Defendants, Appellees,

BRANVILLE G. BARD, JR., in his individual capacity,

Defendant.

APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE DISTRICT OF MASSACHUSETTS

[Hon. Angel Kelley, U.S. District Judge]

Before

Gelpí, Lipez, and Howard, Circuit Judges.

Jack Bartholet, with whom Harold Lichten and Lichten & Liss-Riordan, P.C. were on brief, for appellant.

Kate M. Kleimola, Assistant City Solicitor, City of Cambridge Law Department, for appellees.

Peter J. Duffy and Pollack Solomon Duffy LLP on brief for the Cambridge Police Patrol Officers Association and the Cambridge Police Superior Officers Association, amici curiae.

Page 2Eric R. Atstupenas on brief for the Massachusetts Chiefs of Police Association, Inc., amicus curiae.

August 15, 2025

Page 3LIPEZ , Circuit Judge . Contending that he was

unconstitutionally disciplined in retaliation for exercising his First Amendment rights, Cambridge Police Officer Brian Hussey sued the City of Cambridge, Massachusetts, and the Commissioner of the Cambridge Police Department (collectively, the "Department") under 42 U.S.C. § 1983. The district court granted summary judgment for the Department, finding that its interest in regulating Hussey's speech outweighed the relevant free speech interests. After careful review, we affirm.

The details of the tragic death of George Floyd on May 25, 2020, are well known. Floyd, a Minneapolis, Minnesota resident and father, went to a convenience store to buy cigarettes. Believing Floyd used a counterfeit twenty-dollar bill to make his purchase, a store employee contacted the police. As seen in the widely circulated video of Floyd's arrest that followed, Minneapolis Police Officer Derek Chauvin knelt on Floyd's neck for more than nine minutes, ignoring Floyd's pleas until his body went limp. Floyd's last words were "I can't breathe."¹

The murder of George Floyd led to a public reckoning on issues of racism and policing and sparked historic protests around Page 4the world.² In the City of Cambridge, protesters took to the streets by the thousands to demand action addressing police brutality and impunity. See Hussey v. City of Cambridge, 720 F. Supp. 3d 41, 47 & n.3 (D. Mass. 2024) (citing Marc Levy, Protest Draws Thousands to Hear the Challenges of Reforming Police, Education, Other Institutions, Cambridge Day (June 7, 2020), https://www.cambridgeday.com/2020/06/07/protest-draws-thousandsto-hear-the-challenges-of-reforming-police-education-otherinstitutions/ [https://perma.cc/37UE-DQUB]). In nearby Boston, racial-justice protesters were allegedly met with unreasonable force by the Boston Police Department, leading to a lawsuit asserting constitutional violations. See id. at 47 & n.4 (citing Huffman v. City of Boston, No. 21-cv-10986, 2022 WL 2308937, at *1-3 (D. Mass. June 27, 2022)). These protests and related intensive scrutiny of law enforcement continued for many months, well into 2021. See id. at 47 & n.5 (citing Protesters at Boston Rallies Call for Justice for George Floyd, Action on Police Killing Cases, NBC 10 Boston (Mar. 6, 2021), https://www.nbcboston.com/news/local/rally-in-boston-to-callfor-action-on-police-killing-cases/2321508/Page 5 [https://perma.cc/358H-RQQ7]). On February 24, 2021, the George Floyd Justice in Policing Act of 2021 (the "Act") was introduced in the U.S. House of Representatives, the purpose of which was "[t]o hold law enforcement accountable for misconduct in court, improve transparency through data collection, and reform police training and policies." H.R. 1280, 117th Cong. (2021).

Hussey became an officer of the Cambridge Police Department in 1998, working as a patrol officer for the first decade of his career. In 2009, he joined the Department's Special Investigations Unit ("SIU"), where he participated in hundreds of drug investigations. Much of his work in the SIU involved gaining the trust of confidential informants, who were often current or former drug users, and speaking with current drug users throughout the City of Cambridge. Hussey was part of the SIU for about ten years, returning to his role as a patrol officer in 2020.

On February 25, 2021, at 8:08 a.m., Hussey shared to his personal Facebook page an article titled "House Democrats Reintroduce Police Reform Bill in Honor of George Floyd." Along with the article, he posted the following comment: "This is what it's come to . . . 'honoring' a career criminal, a thief and druggie . . . the future of this country is bleak at best." Hussey made this post from his personal phone while at home. A screenshot taken about one hour after the post was made shows it had two comments, to which Hussey did not respond. Hussey deleted the Page 6post a few hours after he shared it, testifying that he did so because the post did not generate much conversation.

Hussey's Facebook account was "restricted," meaning only people Hussey accepted as "friends" could view his posts, and he did not accept friend requests from people he did not know. At the time of the posting, Hussey had 674 Facebook friends, including roughly ninety current or former members of the Cambridge Police Department. While Hussey did not identify himself as a police officer on his Facebook page, most of his Facebook friends were aware of his profession.

Around March 3, approximately six days after Hussey made the Facebook post, then-Commissioner of the Cambridge Police Department Branville G. Bard, Jr. was contacted by a senior officer of the Cambridge chapter of the NAACP. The NAACP officer alerted Bard to Hussey's post, which had been screenshotted and shared with the NAACP soon after Hussey posted it. Shortly thereafter, Bard and the Cambridge City Manager met via videoconference with the senior NAACP officer who initially contacted Bard, another individual who was an NAACP member and community organizer, and a third individual, the former mayor of Cambridge, to discuss the post. Bard testified that, during this meeting, the three individuals were "alarmed and concerned," believing that Hussey's post "called into question" the Cambridge Police Department's "ability to serve in a bias[]-free manner." Bard requested and Page 7was sent a copy of Hussey's post, which he shared with the Cambridge Police Department's Professional Standards Unit ("PSU") to investigate whether the post violated Department policy.

Hussey was placed on administrative leave while the PSU investigated his post. During this time, he received several messages of support from fellow officers of the Cambridge Police Department.

On March 7, as part of the PSU's inquiry, Hussey provided a written statement in which he expressed the following:

What happened to George Floyd on May 25, 2020 never should have happened . . . he did not deserve to die. Derek Chauvin is a disgrace to the badge and probably never should have worn one in the first place. The same goes for the officers who stood by and did nothing. I am 100% in favor of police reform.

He explained further:

That being said, the one thing I disagree with is naming a police reform bill in "honor" of George Floyd. I understand that this incident, the proverbial straw that broke the camel's back, led to the calls for police reform, hence the naming of the bill in his "honor." While George Floyd did not deserve to have his life taken away that day, he was still a violent criminal. I feel that attaching the name of a violent career criminal, in "honor," to a reform bill aimed at the betterment of policing is a disservice to the spirit of the bill.

