Introduction: The Endemic Problem of Self-Serving Politicians Seeking to Condition Government Employment on Public Servants Ceding Their Constitutional Right to Freedom of Expression

Since the dawn of democracy and democratic self-government, politicians inevitably and invariably covet praise and despise criticism. It is in their nature to claim the credit when things go well but to seek to escape the blame (and responsibility) when things go badly. When public criticism is particularly well-informed—and hence effective—because a government employee publicly discloses highly relevant, deeply embarrassing information held only by the government itself, the criticism stings even more sharply (because it has the ring of truth).

Structural political dynamics associated with seeking re-election will lead politicians, of every ideological stripe, to attempt to silence government employees by prohibiting them from providing the body politic with particularly well-informed criticisms of the government and its policies.1 1. Ronald J. Krotoszynski, Jr., Whistleblowing Speech and the First Amendment, 93 Ind. L.J. 267, 270 (2018) (“Since time immemorial . . . government officers will race to claim responsibility for successes but are far more reticent to acknowledge—much less take responsibility for—government failures”) [hereinafter Krotoszynski, Whistleblowing Speech and the First Amendment]. Thus, it should not be surprising that in democracies around the world, persons holding elected office routinely adopt laws and regulations designed to stifle public criticism from those best positioned to empower voters with the information that they require in order to use their ballots to secure government accountability.2 2. Ronald J. Krotoszynski, Jr., Transparency, Accountability, and Competency: An Essay on the Obama Administration, Google Government, and the Difficulties of Securing Effective Governance, 65 U. Miami L. Rev. 449, 454, 469 (2011) (noting that “systemic failures of governance are not particularly rare, which is a very good reason indeed to spend considerable time and energy thinking about issues associated with administrative competence” and arguing that “all presidential administrations, regardless of political party, are prone to suppress bad news whenever possible”).

Simply put, politicians will attempt to leverage the accident of working for the government to impose an omerta on government employees regarding the public disclosure of true, but highly embarrassing, information. Indeed, to not do so, when such self-serving action would otherwise be lawful, arguably constitutes a form of political malpractice. In consequence, judges have a critical role to play in safeguarding the speech rights of government employees from governments bent on compelling their silence as the price of working for the state.3 3. See, e.g., McAuliffe v. Mayor of New Bedford, 155 Mass. 216, 220, 29 N. E. 517, 517 (1892) (holding that John J. McCauliffe, a local police officer, “may have a constitutional right to talk politics, but he has no constitutional right to be a policeman”). Justice Oliver Wendell Holmes, Jr., then serving on the Supreme Judicial Court of Massachusetts, framed government employment as a mere privilege that government employers could, if they wished, condition on the cession of government employees’ constitutional right under the First Amendment to participate in the process of democratic deliberation. See id. (“There are few employments for hire in which the servant does not agree to suspend his constitutional rights of free speech as well as of idleness by the implied terms of his contract. The servant cannot complain, as he takes the employment on the terms which are offered him.”). Applying these principles to the case at bar, Holmes concluded that “[o]n the same principle the city may impose any reasonable condition upon holding offices within its control.” Id. This approach generally corresponds to—and frames—how the federal courts treated government employer restrictions on the expressive activities of public employees from December 8, 1791 (the date when the First Amendment became part of the federal Constitution) to June 3, 1968, the date that the Supreme Court issued its landmark decision in Pickering v. Board of Education, 391 U.S. 563 (1968). Compare Pickering, 391 U.S. at 568 (holding that “teachers may [not] constitutionally be compelled to relinquish the First Amendment rights they would otherwise enjoy as citizens to comment on matters of public interest” and explaining that “[t]he problem in any case is to arrive at a balance between the interests of the teacher, as a citizen, in commenting upon matters of public concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees”) with Adler v. Bd. of Educ., 342 U.S. 485, 492 (1952) (rejecting any constitutional protection for the speech and association rights of state government employees and holding that “[i]t is clear that such persons have the right under our law to assemble, speak, think and believe as they will” but “[i]t is equally clear that they have no right to work for the State in the school system on their own terms”); United Pub. Workers of Am. v. Mitchell, 330 U.S. 75, 94–96, 100 (upholding the Hatch Act’s prohibition of most forms of political activity by federal civil service employees against First Amendment objections and explaining that “Congress may reasonably desire to limit party activity of federal employees so as to avoid a tendency toward a one-party system”).

In the absence of some sort of external constraint, all of the relevant political incentives run in favor of imposing a code of silence on government employees as a condition of their employment at a government office. The coerced silence of public employees who possess information relevant to voters, in turn, means that voters will not have the information requisite to render prudent electoral choices.4 4. See Heidi Kitrosser, The Special Value of Public Employee Speech, 2015 Sup. Ct. Rev. 301, 330 (2015) (arguing that “government employees are crucial safety valves for protecting the people from abuse and incompetence, given their unique access to information and to a range of avenues for transmitting the same”). Voters cannot hold the government accountable for misdeeds they know nothing about.5 5. Obviously, there is an important role for the institutional press in making relevant information available to voters. As the saying goes, “if a tree falls in the forest, but no one sees it. . .” Democratic self-government relies on a feedback loop involving government officials, the press, and voters—with the press empowering voters with the information necessary to use the electoral process as a means of holding their government accountable. See David A. Anderson, The Origins of the Press Clause, 30 UCLA L. Rev. 455, 537 (1983) (positing that for the framers, including James Madison, who drafted the Bill of Rights, including the First Amendment, “freedom of the press was inextricably related to the new republican form of government and would have to be protected if their vision of government by the people was to succeed”); see Sonja R. West, Press Exceptionalism, 127 Harv. L. Rev. 2434, 2434–37 (2014) (arguing that the press can and does play a critical role in the process of democratic self-government and empowers voters to hold the government accountable through the electoral process). When the press reports truthful information provided by government employees about government misconduct, voters can and will use that information in assessing their electoral choices; in a real and meaningful way, information, in the hands of the voters, is power.6 6. See Sonja R. West, Awakening the Press Clause, 58 UCLA L. Rev. 1025, 1033, 1041–47, 1069–70 (2011) (noting “the common intuition that there does exist a press that performs a special role in our democracy,” positing that performance of this important role “is deserving of constitutional status outside the shadow of the Speech Clause,” and arguing that the press must be able to access and disseminate truthful information so that voters can use the electoral process to secure government accountability).

On the Necessity—and the Difficulty—of Reconciling the Conflicting Constitutional Imperatives of the Government as Manager and Government Employees as Citizens

An important question that necessarily arises from these general principles of democratic self-government: Does a constitutional commitment to the freedom of expression in a democracy require the conveyance of meaningful constitutional free speech protections to government employees who blow the whistle on misconduct or malfeasance within the government office that employs them? In my view, the answer to this question is clearly “yes.” As Professor Helen Norton cogently argues, government employees empower We the People to hold government accountable when they blow the whistle, and “[w]e lose this essential check when the [Supreme] Court interprets the Constitution to permit the government to fire its workers for telling the truth about their jobs—particularly when that truth includes their reports of government mismanagement, deceit, or corruption.”7 7. Helen Norton, The Government’s Speech and the Constitution 66–67 (2019).

If the potential adverse consequences of a government employee choosing speech over silence are too severe, then most public employees will embrace a posture of prudent silence in order to avoid those adverse professional consequences—up to and including the loss of their government jobs (coupled with significantly diminished employment prospects going forward). The balance of this essay constitutes an extended argument that judicial protection of whistleblowing speech is essential to the effective use of the electoral process as a principal means of securing government accountability.

