Harvard Law Review Harvard Law Review Harvard Law Review

First Amendment: Speech

Heffernan v. City of Paterson


Individuals do not lose all of their First Amendment protections while working for the government, but those protections are limited in important yet often unpredictable ways.1×1. See, e.g., Garcetti v. Ceballos, 547 U.S. 410, 417 (2006). In a line of public-employee-retaliation cases, the Supreme Court had balanced the First Amendment interests of public employees against the government’s interests as employer, with results heavily favoring the government’s side.2×2. See Catherine L. Fisk & Erwin Chemerinsky, Political Speech and Association Rights After Knox v. SEIU, Local 1000, 98 Cornell L. Rev. 1023, 1066 (2013). But last Term, in Heffernan v. City of Paterson,3×3. 136 S. Ct. 1412 (2016). the Supreme Court held that when a public employer demotes an employee “to prevent the employee from engaging in” protected speech,4×4. Id. at 1418. the employee can bring a § 19835×5. 42 U.S.C. § 1983 (2012). challenge even if the employer was mistaken in its belief “that the employee had engaged in protected speech.”6×6. Heffernan, 136 S. Ct. at 1418 (emphasis omitted). The case fits with a contested theory of the First Amendment: that it exists to ferret out an illicit government motive. Although the Court refused to “lay down a general standard” in previous retaliation cases,7×7. Pickering v. Bd. of Educ., 391 U.S. 563, 569 (1968). when presented with a case where the only possible reason for liability was an illicit government motive, the Court found that motive alone was sufficient to show a constitutional injury. While its use might remain questionable for First Amendment cases involving legislative action, Heffernan shows why this focus on motive fits well in the executive-action context.

In 2005, Jeffrey Heffernan was a detective for the City of Paterson, New Jersey, working in the office of the Chief of Police.8×8. Heffernan, 136 S. Ct. at 1416. The mayor of Paterson was then running for reelection against Lawrence Spagnola, Heffernan’s good friend.9×9. Id. During the campaign, Heffernan’s bedridden mother asked him to pick up a Spagnola campaign sign for her.10×10. Id. He went to a campaign distribution point, where “[o]ther members of the police force saw him, sign in hand, talking to campaign workers.”11×11. Id. The next day, Heffernan was demoted in response to what his supervisors “thought was his ‘overt involvement’ in Spagnola’s campaign.”12×12. Id.

Heffernan filed a § 1983 lawsuit claiming that the city and its agents had retaliated against him for engaging in protected free speech and association.13×13. See Heffernan v. City of Paterson, 2 F. Supp. 3d 563, 570, 576 (D.N.J. 2014). The District Court of New Jersey entered judgment for the defendants,14×14. The case’s winding path to the final district court decision included a jury trial, multiple motions for summary judgment, and three judges. See id. at 567–69. finding that Heffernan had not actually engaged in conduct protected by the First Amendment15×15. Id. at 571, 574, 580 (finding that because Heffernan did not actually engage in political campaigning or expressive conduct or association, his actions were not protected). and that under binding Third Circuit precedent “[n]o First Amendment claim arises from retaliation based on an employer’s mistaken belief that the employee engaged in protected speech.”16×16. Id. at 580 (citing Ambrose v. Twp. of Robinson, 303 F.3d 488, 496 (3d Cir. 2002); Fogarty v. Boles, 121 F.3d 886, 891 (3d Cir. 1997)). Heffernan appealed, arguing that the district court had erred because (1) it should not have even considered the defendants’ motion for summary judgment on his free-association claim; (2) a jury reasonably could have found that Heffernan actually exercised his First Amendment rights; and (3) he could obtain relief based on a “perceived-support” theory.17×17. Heffernan v. City of Paterson, 777 F.3d 147, 151, 153 (3d Cir. 2015). Accepting the perceived-support theory would mean allowing Heffernan to proceed to trial “where the employer’s retaliation [was] traceable to a genuine but incorrect or unfounded belief that the employee exercised a First Amendment right.” Id. at 153.