The PSU also interviewed Hussey (and only Hussey) as part of its investigation. During this interview, Hussey acknowledged that he never called persons suffering from substance use disordersPage 8 "druggie[s]" to their faces, noting that "it would be unprofessional to do so."

On April 14, approximately six weeks after Hussey was placed on administrative leave, the PSU released a Final Report of Investigation, concluding that Hussey's post violated the Cambridge Police Department's policy against "[d]iscourtesy, rudeness, or insolence to any member of the public" and its rule that officers must "[b]e courteous and act professionally at all times." Two weeks later, and two months after Hussey made his Facebook post commenting on the Act, Bard informed Hussey that he would be suspended without pay for four days because of his rule and policy violations. During his deposition, Bard explained that he considered Hussey's Facebook post "damaging to the reputation of the Cambridge Police Department" and especially harmful "in the context of the national climate." Bard further testified that he believed it was important for the Cambridge Police Department to maintain a positive reputation within the community "[b]ecause that's the only way we can function properly and do our job. We have to be seen as trustworthy and legitimate and bias-free."

Asserting that he was disciplined in retaliation for exercising his First Amendment rights, Hussey filed this suit in federal court under 42 U.S.C. § 1983.³ Following discovery, the Page 9Department moved for summary judgment, arguing in relevant part that its interest in regulating Hussey's speech far outweighed the countervailing free speech interests. Hussey filed a cross-motion for partial summary judgment on the question of the Department's liability for violating his First Amendment rights.

The district court granted summary judgment for the Department and denied Hussey's cross-motion, emphasizing in particular the contemporaneous civil unrest in concluding that Hussey's suspension did not violate his constitutional rights. This appeal followed.

We review the district court's grant of summary judgment de novo. See Satanic Temple, Inc. v. City of Boston, 111 F.4th 156, 167 (1st Cir. 2024). Summary judgment should be granted when the record reveals "no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." Fed. R. Civ. P. 56(a). A dispute is genuine when "the evidence is such that a reasonable jury could resolve the point in the favor of the non-moving party," and a material fact is one that "has the potential of affecting the outcome of the case." Hamdallah v. CPC Carolina PR, LLC, 91 F.4th 1, 16 (1st Cir. 2024) (quoting Page 10Taite v. Bridgewater State Univ., Bd. of Trs., 999 F.3d 86, 93 (1st Cir. 2021)). "In opposing a motion for summary judgment, the plaintiff 'bears "the burden of producing specific facts sufficient to"' defeat summary judgment." González-Cabán v. JR Seafood Inc., 48 F.4th 10, 14 (1st Cir. 2022) (quoting Theidon v. Harvard Univ., 948 F.3d 477, 494 (1st Cir. 2020)). "While we resolve all reasonable inferences in favor of the nonmoving party, we must ignore conclusory allegations, improbable inferences, and unsupported speculation." Viscito v. Nat'l Plan. Corp., 34 F.4th 78, 83 (1st Cir. 2022) (quoting Garcia-Garcia v. Costco Wholesale Corp., 878 F.3d 411, 417 (1st Cir. 2017)).

The First Amendment guarantees the right of the people to speak on matters of public concern, and individuals do not lose that right simply by choosing to work for the government. See Curran v. Cousins, 509 F.3d 36, 44 (1st Cir. 2007) (citing Connick v. Myers, 461 U.S. 138, 142 (1983)). However, as the Supreme Court has explained, "[g]overnment employers . . . need a significant degree of control over their employees' words and actions; without it, there would be little chance for the efficient provision of public services." Garcetti v. Ceballos, 547 U.S. 410, 418 (2006). "Public employees, moreover, often occupy trusted positions in society. When they speak out, they can express views that Page 11contravene governmental policies or impair the proper performance of governmental functions." Id. at 419.

We use a three-part test "to balance the competing interests of the government employer and the employee." MacRae v. Mattos, 106 F.4th 122, 133 (1st Cir. 2024). At step one, we assess whether the employee "spoke as a citizen on a matter of public concern." Gilbert v. City of Chicopee, 915 F.3d 74, 82 (1st Cir. 2019) (quoting Curran, 509 F.3d at 45). "In making this determination, we ask whether the 'speech' underlying [the employee's] claim was made 'pursuant to [the employee's] official duties.'" Id. (quoting Garcetti, 547 U.S. at 421). If so, the employee has no First Amendment claim because, "generally, '[r]estricting speech that owes its existence to a public employee's professional responsibilities does not infringe any liberties.'" Id. (alteration in original) (quoting Garcetti, 547 U.S. at 421-22). If we instead conclude that "the employee did speak as a citizen on a matter of public concern," we move to step two. MacRae, 106 F.4th at 133.

At step two, we must determine "whether, when balanced against each other, the First Amendment interests of the plaintiff and the public outweigh the government's interest in functioning efficiently." Mihos v. Swift, 358 F.3d 91, 102 (1st Cir. 2004). This assessment is commonly referred to as the Pickering balancing test. See Pickering v. Bd. of Educ., 391 U.S. 563, 568 (1968);Page 12 see also, e.g., Mihos, 358 F.3d at 103. If the balance "tip[s] in the employer's favor, the inquiry ends there, and the employee's speech is not constitutionally protected." MacRae, 106 F.4th at 133. If, however, the balancing test favors the employee, "the employee's speech 'is protected speech under the First Amendment' and '[t]he analysis then proceeds to the third step.'" Id. (alteration in original) (quoting Ciarametaro v. City of Gloucester, 87 F.4th 83, 88 (1st Cir. 2023)).

At the third step, we ask "whether the protected speech was a 'substantial or motivating factor in the adverse employment decision.'" Ciaramentaro, 87 F.4th at 88 (quoting Curran, 509 F.3d at 45). Even if the test favors the employee at all three steps, the employer "has the opportunity 'to prove by a preponderance of evidence that "it would have reached the same decision regarding the adverse employment event even in the absence of the protected conduct."'" MacRae, 106 F.4th at 133 (quoting Stuart v. City of Framingham, 989 F.3d 29, 35 (1st Cir. 2021)).

In this case, the first step of the three-step test is undisputed -- that is, the parties agree that Hussey spoke as a citizen on a matter of public concern. Although we note that the third step is also undisputed -- that is, the parties agree that Hussey's speech motivated the Department's decision to discipline him -- our inquiry here ends at step two because, as we shall explain, the Pickering balancing test favors the Department, and Page 13Hussey's speech is not constitutionally protected. Accordingly, we turn now to the Pickering balancing test.