It is obvious, on the one hand, that democracy suffers when public employees cannot participate in the process of democratic deliberation without putting their livelihoods at risk. As Professor Norton observes, in the contemporary United States, the “[f]ederal, state, and local governments employ more than 20 million workers,” and “public employees’ speech is often important not only to the employees themselves but also to the general public.”8 8. Id. at 60. If government employment may constitutionally be conditioned on public workers surrendering their expressive freedoms under the First Amendment, the damage to the process of democratic deliberation would be vast.

On the other hand, however, citizens rightfully expect the government to function and to do so on a reliable basis. As Professor Robert C. Post explains, “[m]anagerial authority over speech is necessary for an institution to achieve [its] goals” and, accordingly, “[a] government institution’s interest in internally regulating speech is therefore its interest in the attainment of the very purposes for which it has been established.”9 9. Robert C. Post, Constitutional Domains: Democracy, Community, Management 236 (1995). What is more, he adds that “this [government] interest remains the same whether or not its [employees] consent to the exercise of government authority.”10 10. Id.

As Professor Margaret O’Brien notes, “[t]he goal of civil service neutrality is widely recognized as important, and in restricting [the] political activity of public servants, there is an implied assumption that these restrictions are a prerequisite to political impartiality.”11 11. Margaret O’Brien, Public Employees Restrictions on Political Activity in Canada, Australia, and the United Kingdom, 15 Int’l & Comp. L.J. 319, 328 (2021). Even so, however, she adds that “there is disagreement on the necessity of formal regulations or laws to achieve this end.”12 12. Id. It bears noting that Professor Kathleen Sullivan agrees that the government has a legitimate managerial necessity for regulating government employees’ speech while on the job and that such regulations should not automatically be deemed inconsistent with the First Amendment. See Kathleen M. Sullivan, Unconstitutional Conditions, 102 Harv. L. Rev. 1413, 1459–61, 1503–04 (1989). She explains that “conditions on public employee speech also may be upheld because some overriding government purpose justifies them, not because they do not pressure rights.” Id. at 1503. The key for Professor Sullivan, as for Professor Post, is that a reviewing court must impose a meaningful burden of justification on the government employer’s invocation of a managerial necessity to regulate or ban public servants’ speech rights. See id. at 1504 (arguing that limits on government employee speech, such as civil service regulations that limit partisan political activity, “should be treated as infringing speech and thus in need of strong justification, but as arguably justified by the need for an efficient or depoliticized bureaucracy”).

Professor Post goes on to explain that “[w]hen the state acts to govern the speech of the general public, it often truncates the very process of discussion and exchange by which public ends and public actions are determined,”13 13. Post, supra note 10, at 265. thereby compromising—violating—core First Amendment values and social objectives. By way of contrast, however, “[w]hen the state acts internally to manage restrictions within its own institutions . . . public ends are taken as given.”14 14. Id. at 266. The question in any given case regarding whether government employee speech should enjoy First Amendment protection thus depends critically on whether the government is simply minding its own store or is instead attempting to distort or disrupt the process of democratic deliberation essential to the process of democratic self-government. Thus, Professor Post concludes that “First Amendment restrictions on the internal management of speech thus turn in the main on whether the management is necessary in order to attain organizational purposes.”15 15. Id.

In sum, the government clearly possesses a legitimate “managerial domain,”16 16. Id. at 242–44, 265–66. Professor Helen Norton argues that, in general, the government should enjoy a categorical power to regulate public employees’ speech only when those employees are specifically hired to speak for the government itself: “A rule that better accommodates the weighty interests at stake here would permit the government to claim the power to control the speech of its employees as its own only when it has specifically commissioned or hired those employees to deliver a transparently governmental viewpoint for which the public can hold it accountable.” Norton, supra note 8, at 64. In all other circumstances, Norton posits that “public employee expression that does not meet this demanding transparency-based test for government speech should continue on to the long-standing balancing test that the Court applied to public employees’ First Amendment claims for decades before Garcetti.” Id. at 65. This approach would seem to be more protective of public worker’s speech rights than Professor Post’s proposed solution, which would allow government employers to restrict government employees’ speech rights in the service of legitimate managerial objectives. See Post, supra note 10, at 261–66. and this managerial domain is essential to achieving important government policies across myriad areas—including public safety, health care, education, defense, and national security. If the First Amendment conveyed absolute protection on government employees to speak whenever and however they wished, within or outside government workplaces, the business of government could not, and would not, go on.17 17. Bi-Metallic Invs. Co. v. Colorado, 239 U.S. 441, 445 (1915) (“There must be a limit to individual argument in such matters if government is to go on.”). To be sure, in his Bi-Metallic opinion, Justice Holmes was writing about the specific issue of an individualized right to be heard in a legislative proceeding involving a proposed general property tax increase—as opposed to a right to be heard in an individualized proceeding to ascertain the specific value of a particular parcel for the determination of an individual property owner’s property tax liability for the cost of municipal improvements. See id. at 443–45. But cf. Londoner v. Denver, 210 U.S. 373, 380–81 (1908) (holding that a property owner has a right to be heard when a municipal government apportions the cost of road and sidewalk improvements to specific parcels of land). The observation, however, also holds equally true when the government, as a manager, seeks to organize government workplaces in order to best achieve the offices’ purposes and objectives. Such an approach would be completely untenable; no constitutional court would write it into a domestic legal system’s constitutional law.

Striking and then holding an appropriate balance is, accordingly, essential. Government must be able to organize its workplaces, and workforces, efficiently and effectively. However, at the same time, the price of government employment cannot include a complete and total loss of government workers’ First Amendment rights as citizens and participants in the project of democratic self-government.18 18. See Kitrosser, supra note 5, at 328–30 (arguing that a categorical rule that excludes any public employee speech that arguably falls within the scope of the speaker’s on-the-job duties will create an unacceptable chilling effect that denies We the People information essential to securing government accountability via the electoral process); see generally Norton, supra note 8, at 64–67 (positing that First Amendment rules should not categorically exclude public employees’ speech from First Amendment protection unless they are clearly speaking for the government itself in some sort of official capacity). Striking this balance and then successfully maintaining it over time constitute critically important—but also extremely difficult—judicial tasks.

If judges adopt a categorical approach to these questions, they will invariably strike a balance that unduly favors the government as a manager. Yet, if government employees must be able to maintain their agency as speakers and participants in democratic deliberation, a balance that simply declares “the government as an employer and manager always wins” simply will not do. In consequence, any test for establishing and enforcing the requisite constitutional boundary line will necessarily involve a careful, contextual, and highly individualized judicial effort to reconcile the respective conflicting interests at stake. A categorical approach will also likely disregard completely the collective interest that voters have in access to government employee speech in general and whistleblowing government employee speech in particular.19 19. By “whistleblowing speech,” I mean speech that discloses to the general public important information about the operation of a government agency, and especially information that relates to illegal activity, misconduct, or malfeasance within the government agency where the whistleblowing government employee works. Whistleblowing government employee speech, unlike more generic forms of government employee speech, “possesses an essential nexus to the electoral process’s core function of holding government accountable to the electorate for its actions.” Krotoszynski, Whistleblowing Speech and the First Amendment, supra note 2, at 302. This characteristic demonstrates the public good nature of government employees’ whistleblowing speech; it has social value not merely because it facilitates the ability of public workers to exercise agency as citizens by participating in the project of democratic deliberation but has heightened social value because it empowers We the People to use our votes to enforce government accountability. See Julian W. Kleinbrodt, Note, Pro-whistleblower Reform in the Post-Garcetti Era, 112 Mich. L. Rev. 111, 113 (2013) (arguing that “[w]histleblower speech is critically important because it helps ensure a well-functioning democracy” and observing that information provided by government employees can be crucial to securing accountability); Diane Norcross, Comments: Separating the Employee From the Citizen: The Social Science Implications of Garcetti v. Ceballos, 40 Balt. L. Rev. 543, 570–72 (2011) (arguing that whistleblowing speech can greatly facilitate the ability of voters to hold the government accountable for its actions); see also Krotoszynski, Whistleblowing Speech and the First Amendment, supra note 2, at 272–75 (distinguishing whistleblowing government employee speech from more generic forms of government employee speech and discussing its potential importance to the process of democratic deliberation).