The Third Circuit affirmed.18×18. Id at 149. Writing for a unanimous panel, Judge Vanaskie19×19. Judge Vanaskie was joined by Judges Greenberg and Cowen. found that with respect to Heffernan’s actual-speech and actual-association arguments, because he had disavowed any intention to campaign, “no room exist[ed] for a jury to find that Heffernan intended to convey a political message when he picked up the sign.”20×20. Heffernan, 777 F.3d at 153. In response to Heffernan’s argument that a previous Third Circuit opinion in the case precluded consideration of a summary judgment motion on this claim, the court held that this was a misunderstanding of its opinion, which “had no bearing on [the defendants’] right to contest the sufficiency of Heffernan’s evidence on his free-association claim through a motion . . . for summary judgment.” Id. at 152. On his perceived-support claim, the court refused to “eliminate a traditional element of a First Amendment retaliation claim — namely, the requirement that the plaintiff in fact exercised a First Amendment right.”21×21. Id. at 153. Eliminating this requirement was “squarely foreclosed” by binding precedent.22×22. Id. (citing Ambrose v. Twp. of Robinson, 303 F.3d 488, 496 (3d Cir. 2002); Fogarty v. Boles, 121 F.3d 886, 891 (3d Cir. 1997)). Furthermore, other circuits that had encountered the issue of factual mistake in the public employee–free speech context had likewise denied the employees’ claims.23×23. The Fifth, Seventh, and Ninth Circuits had previously decided that retaliation claims in free-speech cases must be “prompted by an employee’s actual, rather than perceived, exercise of constitutional rights.” Id. (citing Wasson v. Sonoma Cnty. Junior Coll., 203 F.3d 659, 662 (9th Cir. 2000); Jones v. Collins, 132 F.3d 1048, 1054 (5th Cir. 1998); Barkoo v. Melby, 901 F.2d 613, 619 (7th Cir. 1990)). With respect to the argument that these precedents didn’t foreclose relief based on perceived free association (as opposed to speech), the court found that a Sixth Circuit case relied on by Heffernan, which had found that “the employer’s mere assumption of an affiliation, whether founded or not, was sufficient for the [employees’] claim to proceed,” could not be reconciled with relevant Third Circuit precedent. Id. at 154; see Dye v. Office of the Racing Comm’n, 702 F.3d 286, 299–300 (6th Cir. 2012).

The Supreme Court reversed and remanded.24×24. Heffernan, 136 S. Ct. at 1419. Writing for the Court, Justice Breyer25×25. Justice Breyer was joined by Chief Justice Roberts and Justices Kennedy, Ginsburg, Sotomayor, and Kagan. held that a demotion based on a mistaken belief “that the employee had engaged in protected speech” provides grounds for a § 1983 claim by depriving the employee of a constitutional right.26×26. Heffernan, 136 S. Ct. at 1418 (emphasis omitted). In making this determination, the Court assumed that Heffernan had not actually engaged in activity protected by the First Amendment.27×27. Id. at 1417. Heffernan had presented the case under this assumption. See Brief for Petitioner at i, Heffernan, 136 S. Ct. 1412 (No. 14-1280). The issue was then to resolve the circuit split about whether a perceived-support theory could support Heffernan’s claim.28×28. Heffernan, 136 S. Ct. at 1416.

Justice Breyer first concluded that neither the text of the statute nor precedent answered the question. Looking to the statutory text, he reasoned that although § 1983 authorizes a suit by someone deprived of a right “secured by the Constitution,”29×29. Id. at 1417 (quoting 42 U.S.C. § 1983 (2012)). it does not speak to whether that right “primarily focuses upon (the employee’s) actual activity or . . . primarily focuses upon (the supervisor’s) motive, insofar as that motive turns on what the supervisor believes the activity to be.”30×30. Id.