As indicated, but stated more specifically, at the second step of our three-step inquiry, we must "attempt[] to 'balance the value of an employee's speech -- both the employee's own interests and the public's interest in the information the employee seeks to impart -- against the employer's legitimate government interest in preventing unnecessary disruptions and inefficiencies in carrying out its public service mission.'" Bruce v. Worcester Reg'l Transit Auth., 34 F.4th 129, 138 (1st Cir. 2022) (alteration in original) (quoting Decotiis v. Whittemore, 635 F.3d 22, 35 (1st Cir. 2011)). In conducting this inquiry, we recognize that "the stronger the First Amendment interests in the speech, the stronger the justification the employer must have." MacRae, 106 F.4th at 136 (quoting Curran, 509 F.3d at 48). In weighing the countervailing interests, we consider a host of factors, including "(1) 'the time, place, and manner of the employee's speech,' and (2) 'the employer's motivation in making the adverse employment decision.'" Bruce, 34 F.4th at 138 (quoting Decotiis, 635 F.3d at 35). If "we determine that the employee 'face[d] only those speech restrictions that are necessary for [the] employer[] to operate efficiently and effectively,' then the defendants'Page 14 restrictions on speech were adequately justified." Id. (first and third alterations in original) (citation omitted).

The Pickering balancing inquiry is a question of law for the court, but "it is also a 'fact-intensive' inquiry." MacRae, 106 F.4th at 136 (quoting Fabiano v. Hopkins, 352 F.3d 447, 457 (1st Cir. 2003)). In other words, where the material facts are undisputed, we may determine the outcome of the Pickering inquiry as a matter of law at the summary judgment stage. If, however, material factual disputes remain, summary judgment is improper. Cf. Reuland v. Hynes, 460 F.3d 409, 419 (2d Cir. 2006) (holding that genuine "factual disputes underlying Pickering balance must be submitted to the jury").

Hussey argues that the district court erred in holding that the Department's interest in preventing unnecessary disruptions outweighed his First Amendment interests. Specifically, he asserts that (1) the value of his speech was high because his comment touched on important public issues concerning pending legislation ; (2) the Department's mere prediction of disruption, absent evidence of "actual disruption," should not outweigh the relevant free speech interests ; (3) even if actual disruption is not required, the Department's prediction of disruption was unreasonable; and (4) the Department's interest should be accorded little value because the Department was motivated not by any reasonable concern for disruption but by Page 15impermissible viewpoint discrimination or, at the least, there is a genuine dispute of fact as to its motivation. We address each argument in turn.

We begin with Hussey's assertion that his comment on the Act is entitled to significant weight in the Pickering balance. Hussey's post, in which he objected to the naming of pending federal legislation, touched on important political issues. That conclusion is bolstered by the fact that the legislation was intended to address such important subjects as police reform and racial inequity and that, as a law enforcement officer, Hussey could potentially offer a valuable perspective on those subjects. See Garcetti, 547 U.S. at 419 (acknowledging "the importance of promoting the public's interest in receiving the well-informed views of government employees engaging in civic discussion"). Moreover, as the district court rightly determined, "given the galvanizing role George Floyd's murder played in the widespread protests in 2020 and thereafter, there is value in the public's continued discussions of his life and legacy." Hussey, 720 F. Supp. 3d at 54.

Normally, considering the above, Hussey's and the public's interest in Hussey's speech would weigh heavily in his favor because, as the Supreme Court "has frequently reaffirmed[,] . . . speech on public issues occupies the 'highest rung of the Page 16hierarchy of First Amendment values,' and is entitled to special protection." Connick, 461 U.S. at 145 (quoting NAACP v. Claiborne Hardware Co., 458 U.S. 886, 913 (1982)). However, we have recognized that speech commenting on public "issues in a mocking, derogatory, and disparaging manner" is accorded less weight in the balancing test. MacRae, 106 F.4th at 137; see also Curran, 509 F.3d at 49 ("Speech done in a vulgar, insulting, and defiant manner is entitled to less weight in the Pickering balance."). The district court concluded that Hussey's comment was "inflammatory and insulting," and, for that reason, it found that the post's value was "lessened." Hussey, 720 F. Supp. 3d at 55.

Hussey argues on appeal that his comment "does not even come close" to constituting the "extreme" insulting or defiant speech to which we assign diminished value in the Pickering balance. Specifically, he contends that the term "druggie" is not offensive and that none of the individual words he used was inherently derogatory, profane, or racially charged. Our dissenting colleague agrees with Hussey's contention but goes a step further, arguing that because "Hussey's chosen terminology" is not "widely accepted to meet the same level of vulgarity as an insidious moniker like the n-word," Hussey's post cannot be considered "mocking" or "disparaging" under our precedents.

Hussey and the dissent are incorrect, however, in arguing that the specific words Hussey used needed to be Page 17intrinsically vulgar or racist for his speech to be considered insulting. Rather, we consider the words in context and assess whether the manner in which the message as a whole was conveyed "mock[s]" and "disparag[es]." MacRae, 106 F.4th at 137. For example, we recently found that a meme posted online concerning Dr. Rachel Levine, the then-United States Assistant Secretary for Health, and a transgender woman, "was clearly insulting and disparaging when it included the following text: '"I'm an expert on mental health and food disorders." . . . says the obese man who thinks he's a woman.'" Id. at 127, 137. Even though no individual word in the speech at issue was innately offensive, we concluded that the message was derogatory toward transgender individuals and therefore "not accorded the highest value by the First Amendment." Id. at 137. In light of this precedent, 4 we do not understand the basis for our dissenting colleague's "insidious moniker" rule.

Page 18When we evaluate whether the message of Hussey's post was derogatory or disparaging by considering the words he used in context, as our caselaw instructs, we conclude, as did the district court, that his speech is not entitled to "special protection." Connick, 461 U.S. at 145. Hussey's post referred to George Floyd as "a career criminal," "a thief," and a "druggie" and suggested that "the future" of a country that would honor Floyd by naming legislation after him "is bleak at best." This post -- because of the words Hussey chose -- disparaged both George Floyd and those who rallied in response to the horrific circumstances of his death.⁵

To be sure, in his statement to the PSU, Hussey acknowledged that George Floyd's murder was "the proverbial straw that broke the camel's back" and "led to the calls for police reform, hence the naming of the bill in his 'honor.'" In its Final Report of Investigation, the PSU stated that "[i]t is unfortunate Page 19Officer Hussey did not mention his . . . support of police reform" in his post, which could have "provide[d] additional context to his actions and beliefs." Bard similarly testified that he wished Hussey's statement to the PSU "was the content of his Facebook post because then we wouldn't be sitting here today." But, of course, Hussey's post did not express his views on the Act's title in the same manner as his statement to the PSU.

Despite the alarm sounded by our dissenting colleague, our conclusion does not mean that Hussey's comment is not entitled to protection at all or that his interest in making it does not weigh in his favor in the Pickering balancing test. Nor are we creating "a free pass for governments to police . . . viewpoints," a warning by the dissent that discounts the ability of courts to draw meaningful lines when applying a doctrine. Rather, the disparaging nature of Hussey's speech simply means that it is not on the "highest rung" of the First Amendment "hierarchy" as we consider the strength of the Department's countervailing interest. Connick, 461 U.S. at 145 (quoting Claiborne, 458 U.S. at 913).

