by Bex Rothenberg-Montz*

Although members of the general public enjoy a right to free speech under the First Amendment, government employees’ speech is more constrained. Courts determine whether a government employee’s speech is protected by balancing the interests of the government and the employee. However, in practice, the deference afforded to the government typically tips the scales. Because courts defer to the government’s interests without requiring substantiating evidence, the government is able to stifle employee speech and erode First Amendment protections for its employees. Consequently, this deference conditions free speech protections upon public employment status. Abandoning this deference will bring First Amendment jurisprudence in line with the principles that animated its enactment.

Over 20.9 million people work for federal, state, or local governments in the United States.1 Although these individuals do not explicitly forfeit their constitutionally-protected right to free speech upon accepting government employment,2 government employees enjoy fewer free speech protections than non-employees.3 The First Amendment’s protections do not apply to government employees’ speech when it is made pursuant to their official duties4 and the speech does not address a matter of public concern.5 Even when these two preconditions are met, courts still apply a balancing test to determine whether the government may curtail the speech.6

This balancing test was introduced by the Supreme Court in Pickering v. Board of Education, which announced that public employees have free speech protections and established the boundaries of those protections.7 In Pickering, the Court held that the government may not regulate its employees’ speech when “the interests of the [employee], as a citizen, in commenting upon matters of public concern” outweigh “the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.”8 To determine whether speech impairs the efficiency of the government’s public service, courts consider “whether the statement impairs discipline by superiors or harmony among co-workers, has a detrimental impact on close working relationships for which personal loyalty and confidence are necessary, or impedes the performance of the speaker’s duties or interferes with the regular operation of the enterprise.”9

The Court later expanded upon the Pickering balancing test in Connick v. Meyers and implicitly held that the government’s interests do not need to be supported by evidence.10 In Connick, Assistant District Attorney Meyers was unhappy with her supervisor’s decision to transfer her and circulated a questionnaire around her office soliciting her coworkers’ views about certain issues, including office transfer policy, office morale, and whether employees felt pressured to work on political campaigns.11 In response, the District Attorney dismissed Meyers.12 The government later justified the dismissal by claiming Meyers’ questionnaire was an “act of insubordination which interfered with working relationships.”13

The Court held that the government could silence public employees, if it “reasonably believed [the speech] would disrupt the office, undermine [supervisors’] authority, and destroy close working relationships.”14 However, the Court did not examine any evidence that Meyers’ speech caused such a disruption. Nonetheless, the Court gave “a wide degree of deference to the employer’s judgement.”15 Therefore, under the Connick test, courts “defer[] to a public employer’s mere anticipation of disruption as grounds for employee discipline” without requiring the government to present supporting evidence.16

Courts’ deference to the government is particularly harmful in the age of social media because courts take into account the public’s reaction in addition to that of coworkers and superiors when determining whether speech is disruptive.17 The potential for public disruption – and with it the credibility of the government’s “reasonable” belief of disruption – is made more likely by the lurking potential that government employees’ social media posts will go viral.18 Under the Court’s current guidance, the government can cite an amorphous “reasonable” belief that either the public or government employees will have a negative reaction to speech and a court will rely upon these interpretations without investigation.19


This deference is in stark contrast with free speech cases involving members of the general public, where courts do not defer to the government’s prediction of harm.

For example, in Cornelius v. NAACP Legal Defense and Educational Fund, Inc., the National Association for the Advancement of Colored People (“NAACP”) claimed the federal government violated the First Amendment by excluding it from a charity drive aimed at federal employees, the Combined Federal Campaign (“CFC”).20 Under First Amendment jurisprudence, the government may regulate the public’s speech in a non-public forum – which includes publicly owned facilities that have never been designated for indiscriminate expressive activity by the general public.21 Free speech restrictions in a nonpublic forum must be “reasonable in light of the purpose served by the forum” and “viewpoint neutral.”22 Since the Connick court found the CFC was a nonpublic forum, it addressed whether the government’s restriction was “reasonable in light of the purpose served by the forum.”23 Rather than deferring to the government’s claims that allowing for the NAACP to campaign at the CFC would cause “controversy” and “unwelcome disruption” which might jeopardize the CFC,24 the Court conducted its own analysis. It looked to contemporaneous expressions of concern about including “political” charities in the CFC, citing complaints about the inclusion of such charities in the CFC, and pointing to evidence that the government was struggling to obtain employee contributions.25 Only after a thorough discussion of the government’s evidence of disruption did the Court find the restriction to be “reasonable.”26