Turning to precedent, Justice Breyer noted that some decisions “used language that suggests the ‘right’ at issue concerns the employee’s actual activity.”31×31. Id. However, these cases did not present the issue of factual mistake.32×32. Id. at 1417–18 (discussing Garcetti v. Ceballos, 547 U.S. 410 (2006), Connick v. Myers, 461 U.S. 138 (1983), and Pickering v. Bd. of Educ., 391 U.S. 563 (1968)). Instead, “[i]n each of these cases, the only way to show that the employer’s motive was unconstitutional was to prove that the controversial statement or activity — in each case the undisputed reason for the firing — was in fact protected by the First Amendment.”33×33. Id. at 1418. More on point was Waters v. Churchill,34×34. 511 U.S. 661 (1994). in which the plurality noted that the constitutional inquiry focused on “the employer’s motive” in dismissing an employee, “and in particular the facts as the employer reasonably understood them.”35×35. Heffernan, 136 S. Ct. at 1418 (citing Waters, 511 U.S. at 679–80 (plurality opinion)). Thus, a government employer could discharge an employee for what it reasonably thought was speech “not on a matter of public concern,” even if it was mistaken.36×36. Waters, 511 U.S. at 681 (plurality opinion). The employer in Waters, in other words, had erroneously but reasonably believed the employee had not engaged in protected speech. Id. Likewise here, the majority reasoned, where the employer erroneously believed the employee had engaged in protected speech, the employer’s motive should be equally decisive.37×37. Heffernan, 136 S. Ct. at 1418. “After all, in the law, what is sauce for the goose is normally sauce for the gander.”38×38. Id.

Justice Breyer further noted that the Court’s holding tracked the language of the First Amendment, which “begins by focusing upon the activity of the Government.”39×39. Id. It thus stood to reason that the police department could be sued after it “acted upon a constitutionally harmful policy” — one that “‘abridge[d] the freedom of speech’ of employees” who knew about this policy — “whether Heffernan did or did not in fact engage in political activity.”40×40. Id. at 1418–19 (quoting U.S. Const. amend. I).

Furthermore, Justice Breyer argued that a rule imposing liability on the employer better addressed the real constitutional harm at issue — discouraging employees from engaging in protected activity.41×41. Id. at 1419. After all, “[t]he discharge of one tells the others that they engage in protected activity at their peril.”42×42. Id. The Court cited to political patronage cases to support the proposition that discharge based on protected activity inhibits the rights of all employees. Id. (citing Elrod v. Burns, 427 U.S. 347, 359 (1976)). Moreover, Justice Breyer observed that liability would “not normally impose significant extra costs” for the employer, since “[t]o win,” an employee would “have to point to something more than his own conduct” to demonstrate the requisite illicit motive.43×43. Id. None of this reasoning meant that Heffernan automatically won. Rather, because it was unclear whether the police department demoted Heffernan pursuant to a neutral policy prohibiting officers from “overt involvement in any political campaign” rather than an unconstitutional policy, the Court reversed and remanded for further proceedings.44×44. Id. Such policies may be constitutional. See U.S. Civil Serv. Comm’n v. Nat’l Ass’n of Letter Carriers, 413 U.S. 548, 564 (1973).

Justice Thomas dissented,45×45. Justice Thomas was joined by Justice Alito. arguing that “federal law does not provide a cause of action to plaintiffs whose constitutional rights have not been violated.”46×46. Heffernan, 136 S. Ct. at 1420 (Thomas, J., dissenting). He contended that the threshold inquiry to determine “whether a public employer violated the First Amendment rights of its employees” was whether “the employee has . . . spoken on a matter of public concern.”47×47. Id. Heffernan’s claim should fail at the first step, he reasoned, because Heffernan did not actually exercise his First Amendment rights — and thus they could not have been infringed upon.48×48. Id. at 1421. Justice Thomas also took issue with the majority’s textual analysis, observing that the entire Bill of Rights focuses upon government activity, not just the First Amendment.49×49. Id. at 1422–23 (quoting District of Columbia v. Heller, 554 U.S. 570, 636 (2008)). Finally, Justice Thomas attempted to distinguish Waters, in which the plurality stated that it had “never held that it is a violation of the Constitution for a government employer to discharge an employee based on substantively incorrect information.”50×50. Id. at 1423 (quoting Waters v. Churchill, 511 U.S. 661, 679 (1994) (plurality opinion)). The employee in Waters, Justice Thomas noted, argued that she was actually engaged in protected activity, whereas Heffernan could not say the same.51×51. Id.