The Department has an "interest in the effective and efficient fulfillment of its responsibilities to the public." Id. at 150. And a government employer has a "strong interest in 'preventing unnecessary disruptions and inefficiencies in carrying out its public service mission.'" MacRae, 106 F.4th at 137Page 20 (quoting Díaz-Bigio v. Santini, 652 F.3d 45, 53 (1st Cir. 2011)). Moreover, the government employer's interest in regulating speech to prevent disruption is "heighten[ed]" when the employer is a law enforcement agency due to "the special degree of trust and discipline required in" that context. Curran, 509 F.3d at 50 (quoting O'Donnell v. Barry, 148 F.3d 1126, 1135 (D.C. Cir. 1998)); see also Lalowski v. City of Des Plaines, 789 F.3d 784, 792 (7th Cir. 2015) ("[D]eference to the employer's judgment regarding the disruptive nature of an employee's speech is especially important in the context of law enforcement." (quoting Kokkinis v. Ivkovich, 185 F.3d 840, 846 (7th Cir. 1999))). Nevertheless, Hussey offers several reasons to diminish the Department's interest in regulating his speech.

Hussey argues that, notwithstanding the Department's legitimate interest in preventing disruption to its ability to carry out its mission, it should not have been permitted to take adverse action against him absent a showing of "actual disruption" resulting from his speech. Asserting that we "have been unclear on precisely when actual disruption is needed," Hussey invites us to "make clear that where the primary purpose of speech is to discuss a matter of public concern, the government must show an actual disruption." Allowing government employers to regulate political speech based only on predicted disruption, Hussey warns, Page 21would "invite governmental abuses" and violate "the intent of the First Amendment." Because there is no evidence in the record that his post caused actual disruption in the community or within the Department, Hussey contends, the district court erred in concluding that the Pickering balance weighed in the Department's favor.

We are puzzled by Hussey's assertion that our caselaw is unclear as to the necessary showing of disruption in this context. We have long recognized that "[a]n employer need not show an actual adverse effect in order to terminate an employee under the" Pickering balancing test. Curran, 509 F.3d at 49. Rather, as we have repeatedly stated, "an employer may consider a speech's potential to disrupt." Davignon v. Hodgson, 524 F.3d 91, 105 (1st Cir. 2008). And this is so even when a government employee's speech touches upon important political issues. MacRae, 106 F.4th at 137-38. Our established precedent is consistent with the Supreme Court's direction to "give[] substantial weight to government employers' reasonable predictions of disruption, even when the speech involved is on a matter of public concern," Waters v. Churchill, 511 U.S. 661, 673 (1994), and is in line with how other circuits have applied the Pickering balancing test, see, e.g., Moser v. L.V. Metro. Police Dep't, 984 F.3d 900, 909-10 (9th Cir. 2021); Gillis v. Miller, 845 F.3d 677, 687 (6th Cir. 2017); Munroe v. Cent. Bucks Sch. Dist., 805 F.3d 454, 472 (3d Cir.

Page 222015); Anzaldua v. Ne. Ambulance & Fire Prot. Dist., 793 F.3d 822, 833-34 (8th Cir. 2015); Anemone v. Metro. Transp. Auth., 629 F.3d 97, 115 (2d Cir. 2011).⁶

We see no reason to depart from our past decisions by requiring government employers to wait for actual disruptions to their operations before disciplining employees for their speech. Indeed, we recently rejected a similar invitation to adopt such a requirement, noting that the argument for doing so does not "make[] much sense given our caselaw." MacRae, 106 F.4th at 138. The rule in this circuit, as we explain further below, requires a government employer's prediction of disruption to "be reasonable based upon the record." Id. That rule strikes an appropriate balance . It safeguards employees from the "governmental abuses" that Hussey fears without forcing the government "employer to allow events to unfold to the extent that the disruption of the office and the destruction of working relationships is manifest before taking action." Connick, 461 U.S. at 152. Accordingly, we reaffirm our existing application of the Pickering balancing test Page 23that assesses the reasonableness of the government employer's prediction of disruption. See MacRae, 106 F.4th at 138.

Hussey next contends that, even absent a requirement that the Department show evidence of actual disruption, the district court erred in favoring the Department in the Pickering balance because the Department's prediction of disruption was unreasonable. To be sure, "mere speculation of disruption" is not enough. Id.; see also Davignon, 524 F.3d at 105 ("The 'mere incantation of the phrase "internal harmony in the workplace" is not enough to carry the day.'" (quoting Gustafson v. Jones, 290 F.3d 895, 911 (7th Cir. 2002))). Instead, we consider whether the government employer can point to "specific facts and circumstances . . . that support [its] prediction." MacRae, 106 F.4th at 140.

In making his claim of error, Hussey primarily argues that the Department could not have reasonably anticipated his speech would negatively impact internal workplace harmony. He notes, for example, that his speech did not "encourage[] insubordination" or "encourage[] strikes or work stoppage," he "did not personally attack any of his coworkers," and the only reaction to his post amongst his fellow officers was support. While we agree with Hussey that there is scant evidence in the record to support a prediction of disruption to the Cambridge Police Department's internal operations, the absence of such Page 24evidence here is beside the point. The Department does not claim that it disciplined Hussey because his speech could potentially interfere with internal management. Rather, the Department maintains that Hussey's suspension was motivated by its concern that Hussey's post would undermine the Department's relationship of trust with the public.

As multiple circuits have recognized, "[p]olice departments . . . have a strong interest in maintaining a relationship of trust and confidence with the communities they serve." Hernandez v. City of Phoenix, 43 F.4th 966, 981 (9th Cir. 2022); see also Lalowski, 789 F.3d at 792 (noting that police officers have a "responsibilit[y] . . . to foster a relationship of trust and respect with the public"). "The effectiveness of a city's police department depends importantly on the respect and trust of the community and on the perception in the community that it enforces the law fairly, even-handedly, and without bias." Locurto v. Guiliani, 447 F.3d 159, 178 (2d Cir. 2006) (quoting Pappas v. Guiliani, 290 F.3d 143, 146 (2d Cir. 2002)); see also Bennett v. Metro. Gov't of Nash. & Davidson Cnty., 977 F.3d 530, 544 (6th Cir. 2020) ("[D]iverse constituents . . . need to believe that those meant to help them in their most dire moments are fairminded, unbiased, and worthy of their trust."). Because the Department claims that it disciplined Hussey due to its belief that his post could "tear down . . . th[e] trust [the Department]Page 25 spent a long time building with [the] community," it is the reasonableness of that prediction that we must assess.