The Problems with Pickering (and Connick, Garcetti, and Roe to Boot)

The U.S. Supreme Court’s efforts to draw and enforce a workable line of division between the speech rights of government employees and the managerial necessities of government employers presents something of a mixed jurisprudential picture. Pickering v. Board of Education,20 20. 391 U.S. 563 (1968). decided in 1968, adopts a balancing test that weighs the government’s interest in managing its workplaces against the First Amendment rights of government workers.21 21. Id. at 568 (“The problem in any case is to arrive at a balance between the interests of the teacher, as a citizen, in commenting upon matters of public concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.”). Pickering’s approach, as developed in subsequent Supreme Court decisions, (1) overtly weighs only two of the three relevant interests at stake in cases involving speech by public employees and largely ignores the wider body politic’s interest in government employee speech, (2) fails to disentangle generic government employee speech from the distinguishable subset of government employee whistleblowing speech, (3) systematically underprotects government employee speech by applying a narrow test of “matter of public concern” before affording such speech any protection under the Free Speech Clause,22 22. See Connick v. Myers, 461 U.S. 138, 147–49 (1983) (holding that only speech related to a matter of public concern, narrowly defined to exclude speech related to workplace policies and working conditions, triggers the application of Pickering’s protection of government employee speech); see also City of San Diego v. Roe, 543 U.S. 77, 84 (2004) (holding that sexually-explicit performance art by an off-duty San Diego, California, police officer did not constitute speech about a matter of public concern and opining that the city’s police department could fire Officer Roe without having to meet any burden of justification whatsoever under the First Amendment). (4) categorically excludes from any First Amendment protection government employee speech that falls within the scope of a public servant’s work-related duties,23 23. Garcetti v. Ceballos, 547 U.S. 410, 426 (2006) (holding that Pickering does not apply if a government employee speaks about a matter of public concern incident to the employee’s job-related duties because “[o]ur precedents do not support the existence of a constitutional cause of action behind every statement a public employee makes in the course of doing his or her job”). and (5) arguably enables a kind of heckler’s veto24 24. Terminiello v. Chicago, 337 U.S. 1, 4–5 (1949) (holding that an audience’s hostile reaction to a speaker is not a valid basis, consistent with the Free Speech Clause of the First Amendment, for silencing an unpopular speaker); see Harry Kalven, Jr., The Negro and the First Amendment 140–60 (1965) (coining the phrase “heckler’s veto” and explaining, in some detail, why a hostile audience reaction to speech about a matter of public concern cannot serve as a constitutionally-acceptable basis for silencing a would-be speaker). because a government employee’s right to reinstatement is contingent on how other employees will react to the fired employee’s presence in the government workplace (meaning that if a government employee’s continued presence in the government workplace causes material disruption—or might cause material disruption—the government employer may fire the employee without violating the First Amendment’s Free Speech Clause).25 25. See Pickering, 391 U.S. at 572–73.

Perhaps most important is judicial recognition of the fact that the relevant interests that constitutional courts, both in the United States and elsewhere as well, must accommodate and reconcile are not limited to those of government employers and public employees. When whistleblowing speech is at issue, these interests also encompass an underappreciated, and insufficiently protected, collective interest on the part of We the People in possessing the information essential to holding government accountable through voting and elections. Thus, whistleblowing speech—that is, speech that discloses wrongful behavior by the government26 26. See supra note 20.— constitutes not only a form of self-expression by government employees but also a public good that benefits the body politic as a whole.

A coherent and comprehensive theory of freedom of speech in a democracy must take account of the value of whistleblowing speech to the voters. Unfortunately, in the United States today, First Amendment doctrine does not meaningfully distinguish between a government employee’s generalized personal autonomy interest in speaking truth to power and the community’s more granular, and arguably more important, collective interest in speech essential to using the electoral process to hold the government accountable for its actions.

Alexander Meiklejohn, arguably the leading proponent of the democratic self-government theory of the First Amendment, explains that “[w]hen a free man is voting, it is not enough that the truth is known by someone else, by some scholar or administrator or legislator.”27 27. Alexander Meiklejohn, Free Speech And Its Relation to Self-Government 88 (1948). Instead, “[t]he voters must have it, all of them.”28 28. Id. Democratic self-government cannot function properly if the voters lack the information necessary to make rational electoral decisions. Because of this fact of democratic life, Meiklejohn posits that “[t]he primary purpose of the First Amendment is, then, that all the citizens shall, so far as possible, understand the issues which bear upon our common life.”29 29. Id. at 88–89.

When the government imposes a code of silence on government employees, it denigrates both the government employee’s individual First Amendment interest in speaking and also the collective interest of the potential audience in hearing the speaker’s message.30 30. Kleindienst v. Mandel, 408 U.S. 753, 762–65 (1972); see Citizens United v. Fed. Elec. Comm’n, 558 U.S. 310, 341 (2010) (holding that the government may not “deprive the public of the right and privilege to determine for itself what speech and speakers are worthy of consideration” and that “[t]he First Amendment protects speech and speaker, and the ideas that flow from each”). On the one hand, denying government employees the ability to participate fully and freely as citizens plainly inflicts an individual First Amendment injury. On the other, however, the social cost of silencing government employees extends well beyond this personal autonomy interest; prohibiting them from sharing essential information with the body politic also imposes a collective First Amendment harm on the electorate that undermines the efficacy of elections as a means of securing government accountability.31 31. Anthony Davidson Gray, Public Servants and the Implied Freedom of Political Communication, 49 Fed. L. Rev. 3, 35–36 (2021) (arguing that “the government’s legitimate interests in terms of an independent, impartial and functional public service must be tempered by acknowledgement of the very valuable contribution that public servants can make to important public policy debates, from their specialised knowledge and expertise”). Thus, “a First Amendment violation actually involves harms beyond the speaker because free speech benefits not just the speaker but also the audience.”32 32. Ronald J. Krotoszynski, Jr. & Caprice L. Roberts, Reimagining First Amendment Remedies, 109 Iowa L. Rev. 911, 938 (2024). To date, however, the U.S. Supreme Court has done a relatively poor job of safeguarding either of these important interests under the aegis of the First Amendment’s Free Speech Clause.33 33. Id. at 919–20, 938–39 (arguing that government censorship imposes both individual and collective harms and that First Amendment remedies must take both kinds of harms into account).