Although the line of public-employee-retaliation cases preceding Heffernan seems to be at least partly concerned with government motive, the decision in Heffernan confirms that motive is indeed a central consideration. More so than these cases, Heffernan fits a motive-based model of the First Amendment — a theory of the amendment that identifies it as indirectly testing for an illicit government motive. The relative ease of uncovering motive and the potent chilling effect of even mistaken government actions show why a motive-based model is especially powerful in executive-action cases, in contrast to legislative-action cases.

The First Amendment has several purposes. One is to protect citizens’ speaking opportunities.52×52. See Elena Kagan, Private Speech, Public Purpose: The Role of Governmental Motive in First Amendment Doctrine, 63 U. Chi. L. Rev. 413, 424 (1996) (defining a “speaker-based model”); see also John Fee, Speech Discrimination, 85 B.U. L. Rev. 1103, 1107–08 (2005) (identifying a “speech maximizing value” inherent in the First Amendment, id. at 1107). Another is to ensure that citizens are exposed to a marketplace of ideas as an audience, not just as speakers.53×53. See Kagan, supra note 52, at 424 (terming this approach the “audience-based model”). A third is to limit the government’s ability to regulate citizen speech based on illegitimate reasons.54×54. See id. at 425 (terming this approach the “government-based” or “motive-based model”); see also Larry A. Alexander, Trouble on Track Two: Incidental Regulations of Speech and Free Speech Theory, 44 Hastings L.J. 921, 939 (1993) (“[T]he First Amendment expresses as its primary value that government not preempt individuals’ evaluation of information.”).

Heffernan most closely tracks this third rationale. The First Amendment operates in part so that people will not be discouraged from engaging in protected activities for fear of adverse government action. Under a motive-based model of the First Amendment, famously advanced by then-Professor Elena Kagan, “an action may violate the First Amendment because its basis is illegitimate, regardless of the effects of the action on either the sum of expressive opportunities or the condition of public discourse.”55×55. Kagan, supra note 52, at 426; see also Jed Rubenfeld, The First Amendment’s Purpose, 53 Stan. L. Rev. 767, 769 (2001) (“The actor’s purposes are not relevant to free speech analysis. The state’s purposes, on the other hand, are dispositive.”). The Court’s various tests can thus be understood as indirect tests for illicit motives,56×56. Kagan, supra note 52, at 450–56. For example, on Justice Kagan’s theory, content-based regulations are analyzed under strict scrutiny because these restrictions on speech are more likely to be a result of impermissible motive. See id. at 450. which include privileging favored ideas or acting to protect government actors’ interests against contrary ideas.57×57. Id. at 428–29. As demonstrated by Heffernan, in the public-employee context, the alleged constitutional harm does not depend necessarily on what an employee says, but rather on the motives behind the government’s adverse employment action.