We agree with the district court that the timing of Hussey's speech is central to that assessment and, ultimately, to our conclusion that the Department's prediction was reasonable. Hussey posted about the Act during a period of intense scrutiny of and protest against law enforcement, including within the City of Cambridge and in nearby Boston. As the district court recounted:

The vast majority of protests were peaceful, yet departments across the country met many of those same protests with tear gas, arrests, flash grenades, and more -- sometimes with fatal consequences for demonstrators. Some of the protests turned violent and the ensuing riots set ablaze local businesses, restaurants, news buildings, and even a police department building. A number of police officers sustained injuries during such incidents. In the weeks that followed, as the names of more Black persons, like Breonna Taylor, became synonymous with police brutality and impunity, it poured fuel on the collective fury against police departments and ignited new rounds of protests. These demonstrations were often accompanied by calls to reform, defund, and even abolish police departments.

Hussey, 720 F. Supp. 3d at 57 (footnotes omitted). Given the contentiousness of this period, the Department's concern about its reputation within the community is easy to understand. This contentiousness also explains why the Department would be especially sensitive to a comment disparaging George Floyd, particularly one made by a long-term employee who frequently Page 26interacted with vulnerable community members. See Grutzmacher v. Howard County, 851 F.3d 332, 346 (4th Cir. 2017) ("[T]he more the employee's job requires . . . public contact, the greater the [government's] interest in firing [the employee] for expression that offends [the] employer." (first alteration and omission in original) (quoting McEvoy v. Spencer, 124 F.3d 92, 103 (2d Cir. 1997))). Affording due deference to the Department, see Lalowski, 789 F.3d at 792, we cannot say it was unreasonable to worry, as Bard testified, that "in the context of the national climate," Hussey's post "tore at the fabric of the trust that [the Department] spent a long time building."⁷

To counter this broad context argument, Hussey claims that "nothing about the time, place, and manner of [his] speech had the potential to be disruptive" because he posted on his personal Facebook account, from his personal phone, while he was at home. That claim is unpersuasive for several reasons. For one, speech does not need to occur within the workplace during Page 27working hours for a government employer to predict reasonably that it could disrupt the employer's provision of public services. See, e.g., Curran, 509 F.3d at 41, 47-50 (finding substantial risk of disruption where employee posted on union message board while suspended from employment); Locurto, 447 F.3d at 164-65, 182 (determining employer's prediction of disruption was reasonable where employee's speech occurred outside of work during Labor Day parade).

Additionally, once speech is posted on the internet, the speaker has virtually no control over its distribution, creating the possibility that it will reach a far broader audience far more quickly than speech disseminated in other ways. The facts here clearly indicate as much. Even though Hussey posted on his personal Facebook page, and despite his intention to restrict his Facebook page to people whom he knew, the post was screenshotted within an hour and distributed outside his network of Facebook friends. Far from being "private," Hussey's Facebook post had the reasonable potential to -- and, in fact, quickly did -- reach the broader public with whom the Cambridge Police Department sought to maintain a relationship of trust.⁸

Page 28Hussey also asserts that the Department's prediction of impaired community relations was unreasonable because "there was no evident disruption" during the pendency of the PSU's investigation. Our dissenting colleague accepts this assertion, arguing that, regardless of whether the Department's prediction of disruption was reasonable "at the time the Department learned of Hussey's comment," the absence of "actual disruption" during the PSU's approximately two-month investigation necessarily rendered the Department's prediction unreasonable by the time it suspended Hussey.

In making this argument, the dissent portrays the Department's "first disciplinary act" of placing Hussey on administrative leave as disconnected from the "second disciplinary action" of suspending Hussey. That disconnect is a contrivance. Hussey's placement on leave and suspension were not unrelated. Rather, they were two steps in a disciplinary process triggered by a single event -- Hussey's post -- and the Department's concern Page 29about a breach of public trust with tangible consequences in the post's aftermath.

Thus, in determining whether the Department's prediction of disruption was reasonable when it suspended Hussey, we cannot disregard the circumstances that existed when the Department first acted in response to the community representatives' concerns. That is, the reasonableness of the Department's prediction of disruption cannot turn solely on whether its initial prediction materialized during the time it took the PSU to conduct its investigation. To hold otherwise would be to require the Department to show actual disruption to justify suspending Hussey. But, as we have explained --and as the dissent acknowledges --there is a broad consensus that the absence of actual disruption does not determine the permissibility of an employer's regulation of speech. See supra; see also, e.g., Gillis, 845 F.3d at 681-82, 687 (concluding that employer's prediction of disruption was reasonable where it terminated employee for his speech two months after placing him on leave, notwithstanding that no actual disruption occurred during those months).

The nonoccurrence of actual disruption during the investigation, prior to the imposition of final discipline, is simply one factor to consider as part of the reasonableness inquiry. Here, notably, the probative weight of the absence of Page 30actual disruption is diminished because of another factor relevant to the reasonableness analysis -- the plausible link between the period of calm while the PSU's investigation took place and the investigation itself. As the Department argues, its "decisive action to avoid a loss of the public's trust" reasonably might have shown the community representatives and the public at large that their concerns were being taken seriously, thus tempering their response while the PSU investigated. See Snipes v. Volusia County, 704 F. App'x 848, 852-53 (11th Cir. 2017) (holding in similar context that it was "reasonably possible" that defendant's "swift action" in launching an investigation that led to plaintiff's dismissal prevented actual disruption from occurring). Despite agreeing that "factual developments" taking place "prior to a public employee's discipline" are "relevant to the reasonableness of the employer's prediction of disruption," the dissent ignores the impact of the intervening investigation, instead elevating one factual development above all others. This insistence that the absence of disruption during the investigative period is dispositive of the reasonableness inquiry is nothing more than an actual disruption requirement in a different guise.

Moreover, accepting the argument that the dissent advances would incentivize hasty, underinformed impositions of final discipline rather than the measured decision-making that the stakes for the employer and employee require. Indeed, our Page 31dissenting colleague seems to accept that there would be no First Amendment problem if the Department had suspended Hussey as soon as it learned of the community's concerns but insists that it became impermissible to impose that discipline by the conclusion of the PSU's full investigation solely because the community did not erupt while that investigation played out. That cannot be. An investigation such as the PSU's that may lead to the imposition of discipline against a public servant must be careful and thorough, not rushed. We will not undermine the Department's ability to investigate carefully by placing dispositive weight on a lack of disruption during that investigation in our consideration of the factors pertinent to the reasonable prediction inquiry.⁹

In sum, considering how Hussey disseminated his speech, the broader context in which he spoke, and all other relevant factors, and deferring to the Department, as we must, we agree with the district court that the Department's prediction of disruption was reasonable.

Lastly, Hussey avers that while the Department claims to have suspended him out of concern for potential disruption to its relationship with the public -- a "weighty interest[] on [its] side of the Pickering scale," Mihos, 358 F.3d at 103 --its true purpose was disagreement with the content of his speech, negating the strength of the Department's interest in the Pickering balancing test. See id. (explaining that if employee was terminated because government employer "disagreed with" employee's speech, employer "would have no legitimate governmental interests on [its] side of the scale"). At a minimum, Hussey says he has pointed to sufficient facts to raise a genuine dispute as to the Department's purpose, rendering a grant of summary judgment inappropriate.