Government employee speech claims involve not two but three distinct interests: the public servant’s interest in speaking, the government’s interest in managing its offices and workplaces, and the community’s interest in having access to the government employees’ speech (and particularly to whistleblowing government employee speech). Pickering needs to be refashioned to take account of all three interests—which Canadian constitutional law, as explained in the next principal part, already does.

Building a Better Mousetrap: Some Potential Lessons from Canada

Is a better approach possible? It is. In point of fact, Canada has adopted an approach, rooted in proportionality review,34 34. For an excellent general discussion, overview, and explanation of proportionality review and its centrality to many systems of enforcing human rights around the world, see Vicki C. Jackson, Constitutional Law in an Age of Proportionality, 124 Yale L.J. 3094 (2015). that conveys presumptive constitutional protection under the Charter of Rights and Freedoms35 35. Part I of the Constitution Act, 1982, Canada Act, 1982, ch. 11, sch. B, art. 2, § 2(b) (U.K.), containing the Canadian Charter of Rights and Freedoms [hereinafter Canadian Charter]. For a helpful and comprehensive discussion of the Canadian Charter and its introduction of an entrenched bill of rights coupled with strong-form judicial review, see Paul Bender, The Canadian Charter of Rights and Freedoms and the United States Bill of Rights: A Comparison, 28 McGill L.J. 811 (1983). For an overview of Canadian freedom of expression jurisprudence under Section 2(b) of the Charter, see Richard Moon, The Constitutional Protection of Freedom of Expression (2000). to any and all government employee speech.36 36. O’Brien, supra note 12, at 350 (praising Canada’s careful and context-sensitive approach to reconciling the government’s managerial needs with the constitutional free speech rights of government employees under Section 2(b) of the Canadian Charter of Rights and Freedoms and describing it as “perhaps the best in achieving a [proper] balance” of these competing and conflicting interests). However, under proportionality review, the government may still prevail if it can show that, on the particular facts and circumstances at bar, its action was prescribed by law (rather than merely ad hoc), reflects a rule or policy that is demonstrably necessary in a free and democratic society, and the government’s abridgment of the fundamental right at issue was reasonably tailored with respect to the means used to achieve the government’s pressing and substantial purpose.37 37. See R. v. Oakes, [1986] 1 S.C.R. 103, 136–40 (Can.); see also Ronald J. Krotoszynski, Jr., The First Amendment in Cross-Cultural Perspective: A Comparative Legal Analysis of the Freedom of Speech 41–45 (2006) (describing and discussing proportionality review in Canada under Section 1 of the Canadian Charter and the Oakes test).

The Supreme Court of Canada (SCC) conveyed meaningful protection on government employee speech related to matters of public concern even before the Charter of Rights and Freedoms (Charter) came into force in 1982. In Fraser38 38. Fraser v. Canada, [1985] 2 S.C.R. 455 (Can.). (a pre-Charter decision) and Osborne39 39. Osborne v. Canada, [1991] 2 S.C.R. 69 (Can.). (a post-Charter decision), the SCC made clear that government employees remain citizens and have a legitimate claim to the full protection of Section 2(b) of the Charter. The SCC consistently has engaged in a careful and context-sensitive analysis of legal restrictions that limit or prohibit expressive activity by public employees.40 40. See, e.g., id. at paras. 32–41, 56–61. For a discussion of the regulation of the Canadian civil service workforce, see Pub. Serv. Comm’n of Canada, The 100 Years of the Public Service Commission of Canada 1908–2008, Gov’t Canada, https://www.canada.ca/en/public-service-commission/services/publications/publications/100-years-public-service-commission-canada-1908-2008.html (last visited June 29, 2024).

Section 2 of the Charter provides:

2. Everyone has the following fundamental freedoms:

(a) freedom of conscience and religion;

(b) freedom of thought, belief, opinion and expression, including freedom

of the press and other media of communication;

(c) freedom of peaceful assembly; and

(d) freedom of association.41 41. Canadian Charter § 2.

Most obviously relevant to the question of government employee speech rights is Section 2(b), which safeguards “freedom of thought, belief, opinion and expression,” although Section 2(c), on freedom of assembly, and Section 2(d), on freedom of association, also could be relevant to the protection of government employee speech.

The SCC has defined protected speech in an extraordinarily broad fashion—any human activity designed to communicate a message, save for acts of violence and perhaps also threats of violence, comes within Section 2(b)’s scope of application.42 42. Moon, supra note 36, at 33–52 (describing and explaining the very broad scope of application of Section 2(b) as encompassing virtually any and all human activity intended to convey a message, save for acts of violence and perhaps also threats of violence). The potentially very high social cost of defining protected speech so broadly washes out at the second step of Charter analysis—proportionality review. Once a plaintiff establishes that the government has burdened or denied a protected Charter right, the burden shifts to the government to justify the restriction.

Section 1 of the Charter provides that “The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.”43 43. Canadian Charter § 1. To meet this standard under the governing Oakes test, in addition to having a written statute or regulation on the books (“prescribed by law”), the government must identify a “pressing and substantial” justification for the restriction and also show that the restriction is narrowly tailored to achieve the government’s important interest.44 44. See Moon, supra note 36, at 55–62 (discussing and critiquing Section 1’s application in freedom of expression cases under Section 2(b)); see also Oakes, [1986] 1 S.C.R. 103.

In Canada, a “Values and Ethics Code for the Public Sector” imposes content and viewpoint restrictions on civil servants—regardless of the potential value to voters of truthful information about wrongdoing or malfeasance within a government agency.45 45. For a thoughtful and comprehensive overview of the statutes and regulations that limit the ability of civil servants in Canada to engage in speech activity or stand for office, see O’Brien, supra note 12, at 329–35. The primary statute on point is the Public Service Employment Act (PSEA). Public Service Employment Act, SC 2003, c 22, ss12, 13; see O’Brien, supra note 12, at 330–31 (discussing the PSEA and its operation). Professor O’Brien notes that guidance to government employees “specifies that civil servants should consider the Values and Ethics Code for the Public Sector while carrying out political activities that are not explicitly included in this definition [of a candidacy for a public office].”46 46. O’Brien, supra note 12, at 332. As is the case in Australia, if a government employee’s speech might embarrass a government employer, the speech may serve as a basis for discipline, up to and including discharge. However, that is about as far as the parallels between Australian and Canadian regulation of government employees’ speech rights go.

The applicable regulations implementing the Public Service Employment Act (PSEA) direct a federal civil service employee to consider the nature of the political activity, the nature of the employee’s duties, the level and visibility of the employee’s public duties, and personal visibility.47 47. Pub. Serv. Comm’n of Canada, Political Activities and Non-Partisanship Directorate 2 (2014). Moreover, as Professor O’Brien observes, government “employees do not need to request nor receive permission before engaging in non-candidacy political activities.”48 48. O’Brien, supra note 12, at 332. Even so, a public employee who engages in public speech activity is “responsible for ensuring that their ability to perform their duties is not impaired, and does not appear to be impaired, by their engagement in political activity.”49 49. Id. In addition, the SCC has imposed a burden of justification on the government when it takes adverse action against a government employee based on protected speech activity.