Like government workplaces, nonpublic forums are not generally open to the public and are places where the government has a significant interest in maintaining control.27 However, courts scrutinize government claims of potential disruption in cases where non-public employees speak28 without extending this scrutiny to public employees.29


The differential treatment between public employees and members of the general public is in conflict with the principles underlying Pickering. Prior to Pickering, courts reasoned that it was permissible to impose speech restrictions that would be unconstitutional when inflicted upon members of the general public because public employment was voluntary.30 Pickering renounced this assumption, holding that “[t]he theory that public employment which may be denied altogether may be subjected to any conditions, regardless of how unreasonable, has been uniformly rejected.”31 Instead, the Court reasoned that the government’s restriction was not permissible because “the interest of the school administration in limiting [the employee’s] opportunities to contribute to public debate is not significantly greater than its interest in limiting a similar contribution by any member of the general public.”32 By requiring the government to justify its restrictions, the Court established that government employees do not have limited free speech rights because of their status as public employees.33 In fact, it demonstrated quite the opposite: that, at baseline, public employees must be treated like the general public.34 By applying different levels of deference to the government’s assertion of disruption for public employee and non-employee speech, the Connick court breaks its own rule.


Connick’s deference should be revoked because it is inconsistent with the three principal justifications for free speech. First, protecting unpopular speech is fundamental to the judiciary’s goal of preserving a rich marketplace of ideas, in which the First Amendment allows for truth to be revealed through free and fair dialogue.35 Under the market-place of ideas theory, protecting public employees’ free speech ensures robust public discourse in which many diverse voices can debate and compete with one another.36 The Court has already acknowledged “the importance of promoting the public’s interest in receiving the well-informed views of government employees engaging in civic discussion,”37 because public employees are “likely to have informed opinions as to the operations of their public employers, operations which are of substantial concern to the public.”38 Connick’s deference undermines this justification for free speech rights by silencing the views of individuals who have special insight into government functioning.

Second, protecting unpopular speech protects democracy.39 Democratic governance is participatory and therefore contingent upon some level of free speech.40 There is near-universal agreement that free speech is required to maintain a democracy.41 When elections are regularly won by less than 5% of the vote,42 allowing the party in power to silence 8% of the population43 without justification is contrary to the democratic process.

Finally, protecting free speech rights safeguards the rights of personal expression and dignity.44 Public employees have the same dignitary interests as members of the general public.45 A jurisprudence which allows the government to regulate 20.9 million people’s speech with impunity, by requiring no evidence to support its assertions, works counter to those individuals’ dignitary interests by any measure. These three principles can be safeguarded once more by rejecting Connick’s deference.

In Connick, the Supreme Court diluted the Pickering balancing test by requiring absolute deference to governmental claims of even anticipated disruption. This deference undermines the premise that free speech protections are universal. Extending to public employees the scrutiny exacted when the government regulates the general public’s speech would allow courts to safeguard the principles underlying free speech protections, rather than abdicating this role through undue deference.

* Bex Rothenberg-Montz is a J.D. Candidate (2023) at New York University School of Law. This Contribution arose from the problem presented at the 2022 University of Wisconsin-Madison Law School Evans Moot Court Competition and examines an inconsistency discovered while researching the competition’s second question – when is a government employee’s speech protected by the First Amendment?

1. 2021 ASPEP Datasets & Tables, Census Bureau, (reporting 13.580 million local government employees and 5.248 million state government employees in 2021); Cong. Rsch. Serv., R43590, Federal Workforce Statistics Sources: OPM and OMB 1 (2022) (finding 2.1 million civilian federal employees).