First, Hussey finds it relevant that the initial complaint to Bard was made by a "well-connected" community member, i.e., an officer of the NAACP. But Hussey's assertion that the involvement of the NAACP somehow suggests that viewpoint discrimination was at play is precisely the sort of "unsupported speculation" that "we must ignore" when deciding a motion for summary judgment. See Viscito, 34 F.4th at 83. To the contrary, the fact that an NAACP officer and other community activists, including the former Cambridge mayor, complained to the Department about Hussey's post simply reinforces the reasonableness of the Page 33Department's conclusion that the post threatened its relationship with the public. The community leaders' concern, without more, in no way shows that the Department responded to Hussey's comments differently from the way it would have treated any comments calling into question the Department's ability to interact with the public in a "bias-free manner."

Second, taking a more specific approach to his viewpoint discrimination claim, Hussey avers that the Department did not regulate other speech that could have implicated harmony and community trust, demonstrating that he was "singled out" based on the Department's disagreement with the content of his speech. We agree with the district court, however, that the other instances Hussey cites do not raise a genuine dispute as to the Department's purpose here:

• · Hussey claims that the Department did not discipline Commissioner Elow for referring to drug users as "crackheads" in July 2022. But the few details in the record about this comment indicate significantly different circumstances. Most importantly, Commissioner Elow apparently made the comment in person only to Hussey, with no one else present.
• · Hussey points out that Bard could not recall whether he initiated an investigation into a post on the Twitter page of the Cambridge Police Patrol Officers Association suggesting that a violent "purge" would result from a proposed police
• Page 34reform bill. However, Hussey has identified no evidence that the Department had the authority to investigate activity by the police union, a distinct entity.
• · Hussey asserts that he was treated less favorably than an officer who used an expletive to refer to a congressman on the Cambridge Police Department's official Twitter page. But the record shows that, in fact, the Department suspended that officer for five days -- longer than Hussey's suspension.

Third, Hussey contends that he has raised a genuine dispute as to the Department's purpose for imposing discipline because the Department "made no inquiry into who was aware of the speech in question," and there is no evidence that the post "reach[ed] members of the community with substance abuse issues."¹⁰ Even drawing all inferences in Hussey's favor, we do not see how those assertions reasonably call into question the Department's purpose in suspending him. The record indisputably shows that multiple community members were aware of Hussey's speech and were concerned that it evinced bias within the Cambridge Police Department. Hussey also offers no support for requiring the Department to show a threat to its relationship with the specific subset of the public implicated by his use of the pejorativePage 35 "druggie" -- particularly when the overall thrust of his comments went well beyond the issue of substance abuse. Accordingly, Hussey has failed to raise a genuine dispute of fact as to the Department's purpose for its disciplinary action.

In sum, there are no genuine disputes of material fact that prevent us from reaching the following conclusions. First, while Hussey's and the public's First Amendment interests in Hussey's speech are significant, the speech's value was modestly diminished due to its "mocking, derogatory, and disparaging" nature. MacRae, 106 F.4th at 137. Second, the Department's prediction that Hussey's post could undermine its relationship of trust with the community was reasonable. Third, there is no evidence suggesting that the Department's decision to discipline him was driven by anything other than that reasonable prediction. Given the importance of that trusting relationship to the Department's public service mission, we hold that the Department's interest outweighs Hussey's in the Pickering balancing test. Therefore, as explained above, our inquiry ends here. We affirm.

So ordered.

Page 36HOWARD , Circuit Judge, dissenting . There can be no disputing that a police department's management has a powerful interest in maintaining the public's confidence that the department serves without bias the entire community it is sworn to protect. But a government agency is not free to discipline an employee for simply expressing a viewpoint with which the employer disagrees to friends outside of work. Because Hussey's online comment merits no less than the maximum protection afforded by the First Amendment, and the record belies the reasonableness of his suspension, the Pickering balance tilts decidedly in his favor. I would accordingly reverse the district court's grant of summary judgment.

As my colleagues acknowledge, a public employee's speech generally receives the highest level of protection under the First Amendment when that speech addresses a matter of public concern such as, in this case, the text of pending legislation. See Connick v. Myers, 461 U.S. 138, 145 (1983); see also Am. Postal Workers Union, AFL-CIO v. USPS, 830 F.2d 294, 305-07 (D.C. Cir. 1987) (describing terminated public employee's published column criticizing "right to work" legislation as "l[ying] close to, if not at, the top of that spectrum" "of [F]irst [A]mendment values"). This is especially true where, as here, the speaker's employment relates to the matter of public concern at hand, since the public Page 37also has an interest in hearing a perspective informed by professional expertise. Pickering v. Bd. of Educ., 391 U.S. 563, 571-72 (1968) (describing as "essential" a teacher's ability to "speak out freely" on "how funds allotted to the operations of the schools should be spent" given that teachers are "the members of a community most likely to have informed and definite opinions" on such matters); Waters v. Churchill, 511 U.S. 661, 674 (1994) ("Government employees are often in the best position to know what ails the agencies for which they work; public debate may gain much from their informed opinions."). I thus follow a different path than my colleagues because I disagree with their extension here of a tenuous rule, seldom applied in our circuit or elsewhere, to devalue Hussey's and the public's interest in his speech on the ground that the view he expressed was substantively "insulting" and "disparaging."

Occasions on which we have applied a "vulgarity penalty" to diminish a public employee's interest in the Pickering balance are tellingly scarce among our cases. We first borrowed this concept from the Eleventh Circuit in dicta, noting in Jordan v. Carter that "when public-employee expression is done in a '"vulgar, insulting, and defiant" manner' . . . it is entitled to less weight in the Pickering balance." 428 F.3d 67, 73-74 (1st Cir. 2005) (quoting Stanley v. City of Dalton, 219 F.3d 1280, 1290 (11th Page 38Cir. 2000)).¹¹ Still, we acknowledged in that case that we lacked sufficient "particulars relevant to both sides of the balance" to "definitively resolve the constitutional question." Id. at 74.