In Fraser v. Canada,50 50. [1985] 2. S.C.R. 455 (Can.). an important pre-Charter case decided under a general principle that freedom of expression is an important social value that courts must take into account when the government acts in a way that burdens a citizen’s ability to speak, the SCC held that bizarre anti-metric ravings, as well as public opposition to adoption of the Charter, by Neil Fraser, a civil servant in Canada’s Department of National Revenue (Canada’s IRS) required a sufficient justification from the government to be lawful. After giving him multiple warnings to cease his embarrassing public appearances (which Fraser ignored), the Department of National Revenue fired him for bringing the agency into public disrepute.51 51. Id. at paras. 10–11. Fraser appealed this decision, arguing his government employer had violated his free speech rights.

Writing for a unanimous bench, Chief Justice Brian Dickson found that Fraser enjoyed a right to the protection of his off-the-job speech activity, even though neither the Canadian Charter nor an early, statutory bill of rights applied on these facts.52 52. Id. at para. 23. Despite the absence of a constitutional or even a statutory free speech guarantee, Chief Justice Dickson opined, “[t]hat is not to say, however, that this is not, at least in part, a ‘freedom of speech’ case. It is.”53 53. Id. Chief Justice Dickson observed that “‘freedom of speech is a deep-rooted value in our democratic system of government” and serves as “a principle of our common law Constitution, inherited from the United Kingdom by virtue of the preamble to the Constitution Act, 1867.”54 54. Id. at para. 24. Even so, Fraser’s speech deeply embarrassed his government employer, and “although there is not an absolute prohibition against public servants criticizing government policies, Mr. Fraser in this case went much too far.”55 55. Id. at para. 57. The balancing exercise accordingly favored the government.

The second landmark government employee speech case, Osborne v. Canada, arose after the Charter took effect (on April 17, 1982) and therefore directly applies Section 2(b). A group of federal employees, subject to the provisions of the PSEA,56 56. At the time, this statute appeared as R.S.C. 1985, c. P-33. Canada’s Parliament amended the law in 2003. It now appears as SC 2003, c. 22, ss. 12, 13. objected to being prohibited from engaging in electioneering (“political”) activity.57 57. Osborne v. Canada, [1991] 2 S.C.R. 69, paras. 2–9 (reciting the facts and claims of the plaintiffs). The trial court rejected the plaintiffs’ Section 2(b) claims, but the Federal Court of Appeal reversed, holding that the PSEA was unduly vague and did not constitute a “reasonable limit” on the expressive activities of government employees.58 58. See id. at para. 16. The SCC granted review in order to consider the constitutionality of the PSEA’s restrictions on persons working in Canada’s federal civil service. The SCC affirmed the decision of the Federal Court of Appeal.

Writing for the plurality, Justice John Sopinka easily found that the challenged provision of the PSEA, section 33, “which prohibits partisan political expression and activity by public servants under threat of disciplinary action including dismissal from employment, violates the right to freedom of expression in s. 2(b) of the Charter.”59 59. Id. at para. 32. Applying Fraser, Justice Sopinka concludes that in post-Charter Canada, the arguments in favor of conveying broad constitutional protection on government employees’ speech activities are both obvious and compelling. Moreover, he could not “see how it may be said that the oppugned [sic] section in the present case does not constitute a violation of the right to freedom of expression.”60 60. Id. at para. 41. He explains that “[b]y prohibiting public servants from speaking out in favour of a political party or candidate, it expressly has for its purpose the restriction of expressive activity.”61 61. Id.

Even though the federal government has an important and substantial interest in a politically neutral civil service, the PSEA’s restrictions on political speech went far beyond what was necessary to safeguard the government’s interest. Justice Sopinka observes that “[t]he result of this broad general language [in section 33] is that the restrictions apply to a great number of public servants who in modern government are employed in carrying out clerical, technical or industrial duties that are completely divorced from the exercise of any discretion that could be in any manner affected by political considerations.”62 62. Id. at para. 56. Clearly, “the need for impartiality and indeed the appearance thereof does not remain constant throughout the civil service hierarchy.”63 63. Id. Thus, Justice Sopinka concludes that “[t]o apply the same standard to a deputy minister and a cafeteria worker appears to me to involve considerable overkill and does not meet the test of constituting a measure that is carefully designed to impair freedom of expression as little as reasonably possible.”64 64. Id. In light of these considerations, “s. 33 fails the minimum impairment test” and was not saved by operation of Section 1.65 65. Id. at para. 61.

Canada’s jurisprudential approach differs significantly from that followed in both the United States and Australia because, in both theory and in practice, all government employee speech triggers constitutional protection under Section 2(b) of the Charter. For example, the SCC has held that anti-Semitic speech—and school lessons at an Eckville, Alberta, public high school—enjoyed protection under Section 2(b).66 66. R. v. Keegstra, [1990] 3 S.C.R. 697, 713–14, 725–34 (Can.). The fact that James Keegstra essentially abused his teaching position to inculcate hatred of Jewish persons did not exclude his expressive activity from constitutional protection.67 67. Id. at 734.

Canada’s framework for protecting the speech rights of government employees is arguably superior to the U.S. approach for several important reasons.68 68. See infra text and accompanying notes 102 to 108. Canada’s example clearly establishes that it is possible to convey broader constitutional protection for government employees’ speech without turning the courts into a free-form, general forum for government employees’ grievances against their public employers.69 69. See Waters v. Churchill, 511 U.S. 661, 671–77 (1994) (holding that because “the government as employer indeed has far broader powers than does the government as sovereign,” it may discipline a government employee based on the mistaken belief that an employee said something potentially disruptive to the government workplace, even when the employee did not make the statement, provided that the employer could “reasonably” have believed that the employee made the potentially disruptive statement). Canada has built a better mouse trap—and the U.S. federal courts should consider whether incorporating elements of the Canadian approach might provide a useful model for reforming how they weigh the speech rights of government employees, in conjunction with the collective interest of the body politic in government employee speech in general and whistleblowing speech in particular, against the legitimate managerial needs of government employers.

Considered together, Fraser and Osborne establish an open-ended, context-sensitive approach to reconciling the free expression claims of government employees with the managerial necessities of their government employers. Osborne, in particular, is a critically important case because it requires that regulations restricting government employees’ expressive freedoms possess a careful and precise means/end fit. It also bears noting that Keegstra, a hate speech case involving a public high school teacher,70 70. See supra text and accompanying notes 66–67. did not embrace some sort of categorical exclusion of low-value speech from the scope of Section 2(b) but instead simply considered the low value of the speech activity when engaging in proportionality analysis.

In Canada, any and all public employee speech enjoys constitutional protection under the Charter.71 71. Paul Clarke & Robyn Trask, Teachers Freedom of Expression: A Shifting Landscape – Part One – Critical Political Expression to Parents and Others, 22 Educ. & L.J. 303, 304–07, 323–26 (2013). The SCC has never adopted any categorical exclusions, such as a requirement that speech relate to a matter of public concern, or not be related to a government employee’s official duties or the like.72 72. Id. at 112–20. Canada, like the United States, does use a balancing test that weighs the government’s interest in regulating government employees’ speech against the autonomy interest of these individuals as citizens (as well as public servants). And, as with Pickering, the threat of disruption to a government workplace is a relevant factor in the balancing exercise. However, unlike Pickering, the SCC has made clear, in Chief Justice Dickson’s Fraser opinion, that government employee speech of a whistleblowing nature must receive broader and deeper judicial protection.73 73. Fraser, [1985] 2 S.C.R. 455, at para. 46 (holding that Canada’s federal government employees may criticize the government and “publicly express opposition” to its policies when “the Government [was] engaged in illegal acts, or if its policies jeopardized the life, health, or safety of the public servant or others”).