2. See Pickering v. Bd. of Educ., 391 U.S. 563, 568 (1968) (observing that the theory “that teachers may constitutionally be compelled to relinquish the First Amendment rights they would otherwise enjoy as citizens to comment on matters of public interest in connection with the operation of the public schools in which they work . . . has been unequivocally rejected in numerous prior decisions of this Court”).

3. See, e.g., City of San Diego v. Roe, 543 U.S. 77, 80 (2004) (“[A] governmental employer may impose certain restraints on the speech of its employees, restraints that would be unconstitutional if applied to the general public.”).

4. See Garcetti v. Ceballos, 547 U.S. 410, 421 (2006) (“We hold that when public employees make statements pursuant to their official duties, the employees are not speaking as citizens for First Amendment purposes, and the Constitution does not insulate their communications from employer discipline.”).

5. See Connick v. Myers, 461 U.S. 138, 146 (1983) (“[I]f [the employee’s speech] cannot be fairly characterized as constituting speech on a matter of public concern, it is unnecessary for us to scrutinize the reasons for her discharge.”).

6. See Pickering, 391 U.S. at 568.

7. Id.

8. Id.

9. Rankin v. McPherson, 483 U.S. 378, 388 (1987). See also, e.g., Grutzmacher v. Howard Cnty., 851 F.3d 332, 345–46 (4th Cir. 2017) (finding that a fire department commander’s Facebook comments about gun violence were not protected speech because they “impaired Department operations and discipline as well as working relationships within the Department,” and “significantly conflicted with Plaintiff’s responsibilities as a battalion chief”); Volkman v. Ryker, 736 F.3d 1084, 1092 (7th Cir. 2013) (holding that a corrections officer’s criticism of an internal investigation was not protected because it undermined “maintaining order and respect” among correctional officers); Boyd v. Miss. Dep’t of Pub. Safety, 751 F. App’x 444, 450 (5th Cir. 2018) (holding the same where a police officer’s email suggesting promotions favored non-white employees because it “interfered with the operations of the department . . . [when] the email created a ‘racial ruckus’ and [] members of the SWAT team expressed concerns that they did not feel safe operating with [the officer]”).

10. Connick, 461 U.S. at 152.

11. Id. at 140–41.

12. Id. 141.

13. Id. at 151.

14. Id. at 154.

15. Id. at 152; see also Waters v. Churchill, 511 U.S. 661, 673 (1994) (plurality opinion) (noting that courts give “substantial weight to government employer’s reasonable predictions of disruption”).

16. Toni M. Massaro, Significant Silences: Freedom of Speech in the Public Sector Workplace, 61 S. Cal. L. Rev. 1, 4 (1987).

17. See Pickering, 391 U.S. at 570 (assessing whether the teacher’s speech would “foment controversy and conflict among the Board, teachers, administrators, and the residents of the district”); Bennett v. Metro. Gov’t of Nashville & Davidson Cnty., 977 F.3d 530 (6th Cir. 2020) (finding the government’s interest in harmony among coworkers and not being perceived as racist outweighed a 311 operator’s interest in using racial slurs on Facebook).

18. Christina Jaremus, #FiredforFacebook: The Case for Greater Management Discretion in Discipline or discharge for Social Media Activity, 42 Rutgers L. Rec. 1, 34 (“Internet postings would be considered presumptively disruptive to the government’s ability to provide efficient and effective services to the public due to their ability to go viral.”).

19. E.g., Graziosi v. City of Greenville, 985 F. Supp. 2d 808, 814–15 (N.D. Miss. 2013), aff’d sub nom. Graziosi v. City of Greenville Miss., 775 F.3d 731 (5th Cir. 2015) (holding in the alternative that a police officer’s post on social media criticizing the department’s failure to pay for officers’ gasoline was not protected speech because it created a “buzz” around the office).