Since then, we have applied this penalty in only two cases: Curran v. Cousins, 509 F.3d 36 (1st Cir. 2007), and MacRae v. Mattos, 106 F.4th 122 (1st Cir. 2024), cert. denied, 606 U.S. __, 2025 WL 1787664 (June 30, 2025). In the first, Curran, Page 39we considered a series of statements by a disgruntled corrections officer to his coworkers. 509 F.3d at 39-43. His "threatening and menacing" statements included: implying to a supervisor that enforcement of a department "sick-call policy" was akin to Nazi war crimes during World War II; explicitly threatening that "captains and deputies are gonna get shot" when a supervisor inquired about his use of sick time; and writing two extensive posts on an online discussion board for union members in which he analogized supervisors to Adolf Hitler and Nazi officials "that pushed the Jews in" and advocated that other officers "[a]ct brutally" and "plot against" the sheriff like German officials did to Hitler because "[d]eath [comes] before dishonor." Id. at 40-42. In concluding that this "intemperate and extreme language" was "entitled to less weight" because it was "vulgar, insulting, and defiant," we noted that "[i]t [was] difficult to find any First Amendment value to the citizenry" therein. Id. at 49 (emphasis added). Seventeen years later, we revived this penalty for the second time in MacRae, in which we concluded that a series of online posts merited diminished weight in the Pickering balance in part because the plaintiff conceded that the posts were "derogatory towards transgender people." 106 F.4th at 137.

By comparison, it stretches reason to affix this "vulgarity penalty" to Hussey's Facebook post given the factual gulf between those two cases and this one. As a preliminary Page 40matter, one need only skim Curran and MacRae to observe the material differences between those plaintiffs' speech and Hussey's one-sentence post. For one, both Curran and MacRae involved torrents of vitriol dispensed over an extended period: multiple oral statements and online writings over a fourteen-month period in Curran, see 509 F.3d at 40-43, and seven TikTok posts over the course of a year in MacRae, see 106 F.4th at 127-28. Hussey, by contrast, only wrote one comment that he deleted "after a couple of hours." Similarly, Hussey's post is unlike the plaintiffs' statements in Curran and MacRae in that the targeted audience of his speech was not, by all accounts, the group purportedly insulted by it. In Curran, the plaintiff expressly threatened to have his supervisors "shot" to their faces and likened them to Nazis in conversation before advocating their violent demise to his coworkers by metaphor. 509 F.3d at 40. Likewise, in MacRae, the plaintiff's transphobic TikTok posts were part-and-parcel of her inherently public election campaign for her local school board, which she staked in part on her desire to reform the school district's treatment of transgender students. See 106 F.4th at 128. On the other hand, nothing in the record suggests that Hussey directed his speech at the "career criminal[s]," "thie[ves]," and "druggie[s]" he referenced; if anything, he sought (albeit unsuccessfully) to limit its reach to his restricted group of Facebook "friends" and perhaps, at least abstractly, for it to Page 41influence legislative behavior. And quite unlike Curran and MacRae, Hussey did not expressly or impliedly advocate any meaningful harm to the audience of his post nor the group purportedly insulted by it. Instead, he merely sought to change the name (but not, notably, the substance) of pending legislation.

We can set aside the poor fit of the "vulgarity penalty" to these facts, however, because the majority's preliminary error in applying it here is its antecedent assumption that the terms "career criminal," "thief," and "druggie" are so offensive that their use -- in this context or otherwise -- could ever merit the penalty's application. See Slip. Op. at 18. In our prior cases, speech deemed so "vulgar, insulting, and defiant" that it lost protection was of the kind subject to widespread, near-universal social consensus of its contemptibility. See Curran, 509 F.3d at 41 (threatening violence amid comparisons to Nazism); MacRae, 106 F.4th at 128 (targeting schoolchildren). Indeed, other circuits that apply similar standards only unweight the speaker's interest when the language at issue is of the kind incontrovertibly reviled in modern society. See, e.g., Bennett v. Metro. Gov't of Nashville & Davidson Cnty., 977 F.3d 530, 543 (6th Cir. 2020) (noting how "centuries of history . . . make the use of the [n-word] more than just 'a single word'"); Hernandez v. City of Phoenix, 43 F.4th 966, 978 (9th Cir. 2022) (suggesting that "[s]peech that expresses hostility toward racial or religious minorities may be of Page 42particularly low First Amendment value [in] the Pickering balancing test"). But see Hardy v. Jefferson Comm. Coll., 260 F.3d 671, 678-82 (6th Cir. 2001) (holding that even professor's use of "racially vulgar words" such as the "N-word" during "in-class discussion of 'socially controversial words'" receives "paramount" "constitutional protection" (third quoting Bonnell v. Lorenzo, 241 F.3d 800, 823 (6th Cir. 2001)).

By contrast, the record in this case supplies no credible basis on which to conclude, without more, that Hussey's chosen terminology is widely accepted to meet the same level of vulgarity as an insidious moniker like the n-word. In fact, the only authority advanced in this case to support that conclusion was an assertion by the Commissioner -- the government defendant -- that he considered "druggie" to be "a derogatory term." At no point in the course of Hussey's discipline or this litigation, however, has the Department ever identified any authority besides itself that characterizes Hussey's language as materially offensive such that it might suffer diminished protection under the First Amendment.

The consequence of relying on such thin support to condemn a public employee's manner of expression is self-explanatory in this case: my colleagues have simply reached an unsupported result. Less obvious, but more troubling, is the precedent that we set by uncritically accepting a defendant employer's self-serving assurances as to what speech isPage 43 "offensive" or otherwise deserving of our "vulgarity penalty." By the majority's reasoning, no public employee will ever be able to demonstrate maximal interest in his or her speech, so long as the defendant employer asserts -- apparently on no other authority than its own declaration as much -- that it or another group considers the speech offensive.

Indeed, now, what was a narrowly circumscribed rule has transformed into a free pass for governments to police the viewpoints expressed by their employees simply by deeming "derogatory" or "disparaging" the vocabulary essential to expressing certain disfavored opinions and thus subjecting those opinions to only qualified protection under the First Amendment. Slip Op. at 18. My colleagues openly embrace this result, emphasizing that the principal reason that Hussey's speech deserves less protection is that "in context" his "message as a whole" cast aspersions on "both George Floyd and those who rallied in response to the horrific circumstances of his death." Slip Op. at 17-18. Thus the majority opinion, as I read it, goes so far as to suggest that any expression of the view that pending legislation should not be named after Floyd would not merit full protection so long as it criticized the legislation's namesake. That is to say, Hussey's views apparently deserve less protection not because of the terminology with which he expressed them, but because my colleagues disagree. Such textbook viewpoint discrimination is Page 44well outside the Eleventh Circuit-borne vulgarity standard we adopted in Jordan, 428 F.3d at 74, which asks only whether the "nature of his words," rather than the substantive message they convey, implicates the viewpoint-neutral "manner, time, and place" of his speech on which Connick trains our attention. Dartland v. Metro. Dade Cnty., 866 F.2d 1321, 1324 (11th Cir. 1989) (second quoting Connick, 461 U.S. at 152); see also MacRae, 2025 WL 1787664, at *2 (Thomas, J., concurring in denial of certiorari) ("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 on that point -- the manner, time, and place in which Hussey expressed his views -- my colleagues fault him for doing so contemporaneously with ongoing public discourse about Floyd's death, apparently suggesting that Hussey's comment would somehow earn greater protection as the debate settled with time. But such a directive obscures the very purpose of the Supreme Court's framework for analyzing public employees' speech rights, which strives to ensure that public employees "face only those speech restrictions that are necessary," Garcetti v. Ceballos, 547 U.S. 410, 419 (2006), and that "citizens are not deprived of fundamental rights by virtue of working for the government," Connick, 461 U.S. at 147. Like waiting to warn a patient of a drug's side effects Page 45until after the patient has already taken the drug, the majority asks public employees to abstain from commenting on matters of public concern until those matters are no longer of concern to the public.