The Court has explicitly inquired into government motives in political patronage cases, which hold that “patronage practices offend the First Amendment because the government is acting with the purpose of ‘deny[ing] a benefit to a person because of his constitutionally protected speech or associations.’”58×58. Brief for the United States as Amicus Curiae Supporting Petitioner at 10, Heffernan, 136 S. Ct. 1412 (No. 14-1280) (second alteration in original) (quoting Branti v. Finkel, 445 U.S. 507, 515 (1980)). The results in these cases importantly do not depend on the employees’ actual activity, as the Court does not “require plaintiffs in political affiliation cases to ‘prove that they, or other employees, have been coerced into changing, either actually or ostensibly, their political allegiance.’” Heffernan, 136 S. Ct. at 1419 (quoting Branti, 445 U.S. at 517). In contrast, cases involving alleged retaliation against public employees have not been as prominent exemplars of a focus on motive,59×59. Kagan, supra note 52, at 432–33 (noting that her Article does not examine the government’s role as “employer,” but stating that “the concept of illicit purpose should apply in th[is] context[] even more strongly than it does”). in part because the Court treats the government more deferentially when acting as an employer than as a sovereign.60×60. See, e.g., Waters v. Churchill, 511 U.S. 661, 671 (1994) (“[W]e have always assumed . . . that the government as employer indeed has far broader powers than does the government as sovereign.”); see also Fee, supra note 52, at 1155 (“[T]he Speech Clause’s anti-discrimination principle contains large exceptions relating to the government’s roles as educator, employer, and speech promoter . . . .”). But just because the government gets more leeway when it acts as an employer has not and should not prevent executive-action cases from being understood as implicating motives, even if the Court’s ultimate focus has not always been clear.

Take, for example, Pickering v. Board of Education,61×61. 391 U.S. 563 (1968). a foundational public-employee-retaliation case. In Pickering, an Illinois county board of education dismissed a public school teacher for writing a letter to the local newspaper criticizing the board.62×62. Id. at 564. The Court held that the exercise of the “right to speak on issues of public importance may not furnish the basis for . . . dismissal from public employment.”63×63. Id. at 574 (emphasis added). Why the Board chose to fire the teacher, in other words, mattered: once the Court determined that the reason for the Board’s action was the teacher’s protected speech, the dismissal was presumptively improper.64×64. Unless the Board could prove that the employee’s statements were “made either with knowledge of their falsity or with reckless disregard for their truth or falsity,” id. at 573, his statements could not furnish the basis for his dismissal, id. at 574. Although motive was not dispositive in Pickering itself, later cases applying Pickering suggest that it matters whether speech was the “motivating factor” behind the government’s action.65×65. See Mt. Healthy City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274, 287 (1977) (announcing a burden-shifting analysis for consideration of cases in which the government actor may have mixed motives for its actions); see also Melzer v. Bd. of Educ., 336 F.3d 185, 193 (2d Cir. 2003) (“For an employee to prevail he or she must also demonstrate that the speech was the motivating factor causing the public employer to take adverse action.”); Elizabeth Dale, Employee Speech & Management Rights: A Counterintuitive Reading of Garcetti v. Ceballos, 29 Berkeley J. Emp. & Lab. L. 175, 188–89 (2008).

Other major cases that, like Pickering, do not explicitly put motive front and center can also implicate it in some way. In Connick v. Myers,66×66. 461 U.S. 138 (1983). for example, the Court considered the termination of a prosecutor who refused to accept a transfer and who then distributed a questionnaire that her supervisor viewed as an “act of insubordination.”67×67. Id. at 141. The Court denied the prosecutor’s § 1983 claim, finding that public issues were implicated in only a “limited” way and that the plaintiff’s supervisor was not required to “tolerate action which he reasonably believed would disrupt the office.”68×68. Id. at 154. The various factors considered by the Court can be seen as proxies for determining the likelihood of an illegitimate motive. For example, the “nature of the employee’s expression” and the “manner, time, and place” in which it occurred meant there was a low likelihood of an illicit motive,69×69. Id. at 150, 152. but there was ample reason for the employer to act based on legitimate reasons instead.