Things Could Be Worse than Pickering and Its Progeny: The Australian Example and the High Court’s Deeply Problematic Judgment in Banerji

Of course, looking in the opposite direction, things also could be worse than under Connick/Pickering. One could imagine a constitutional system that has a theoretical commitment to protecting the process of democratic deliberation essential to using elections as a means of conveying legitimacy on the government and its policies—but that utterly fails to protect the speech rights of government employees (regardless of the potential relevance of their speech to the process of democratic deliberation). As it happens, such a system is not merely hypothetical—Australia has adopted precisely this (very unfortunate) approach.74 74. See Comcare v. Banerji, (2019) H.C.A. 23, paras. 17, 35–38, 42 (Austl.) (Kiefel, C.J., Bell, Keane, & Nettle, JJ.) (upholding a statute and implementing regulations that prohibit criticism of the government if a government employer objects to an employee’s off-the-clock private speech as “unreasonably or harshly critical” of the government).

Despite holding, in a pair of landmark decisions in 1992, that the constitutional provisions in Australia’s 1901 Commonwealth Constitution guaranteeing voting rights75 75. Commonwealth of Australia Act, 1900 (Imp), 63 & 64 Victoria, c. 12, § 7, 24, 64 & 128 (U.K.), https://www.aph.gov.au/constitution (last visited August 23, 2024) (providing for elections and voting for the federal parliament and for amendments to the Commonwealth Constitution). also gave rise to an implied freedom of political and governmental communication (IFPGC),76 76. Australian Capital Television Pty Ltd v. Commonwealth, (1992) 177 C.L.R. 106 (Austl.); Nationwide News Pty Ltd v. Wills, (1992) 177 C.L.R. 1 (Austl.). and grounding this implied freedom on the structural necessity of a public debate that, borrowing Justice William J. Brennan, Jr.’s wonderful turn of phrase, is “uninhibited, robust, and wide-open” and “may well include vehement, caustic, and sometimes unpleasantly sharp attacks on government and public officials,”77 77. New York Times Co. v. Sullivan, 376 U.S. 254, 270 (1964). public employees who speak out about matters of public concern in Australia must do so at the risk of unemployment.

Moreover, under the current test for IFPGC claims, namely “structured proportionality,” if the government has a legitimate interest, and uses reasonably adapted means to achieve it, the High Court of Australia (HCA) will sustain the government’s speech regulation. This test veritably exudes deference to the elected branches of Australia’s federal and state governments.78 78. Ronald J. Krotoszynski, Jr., Free Speech as Civic Structure: A Comparative Analysis of How Courts and Culture Shape the Freedom of Speech 107–09 (2024) [hereinafter Krotoszynski, Jr., Free Speech as Civic Structure]; see Gray, supra note 32, at 29–31 (criticizing the weakness of the IFPGC standards of review as applied in Banerji and objecting that “I must respectfully disagree that there was no obvious and compelling alternative to achieving the government’s legitimate ends”). Indeed, the HCA held that, when applying proportionality analysis, the government’s preferred balance between regulation and speech should stand “unless the benefit sought to be achieved by the law is manifestly outweighed by the adverse effect on the implied freedom.”79 79. LibertyWorks Inc. v. Commonwealth, [2021] H.C.A. 18, para. 85 (Austl.) (Kiefel, C.J., Keane & Gleeson, JJ.); see Banerji, [2019] H.C.A. 23, at para. 38 (“If a law presents as suitable and necessary in the senses described, it is regarded as adequate in its balance unless the benefit sought to be achieved by the law is manifestly outweighed by its adverse effect on the implied freedom. In this case, that directs attention to the quantitative extent of the burden and the importance of the impugned provisions to the preservation and protection of the system of representative and responsible government mandated by the Constitution.”)

In Comcare v. Banerji,80 80. Banerji, [2019] H.C.A. 23, at paras. 17, 35–38, 42. the HCA had to decide whether Michaela Banerji’s off-the-clock criticisms of her employer, Australia’s Department of Immigration and Citizenship (DIC), provided a valid basis for her discharge under Australia’s Public Service Act (APSA)81 81. Public Service Act 1999 (Cth), §§ 10(1), 13(11), 15(1), 33(1). and the related Australian Public Service Code of Conduct (APSC).82 82. Id. at § 13 (setting out the APS Code) [hereinafter APS Code]. Banerji’s comments were highly critical of the government’s immigration policies under both Labor (Australia’s main progressive political party) and Liberal (Australia’s main conservative political party) governments. However, she offered her criticisms on an anonymous basis using a social media platform (Twitter, now known as X). Thus, a person reading Banerji’s posts would have had no idea (whatsoever) that the speaker actually worked for the DIC. Essentially, the question presented was whether Australia’s implied freedom of political and governmental communication protects off-the-job government employee speech about matters of public concern.

The APSC provides that classified employees “must at all times behave in a way that upholds: (a) the APS Values and APS Employment Principles; and (b) the integrity and good reputation of the employee’s Agency and the APS.”83 83. APS Code, § 13(11)(a) & (b). Implementing regulations provide more granular guidance on exactly what this means. The DIC’s guidelines on departmental employees’ use of social media provided that public comments that would “compromis[e] their ability to fulfil their duties professionally in an unbiased manner (particularly where comment is made about DIAC policy and programs)” or that are so “harsh or extreme in its criticism of the government, a member of parliament or other political party and their respective policies that questions [could arise about] the staff member’s ability to work professionally, efficiently or impartially” could serve as a basis for imposing discipline (up to can including discharge from the DIC.84 84. DIAC Social Media Guidelines, reprinted and quoted in Banerji and Comcare (Compensation), [2018] AATA 892, para. 36 (Apr. 16, 2018) (Austl.), available at https://www.asutax.asn.au/wp-content/uploads/2018/11/Banerji-and-Comcare-2018-AATA-892-16-April-2018.pdf (last visited Oct. 31, 2024). Similarly, more general APS guidelines emphasized that social media comments that are “so harsh and extreme in [their] criticism of the government or its policies that it indicates that the public servant concerned is incapable of professionally, efficiently or impartially performing his or her official duties, can and will serve as a basis for employee discipline, up to and including discharge.85 85. Australian Public Service Commission, Guidelines on Official Conduct of Commonwealth Public Servants 35 (Australian Gov’t Pub. Serv. 3rd ed., 1995), available at https://www.wipo.int/export/sites/www/tk/en/databases/creative_heritage/docs/pub_serv_comm_guidelines.pdf (last visited Oct. 24, 2024) [hereinafter APS Guidelines].

The HCA unanimously upheld the DIC’s decision to fire Banerji for her off-the-clock public policy critiques. Four members of the seven-member court, in their joint opinion, explain that “the impugned provisions . . . present as a plainly reasoned and focused response to the need to ensure that the requirement of upholding the APS Values and the integrity and good reputation of the APS trespasses no further upon the implied freedom than is reasonably justified.”86 86. Banerji, [2019] H.C.A. 23, at para. 42. The joint opinion emphasizes the importance of a politically neutral civil service to the operation of “responsible government,” meaning the staffing of senior executive branch posts by incumbent members of the federal parliament. The Justices posit that “[t]here can be no doubt that the maintenance and protection of an apolitical and professional public service is a significant purpose consistent with the system of representative and responsible government mandated by the Constitution.”87 87. Id. at para. 31.