In our time, when questions of permissible terminology are at the fore of public debate, we should be hesitant to limit the protection afforded a public employee's speech based only on an employer's self-serving appraisal of his terminology, lest we allow the government to censor all but its preferred side of that debate. In my estimation, that is precisely what happened here, and setting aside the Department's disagreement with his views, Hussey's online comment was nowhere near so vulgar as to merit any less than the maximum weight afforded in the Pickering balance.

II.

As for the Department's interest in restricting Hussey's speech, I agree that it need not show that Hussey's post provoked actual disruption to its operations; rather, the Department only need show a "reasonable prediction of disruption." MacRae, 106 F.4th at 138. And I also agree that the Department could ground its predictions of disruption in the effect Hussey's speech was likely to have on the Department's external relationships and activities. See Connick, 461 U.S. at 150 (requiring "full consideration of the government's interest in the effective and efficient fulfillment of its responsibilities to the public");Page 46 Decotiis v. Whittemore, 635 F.3d 22, 35-36 (1st Cir. 2011) (accepting public employer's argument concerning the "risk of disruption and efficiency" based on its interest in "its ability to effectively communicate with the vulnerable population it serves"). Where I part ways with my colleagues on this score is on whether the Department's prediction of disruption here, taking stock of the full record, was reasonable by the time Hussey was suspended without pay in May 2021. In my view, it was not.

Briefly summarized, the disruption predicted by the Department distills to its worry that Hussey's harsh criticism of Floyd amidst widespread public outcry about Floyd's death would undermine the Department's relationships with the Cambridge community writ large and drug users therein. And affording the Department the "heightened" deference owed to it as a law enforcement agency, see Jordan, 428 F.3d at 74 (quoting Moore v. Wynnewood, 57 F.3d 924, 934 (10th Cir. 1995)), this prediction of ensuing mistrust could very well have been reasonable at the time the Department learned of Hussey's comment in February 2021 and first placed him on administrative leave.¹²

Page 47The reasonableness of the Department's prediction dissipated, however, by the time the Department again disciplined Hussey in May, more than two months after he published and deleted his Facebook post. By then, many weeks after a few community members learned of his comment, the Department concedes that it had nonetheless "successfully avoided widespread media publicity and a loss of public trust due to [Hussey's] post." In other words, by the time the Department suspended Hussey without pay, it already knew that its worries of impaired community relationships had not, in the end, materialized. As public interest in an episode such as this one typically declines with time, this admission fatally belies any reasonableness in the Department's supposed prediction that yet-unaffected community relationships were still nonetheless at risk of rupturing. More accurately, by May, the Department's "prediction" had morphed from a prospective, experience-informed judgment call into an unlikely pretext affirmatively disproven by its experiences to date. And while we defer to a law enforcement agency's "reasonable prediction of disruption," MacRae, 106 F.4th at 138 (emphasis added), that deference does not require us to blind ourselves to a record that exclusively contradicts the basis for that prediction.

Accounting for such contradiction in the record does not impose an actual-disruption standard on public employers. Indeed, with respect to the Department's first disciplinary act -- placing Page 48Hussey on administrative leave in February -- any subsequent lack of disruption is irrelevant to our analysis of the Department's prediction of disruption at that time, even if it later learned that its prediction was wrong. But, once the Department had the opportunity to observe over multiple months that these preliminary fears did not come to pass (or had already been sufficiently ameliorated by its preemptive disciplinary step), it was incumbent on the Department to account for that development in considering its second disciplinary act against Hussey -- suspending him without pay -- in May. See Davignon v. Hodgson, 524 F.3d 91, 104-05 (1st Cir. 2008) (reasoning that sheriff's "concerns about potential disruption" did not "have substance" because record evidence contradicted the concerns); cf. MacRae, 106 F.4th at 137-38 & n.16 (noting that "there was little opportunity . . . for actual disruption to have occurred" before the plaintiff was disciplined three days after her speech was publicized). By all accounts, it did not. I thus cannot conclude that any prediction of further disruption by the Department at that time was reasonable as the First Amendment requires when, to reach it, the Department evidently ignored its still-intact public trust and unimpaired community relationships.

My colleagues do not meaningfully contend with the effect of this two-month window on the reasonableness of the Department's prediction, instead suggesting that we are to Page 49consider only the facts extant when the Department initiated its investigation and somehow ignore any that arose as the investigation unfolded. The law supplies no foundation for this self-imposed limitation. We often consider factual developments that arise prior to a public employee's discipline to be relevant to the reasonableness of the employer's prediction of disruption, even if the developments occur contemporaneously with an investigation. See, e.g., Davignon, 524 F.3d at 103 (requiring employer to "engage in further investigation" because he "had strong evidence indicating the plaintiffs' speech was not disruptive" that he did not consider); see also Waters, 511 U.S. at 677-78 (calling for consideration of employer's investigatory "procedures" because "it may be unreasonable for an employer to act based on extremely weak evidence when strong evidence is clearly available"). To hold otherwise contradicts the very purpose of such an investigation and implies that we must uncritically accept as gospel the government's conclusions, even when faced with a wholly contradictory record. Rather than discourage "hasty" decision-making as the majority maintains, 13 see Page 50Slip Op. at 31, such willing ignorance will incentivize savvy public employers to manipulate the timing, scope, and conclusions of such investigations to present an incomplete picture of what they learned about the likelihood of disruption as an investigation unfolded, secure in the knowledge that we have promised to close our eyes to any contrary developments during the pendency of that investigation.

III.

As is commonplace in the era of social media, Hussey expressed a view online that may very well have fared poorly in the court of public opinion and that even he attests, in retrospect, that he wishes he had phrased differently. Even so, "the First Amendment does not permit one side of a debate to use the government to cancel the other side. It allows all perspectives, even the very offensive, to be heard." Noble v. Cincinnati & Hamilton Cnty. Pub. Libr., 112 F.4th 373, 383 (6th Cir. 2024).

Page 51Hussey's expression of one such unpopular perspective on legislation pending before his elected representatives is of the kind quintessentially draped with the utmost protection by the First Amendment. Especially as the Department's justification for punishing him for that speech was demonstrably unreasonable, the Pickering balance tilts decisively in his favor -- a result that the appropriate analysis on even just one side of the scales would produce.

I respectfully dissent.