Moreover, in Waters, the plurality noted that while “[g]overnment action based on protected speech may under some circumstances violate the First Amendment even if the government actor honestly believes the speech is unprotected,”70×70. Waters v. Churchill, 511 U.S. 661, 669 (1994) (plurality opinion). an employer who reasonably believed that an employee’s speech was not protected did not violate the First Amendment in firing her for it.71×71. See id. at 679–80. The employer had fired the plaintiff because it (mistakenly) believed that the plaintiff’s speech was not protected; the Court interpreted its actions as an attempt to keep its workplace from being disrupted.72×72. See id. at 680–81. As Justice Breyer would put it, “it was the employer’s motive, and in particular the facts as the employer reasonably understood them, that mattered” to the First Amendment analysis.73×73. Heffernan, 136 S. Ct. at 1418. But see Waters, 511 U.S. at 677 (plurality opinion) (rejecting the idea that “intent” is required to make out a First Amendment claim).

The Court’s more recent decision in Garcetti v. Ceballos,74×74. 547 U.S. 410 (2006). meanwhile, did not even reach the Pickering balancing test, solely focusing instead on the duty-related nature of the employee’s speech. The Court upheld the reassignment and transfer of Richard Ceballos, a deputy district attorney, which Ceballos argued was in retaliation for a memo he had written on the job alleging prosecutorial wrongdoing.75×75. Id. at 413–15, 424. The Court held that when making “statements pursuant to their official duties, [public] employees are not speaking as citizens for First Amendment purposes.”76×76. Id. at 421. This decision carved out a subset of cases in which the employer is presumed to be merely exercising its “managerial discretion,”77×77. Id. at 423; see id. at 422–23. foreclosing the issue of motive altogether.78×78. See generally Sheldon H. Nahmod, Public Employee Speech, Categorical Balancing and § 1983: A Critique of Garcetti v. Ceballos, 42 U. Rich. L. Rev. 561 (2008) (criticizing the categorical nature of Garcetti with respect to both the First Amendment and § 1983 implications).

Heffernan, however, involved no reason to preclude consideration of motive. The city’s agents were acting to protect the self-interest of the mayor who appointed them, regardless of what Heffernan was actually doing. In a case where the government’s action exerted a clear chilling effect on other employees’ political activity,79×79. See Heffernan, 136 S. Ct. at 1419. the Court found that an illicit motive was all that was necessary to show constitutional harm. The employer’s mistake of fact about its employee’s actions did not change this result.

Heffernan thus confirmed what seemed to be lurking in Pickering and its progeny: motive matters. The Pickering test applies a presumption that if a government employer acts to restrict an employee’s speech as a citizen on a matter of public concern, it is more likely to be acting on impermissible motives than it is when it restricts employment-related speech. In previous cases, the Court used the employee’s constitutionally protected conduct (or what the employer thought was protected conduct) as a proxy for determining that the employer’s adverse action was based on illicit motives.80×80. See id. at 1418. While the Court has narrowed the circumstances in which it finds that speech is made “as a citizen . . . upon matters of public concern”81×81. Pickering v. Bd. of Educ., 391 U.S. 563, 568 (1968). — limiting the circumstances under which it is worried about motive — Heffernan shows that when there’s a reason to be concerned about motive, it can be a dispositive factor.