It necessarily followed from this premise that the government possessed a legitimate purpose in banning civil service employees from engaging in speech activity that criticizes or might embarrass their government employer. To use the magic words, the APSA and APSC, and the implementing regulations, advanced “a legitimate government interest consistent with the system of representative and responsible government” because it was “suitable, necessary and adequate in the balance.”88 88. Id. at paras. 32-47. The joint opinion goes out of its way to emphasize the signal importance of government ministers enjoying loyal minions (like): “Regardless of the political complexion of the government of the day, or its policies, it is highly desirable if not essential to the proper functioning of the system of representative and responsible government that the government have confidence in the ability of the APS to provide high quality, impartial, professional advice, and that the APS will faithfully and professionally implement accepted government policy, irrespective of APS employees’ individual personal political beliefs and predilections.”89 89. Id. at para. 34.

Two glaring problems arise from the HCA’s approach. The first, and most obvious, is that government employee speech can be essential to the process of democratic deliberation. The HCA has taken pains, repeatedly, to emphasize that the IFPGC is not a “personal right” but rather a collective interest and a limitation on the scope of legislative power at the federal and state levels of government: “The freedom operates as a constitutional restriction on legislative power and should not be understood as a personal right.”90 90. LibertyWorks Inc, [2021] H.C.A. 18, at para. 44. It exists “as necessarily implied because the great underlying principle of the Constitution is that citizens are to share equally in political power and because it is only by a freedom to communicate on these matters that citizens may exercise a free and informed choice as electors.”91 91. Id.

Second, even if one takes these structural considerations into account, however, Banerji remains a deeply flawed, and highly problematic, decision.92 92. See Gray, supra note 32, at 31 (“The government’s clear difficulty with what Ms. Banerji did was that it was embarrassing to them. It was critical of their policies.”). Viewpoint discrimination is fundamentally inconsistent with any plausible general theory of freedom of expression because it represents an effort by the government to close off the marketplace of political ideas from ideas that the government dislikes or fears. This is not to say that the government could never justify speech regulations that feature viewpoint, or speaker-based, discrimination—but any and all such regulations should require the most compelling justifications to survive judicial review. If one accepts that the IFPGC does not give rise to a personal right, but instead supports a collective interest in a wide-open and robust process of democratic deliberation, no obvious reason exists for not conveying any meaningful protection on government employees who seek to speak out about matters of public concern—or to blow the whistle on wrongdoing within the agency or department for which they work. Under Banerji, the government has essentially no burden of justification to meet when it silences government workers (at least with respect to criticism, rather than praise, of the government and its policies).93 93. O’Brien, supra note 12, at 339 (noting that the applicable Australian civil service regulations provide that “employees should feel encouraged to make a ‘positive comment on social media about your agency’” but specifically provide that “criticism of the administration or work of your agency will ‘almost always. . .be seen as a breach of the Code’”) (quoting Australian Public Service Commission, Making Public Comment on Social Media: A Guide for APS Employees [hereinafter Making Comment on Social Media]).

Banerji makes Pickering, even as weakened by subsequent decisions including Connick v. Myers,94 94. 461 U.S. 138, 147–49 (1983). Garcetti v. Ceballos,95 95. 547 U.S. 410, 426 (2006). and City of San Diego v. Roe,96 96. 543 U.S. 77, 84 (2004). look incredibly robust. The HCA’s reasoning seems to be that if government employees were free to speak their minds freely about matters of public concern, ministers might not repose faith in them to work as honest brokers in the implementation of the government’s policies. Strangely, however, the APSA and APSC do not prohibit any and all public statements about matters of public concern by government employees covered by Australia’s civil service protections; they only prohibit critical comments.97 97. See Making Comment on Social Media, supra note 94; see also O’Brien, supra note 12, at 339, 350-51 (discussing how Australia’s civil service speech regulations and guidelines only prohibit criticism of the government but affirmatively encourage employees to make public statements of support).

Government employees are quite free to make positive public comments about the work of their departments.98 98. See APS Guidelines (prohibiting critical or embarrassing public comments but permitting civil service employees to praise or laud their government office and its leadership). The APS Guidelines are facially viewpoint-based—they only restrict negative or critical public commentary. See Making Comment on Social Media, supra note 94; see also O’Brien, supra note 12, at 339, 351 (noting that the Australian laws and regulations seek only to prohibit criticism of the government and its policies from government employees). But, even if one assumes all of the limitations, which the HCA repeatedly has emphasized, about the IFPGC protecting a collective or communal interest in speech related to democratic self-government, the Banerji opinion remains deeply problematic because it fails to consider, at all, the interest of the body politic in at least some kinds of government employee speech. It might well be that the process of democratic deliberation could go on and generate sensible electoral judgments without Banerji’s policy critiques of the DIC. However, if the Australian equivalent of a Daniel Ellsberg were to leak documents showing that the government had been lying to the voters for years about the efficacy of crucial government policies (such as a war in Southeast Asia), the collective interest in access to that information would be both profound and obvious.

The viewpoint-based nature of the APSA and APSC makes the Banerji court’s analysis deeply problematic—if not completely misguided.99 99. O’Brien, supra note 12, at 339 (“While neutrality is intended to mean restriction from both negative comment and positive comment, many of the prominent court cases involving restrictions on public employee’s political activity have been regarding criticism of the government.”); id. at 351 (observing that “much of the dispute over these regulations revolves around negative comments made about governmental agencies or their policies” and noting that “the language used in the guidance issued by the APS specifically references negative comments, as opposed to all political comments in general”). As Professor O’Brien observes, “Australia’s application and enforcement of the laws heavily weigh against any criticism of the government.”100 100. Id. Simply put, “the Australian system fails to provide protections for its public servants who wish to engage in political activity, providing that almost any criticism of the government will result in disciplinary action.”101 101. Id. at 355. Professor Anthony Gray, echoing Professor O’Brien’s concerns, objects that “the government’s legitimate interests in terms of an independent, impartial, and functional public service must be tempered by acknowledgment of the very valuable contributions that public servants can make to important public policy debates, from their specialised knowledge and expertise.”102 102. Gray, supra note 32, at 35–36.

Within the rubric of these limitations, it seems obvious that the implied freedom, even if it protects a collective interest in democratic deliberation rather than a personal autonomy interest, must afford at least some protection to government employees who speak out about matters of public concern103 103. O’Brien, supra note 12, at 351–52, 355 (arguing that Australia fails to provide any meaningful protection to government employees and maintains confusing and vague standards regulating government employee speech, leaving employees to guess about what could lead to discharge, because “[t]he Australian courts have also failed to provide protection or assistance to public employees in understanding these restrictions”).—both generally and more specifically when government employees blow the whistle on misconduct within the government agency for which they work.104 104. Gray, supra note 32, at 28 (criticizing the HCA’s Banerji decision for its “lack of explicit recognition in the judgment of the specific value of the speech of public servants in terms of democratic governance and holding governments to account” and complaining that “[a]ll of the judgments recognise the potential for the Code of Conduct provisions to impinge on the freedom of individuals to communicate about political matters” but “there is little express recognition of the particularly valuable role that public servants can play in this space, given their knowledge of the internal workings of government”). To sustain the government’s ban on speech critical of it or its policies, or that might embarrass a government minister, without even a nodding of the judicial head toward the voters’ interest in whistleblowing government employee speech, simply cannot be reconciled with the underlying theory of the IFPGC. Even if government employees do not have any individually protected interest in speaking, for example, as citizens, a serious question remains about whether some kinds of government employee speech might be essential to the use of elections to hold the government accountable (and therefore squarely within the core of the IFPGC).