20. 473 U.S. 788, 790 (1985).

21. See 1 Rodney A. Smolla, Smolla & Nimmer on Freedom of Speech § 8:8 (3d ed. 1996); see also, e.g., Minn. Voters All. v. Mansky, 138 S. Ct. 1876, 1886 (2018) (holding a polling place is a nonpublic forum); Int’l Soc. for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672, 680 (1992) (finding an airport terminal is a nonpublic forum); Greer v. Spock, 424 U.S. 828, 838 (1976) (holding a military base is a nonpublic forum); Adderley v. Fla., 385 U.S. 39, 47–48 (1966) (finding a jail is a nonpublic forum).

22. Cornelius, 473 U.S. at 806 (citing Perry Educ. Ass’n v. Perry Loc. Educators’ Ass’n, 460 U.S. 37, 49 (1983)).

23. Id.

24. Id. at 807.

25. Id. at 810.

26. Id. at 811–13.

27. See Perry Educ. Ass’n, 460 U.S. at 46 (“[T]he State, no less than a private owner of property, has power to preserve the property under its control for the use to which it is lawfully dedicated.” (quoting U.S. Postal Serv. v. Council of Greenburgh Civic Ass’ns, 453 U.S. 114, 129 (1981)).

28. See, e.g., Ne. Pa. Freethought Soc’y v. Cnty. of Lackawanna Transit Sys., 938 F.3d 424, 439 (3d Cir. 2019) (“[W]hen the state seeks to ban particular topics for fear of public controversy, it must make a showing of threatened disruption.”); Tucker v. Cal. Dep’t of Educ., 97 F.3d 1204, 1211 (9th Cir. 1996) (explaining that the government’s restriction on speech is unreasonable when the government “makes at most only a minimal showing that one individual’s speech has disrupted the workplace, or threatens to do so”).

29. See, e.g., Waters v. Churchill, 511 U.S. 661, 673 (1994) (plurality opinion) (“[W]e have consistently given greater deference to government predictions of harm used to justify restriction of employee speech than to predictions of harm used to justify restrictions on the speech of the public at large.”); Kokkinis v. Ivkovich, 185 F.3d 840, 845 (7th Cir. 1999) (“Deference to the employer’s judgment regarding the disruptive nature of an employee’s speech is especially important in the context of law enforcement.”).

30. See, e.g., McAuliffe v. City of New Bedford, 29 N.E. 517, 517 (Mass. 1892) (Holmes, J.) (“The petitioner may have a constitutional right to talk politics, but he has no constitutional right to be a policeman.”).

31. Pickering v. Bd. of Educ., 391 U.S. 563, 568 (1968) (quoting Keyishian v. Bd. of Regents, 385 U.S. 589, 605–06 (1967)).

32. Id. at 573.

33. See Randy J. Kozel, Free Speech and Parity: A Theory of Public Employee Rights, 53 Wm. & Mary L. Rev. 1985, 2011 (2012) (“By rejecting the Holmesian view and its singular focus on the existence of an employment relationship, the Court rendered that fact irrelevant to the constitutional calculus. No longer is it sufficient to answer the question, ‘Why do I have weaker First Amendment rights than my peers?’ with the response, ‘Because you work for the government.’”).

34. See Kozel, supra note 33 (“[E]mployees and other citizens are presumed to be similarly situated for purpose of the First Amendment.”).

35. See John Stuart Mill, On Liberty 34 (Batoche Books 2001) (1859) (“[I]f [an opinion] is not fully, frequently, and fearlessly discussed, it will be held as a dead dogma, not living truth.”); Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290, 295 (1981) (“The Court has long viewed the First Amendment as protecting a marketplace for the clash of different views and conflicting ideas”); Abrams v. United States, 250 U.S. 616, 630 (1919) (Holmes, J., dissenting) (“[T]he best test of truth is the power of the thought to get itself accepted in the competition of the market.”).