In addition, Heffernan illustrates why motive is an especially apt focus for executive-action cases, even as it remains more contested for First Amendment cases involving legislative motive.82×82. See, e.g., Kagan, supra note 52, at 427 n.43 (“In cases involving executive action, the Court routinely speaks in terms of motive. . . . The Court apparently sees the examination of motive in such cases as different in kind from — and less problematic than — the examination of the motives underlying legislation.”). First, much of the opposition to intent-based tests arises from the difficulty of establishing the intent, motive, or purpose of multimember bodies.83×83. See United States v. O’Brien, 391 U.S. 367, 383–84 (1968) (“Inquiries into congressional motives or purposes are a hazardous matter. . . . What motivates one legislator to make a speech about a statute is not necessarily what motivates scores of others to enact it, and the stakes are sufficiently high for us to eschew guesswork.”); see also Richard H. Fallon, Constitutionally Impermissible Legislative Intent, 130 Harv. L. Rev. (forthcoming December 2016); Antonin Scalia, Common-Law Courts in a Civil-Law System: The Role of United States Federal Courts in Interpreting the Constitution and Laws, in A Matter of Interpretation 16–23 (Amy Gutmann ed., 1997). In contrast, in executive-action cases there is often only a single decisionmaker. While the threat of pretextual intent is always present,84×84. See Kagan, supra note 52, at 414, 443. it is far easier to inquire into the motive of one person than an entire legislative body.85×85. The exceptions to the general proposition that intent is difficult to establish for legislative action are cases with egregious evidence of intent. See, e.g., Grosjean v. Am. Press Co., 297 U.S. 233, 251 (1936) (“The form in which the tax is imposed is in itself suspicious. . . . It is measured . . . with the plain purpose of penalizing the publishers and curtailing the circulation of a selected group of newspapers.”). Heffernan’s supervisors even told him “that he was being demoted . . . because of his political involvement” with the challenger in the mayoral campaign.86×86. Heffernan v. City of Paterson, 2 F. Supp. 3d 563, 567 (D.N.J. 2014). “That which stands for a ‘law’ of ‘Congress’” in Justice Breyer’s analogy — “namely, the police department’s reason for taking action” — is far more knowable from a motives perspective than is anything that a legislative body does.87×87. Heffernan, 136 S. Ct. at 1418.

Second, mistakes of fact are more meaningful in executive-action cases than in the standard legislative-intent cases. If a legislature intends to pass a law restricting speech, but is somehow mistaken and passes a benign law, it is not intuitive that anyone’s First Amendment rights have been abridged.88×88. The law’s operation on the books either impermissibly restricts speech or it doesn’t. Cf. O’Brien, 391 U.S. at 385 (“The statute attacked in the instant case has no such inevitable unconstitutional effect, since the destruction of Selective Service certificates is in no respect inevitably or necessarily expressive.”). In contrast, an employer’s adverse actions based on speech can send a meaningful signal even if the employer is mistaken about what it is acting adversely to. Just as Justice Breyer explained, the signal that one’s employer is intent on punishing speech “can cause the same kind, and degree, of constitutional harm whether that belief does or does not rest upon a factual mistake” — it chills speech regardless of its accuracy.89×89. Heffernan, 136 S. Ct. at 1419; see also Connick v. Myers, 461 U.S. 138, 145 (1983).

Justice Breyer’s discussion of these practical constitutional harms shows that the government’s motive is especially important in executive-action cases because motive itself has effects.90×90. Heffernan, 136 S. Ct. at 1419 (discussing the “constitutional implications” of a rule derived primarily from an examination of the government’s reasons for acting). After all, regardless of what Heffernan did, the city construed his acts as protected activity and demoted him because of them.91×91. Setting aside, as the Court did for the purposes of its opinion, the possibility of the city’s having a neutral policy prohibiting all campaigning. See id. This illicit motive leads to constitutional harm for the “speaker” as well as, importantly, what Justice Breyer described as the “constitutional harm . . . in the ordinary case” — other citizens’ now being discouraged from speaking out.92×92. See Heffernan, 136 S. Ct. at 1419; Kagan, supra note 52, at 424–25 (discussing constitutional harms under speaker-based and audience-based models of the First Amendment); Michael L. Wells, Section 1983, the First Amendment, and Public Employee Speech: Shaping the Right to Fit the Remedy (and Vice Versa), 35 Ga. L. Rev. 939, 960 (2001) (“Public employee speech is a particularly vulnerable form of commentary, for in this context the risk is not merely public disapproval, but the loss of one’s livelihood.”).

The rationale behind previous public-employee-retaliation cases has been hard to pin down because of the many factors the Court takes into account. Heffernan, in which the only factor present was the government’s motive, suggests that motive matters more than the Court had let on. And the Court may focus on it especially in executive-action cases because ferreting out illicit government motive is easier in such cases than it is in legislative-intent cases. It is also more important, since government employers like the City of Paterson can send particularly powerful speech-chilling signals simply by demonstrating bad intentions, regardless of whether any protected speech actually occurred.