36. See Brett G. Johnson, The Heckler’s Veto: Using First Amendment Theory and Jurisprudence to Understand Current Audience Reactions Against Controversial Speech, 21 Commc’n. L. & Pol’y 175, 207 (2016) (“The theory of the marketplace of ideas holds that the First Amendment is not a means to effective democratic governance, but to attaining truth.”).

37. Garcetti v. Ceballos, 547 U.S. 410, 419 (2006).

38. City of San Diego v. Roe, 543 U.S. 77, 82 (2004). Compare Bennett v. Metro. Gov’t of Nashville & Davidson Cnty., 977 F.3d 530, 539 (6th Cir. 2020) (allowing restriction when the public employee commented on a topic “of which she had no special insight”), cert. denied, 141 S. Ct. 2795 (2021), and Kokkinis v. Ivkovich, 185 F.3d 840, 844 (7th Cir. 1999) (allowing restriction when a police officer’s “basis for his knowledge of the alleged sex discrimination was minimal at best.”), with Cotarelo v. Vill. of Sleepy Hollow Police Dep’t, 460 F.3d 247, 252 (2d Cir. 2006) (finding that a police officer’s allegation of racial discrimination was “protected activity” when the officer provided specific examples of discrimination and had experienced discrimination himself).

39. Robert C. Post, Racist Speech, Democracy, and the First Amendment, 32 Wm. & Mary L. Rev. 267, 279 (“In protecting public discourse the first amendment serves the purposes of democracy.”).

40. James Weinstein, Participatory Democracy as the Central Value of American Free Speech Doctrine, 97 Va. L. Rev. 491, 498 (2011) (“If an individual is excluded from participating in public discourse because the government disagrees with the speaker’s views or because it finds the ideas expressed too disturbing or offensive . . . then to that extent and with respect to that citizen, the government is no democracy, but rather an illegitimate autocracy.”).

41. Id. at 497 (citing Mills v. Alabama, 384 U.S. 214, 218 (1966) (“Whatever differences may exist about interpretations of the First Amendment, there is practically universal agreement that a major purpose of that Amendment was to protect the free discussion of governmental affairs.”)). But see Toni M. Massaro & Helen L. Norton, Free Speech and Democracy: A Primer for Twenty-First Century Reformers, 54 U.C. Davis L. Rev. 1631, 1639 (2021) (arguing the abundance of speech resulting from technological advances undermines democracy); Alexander Tsesis, Dignity and Speech: The Regulation of Hate Speech in a Democracy, 44 Wake Forest L. Rev. 497, 516–18 (arguing broad allowances for hate speech undermine democratic values). See generally Michel Rosenfeld, Hate Speech in Constitutional Jurisprudence: A Comparative Analysis, 24 Cardozo L. Rev. 1523 (2003).

42. Election Results, 2020: Congressional Elections Decided by 10 Percentage Points or Fewer, Ballotpedia (Feb. 8, 2021),,_2020:_Congressional_elections_decided_by_10_percentage_points_or_fewer (“In 2020, 42 congressional races (five Senate and 37 House) were decided by less than a 5 percent margin.”).

43. The 20.9 million public employees, 2021 ASPEP Datasets & Tables, supra note 1, compose 8.1% of the adults in the United States. See Stella U. Ogunwole et al., Population Under Age 18 Declined Last Decade, Census Bureau (Aug. 12, 2021), (258.3 million people living in the United States are adults).

44. See Alexander Tsesis, Free Speech Constitutionalism, 2015 U. Ill. L. Rev 1015, 1028 (2015) (“One of the most often stated rationales for protecting free speech is society’s obligation to safeguard the right of thoughtful and articulate persons to communicatively exercise their intellectual capacities.”); Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U.S. 749, 787 (1985) (Brennan, J., dissenting) (“[F]reedom of expression is . . . intrinsic to individual liberty and dignity.”). But see Tsesis, supra, at 1029 n.58 (explaining free speech restrictions other democracies impose to protect individual’s humanity).

45. See Tsesis, supra note 44, at 1028 (“[S]peech is a dignitary interest of each autonomous human being.”)