Adopting Canada’s Holistic Balancing Approach Would Better Secure and Advance Core First Amendment Values

The United States would benefit from considering the examples that Canada and Australia provide regarding the critical role of courts in safeguarding the process of democratic deliberation that informs the act of voting. The SCC has embraced an approach that categorically rejects any bright-line rules that fence out particular government employee speech from constitutional protection. Any and all government employee speech enjoys protection under Section 2(b) of the Charter—including Keegstra’s anti-Semitic tirades and Fraser’s intemperate, and quite odd, attacks on metrification and the Charter—and this broad and inclusive approach does not impose unacceptably high costs on government employers, or on the public, because the government may always attempt to justify imposing discipline on an employee, thereby saving the government’s decision from invalidation, by showing a sufficiently important need to limit a public employee’s expressive activities.

Canada’s approach also takes into full account the nature of the government employee’s speech and puts a judicial thumb on the scale when a public servant blows the whistle on wrongdoing or misconduct within a government agency.105 105. Fraser, [1985] 2 S.C.R. 455, at para. 46 (opining that a government employee’s speech would be highly protected if it involved disclosure of “illegal acts, or if [the government’s] policies jeopardized the life, health or safety of the public servant or others”). The speech would also be protected “if the public servant’s criticism had no impact on his or her ability to perform effectively the duties of a public servant or on the public perception of that ability,” even if the public servant’s speech did not involve whistleblowing. See id. This is surely the correct approach—and one that should be incorporated into the Connick/Pickering/Garcetti First Amendment analysis.

If a nation’s domestic courts fail to provide meaningful protection to government employees who engage in whistleblowing speech, persons holding employment with the government will be far less willing to share critically important information about malfeasance and misconduct with the public. Quite clearly, democratic self-government will suffer as a consequence. After all, “[a] rational government employee will not disseminate information about wrongdoing within her department or agency if a not improbable consequence will be the loss of her employment.”106 106. Krotoszynski, Whistleblowing Speech and the First Amendment, supra note 2, at 274.

Since its landmark decision in Pickering,107 107. Pickering v. Bd. of Educ., 391 U.S. 563 (1968). the U.S. Supreme Court has pursued a halting and half-hearted First Amendment jurisprudence that excludes from any constitutional protection a great deal of government employee speech. Public employees who can show that their speech falls within a restrictively defined universe of speech related to a matter of public concern (Connick) and who can also demonstrate that their speech did not relate to their official government duties (Garcetti) will find themselves facing an uncertain outcome under Pickering’s open-ended balancing test. This test also fails to consider expressly the value of the government employee’s speech to the body politic. Simply put, the Pickering test as it presently exists is binary, but to take full account of the community’s interest in whistleblowing speech, it must include that collective interest as an express third factor.

What is more, under Connick/Pickering/Roe, government employers are entirely free, like private employers, to regulate their employees’ speech on or off the clock.108 108. Mary-Rose Papandrea, The Free Speech Rights of “Off-Duty” Government Employees, 2010 BYU L. Rev. 2117, 2145–58 (2010). Under current doctrine, perhaps best represented by Roe,109 109. Roe, 543 U.S. at 83–84. employers have broad authority to impose discipline on government employees for any and all expressive activity; the fact that speech takes place outside the workplace does not mean that it cannot serve as a basis for discipline (up to and including discharge from employment). As the Supreme Court has held in the context of off-campus speech by public school students,110 110. Mahanoy Area Sch. Dist. v. B. L, 141 S. Ct. 2038, 2053 (2021) (“A public school's regulation of off-premises student speech is a different matter. While the decision to enroll a student in a public school may be regarded as conferring the authority to regulate some off-premises speech (a subject I address below), enrollment cannot be treated as a complete transfer of parental authority over a student’s speech.”). Just as enrollment in a public school does not vest public school officials with carte blanche authority to regulate a student’s off-campus, after-hours speech on social media, government employment should not vest the state with comprehensive powers of censorship over public workers. the location of speech activity should affect the balancing exercise.

The Pickering/Connick/Garcetti framework adopts too many categorical rules that all favor, in nearly absolute terms, the government as manager over a public servant’s interests as a citizen and would-be speaker. The SCC’s approach, by way of contrast, never categorically excludes any speech from protection under Section 2(b) of the Charter and also expressly takes into account the electorate’s need for information that sometimes only government employees can provide. If both the government’s managerial interests and government employees’ autonomy interests are to be advanced concurrently, an open-ended balancing exercise—not a series of categorical rules that deny particular forms of government employee speech any First Amendment protection—will be essential to reconciling these important, competing values.

Indeed, perhaps the key difference between the jurisprudential approaches in Canada and the United States is that, in Canada, the government always labors under a duty to justify muzzling a civil servant, whereas, in the United States, categorical rules essentially zero out First Amendment protection for important forms of government employee speech. U.S. First Amendment doctrine currently relies on a series of categorical rules—again, literally all of which favor the government as manager over government employees as would-be speakers.

By way of contrast, for those who might think that the United States goes too far in affording government employees First Amendment protection, Australia provides a strong cautionary note running in the opposite direction. The HCA’s abject deference to the national parliament’s content- and viewpoint-based regulations of government employee speech is difficult, and arguably impossible, to reconcile with normative and policy-based justifications for recognition of the IFPGC in the first place. If a constitutional commitment to using voting and elections to select those who will make public policy necessarily implies limits on the government’s ability to censor political and governmental speech, the contributions of government employees to the process of democratic deliberation merit some consideration (and constitutional protection). Whistleblowing speech, in particular, would seem squarely related to the process of democratic deliberation and to empowering voters to use their ballots as a means of rendering the government accountable to the people.

What is more, Australia’s principal reliance on the political branches to safeguard fundamental rights makes the free and fair nature of the political process even more essential; if elections and voting are the safeguards of all substantive rights in Australia (rather than the courts enforcing substantive constitutional rights via judicial review), it is imperative that voters have access to any and all relevant information.111 111. See Meiklejohn, supra note 28, at 25–26, 88–89. Banerji, however, makes it far less likely that voters will possess critical information if government employees are the only potential source. Expecting public servants to embrace professional martyrdom is hopelessly unrealistic; if we want government employees to step forward when serious misconduct is taking place within a government agency, we must ensure that adequate constitutional protections exist for those employees who step forward and blow the whistle. Banerji sends the unfortunate message that government employees who speak out about matters of public concern must do so at their own personal, professional, and financial risk.

The Connick/Pickering doctrine plainly underprotects both government employees who speak out about matters of public concern (exercising their agency as citizens in a democracy) and government employees who engage in whistleblowing speech (thereby empowering other citizens to play their role in holding government accountable through the electoral process). Subsequent cases, notably including Garcetti and Waters v. Churchill,112 112. 511 U.S. 661, 675–80 (1994) (holding that a government employer may fire a public servant, without violating the First Amendment’s Free Speech Clause, based on a completely erroneous understanding of what an employee actually said provided that the government employer reasonably believed the incorrectly-attributed statements to be true). further undermine First Amendment protections for government employees. In cases like Connick, Garcetti, and Waters, the Supreme Court has expressed concern that the absence of categorical rules that limit the First Amendment exposure of government employers would unduly compromise the government’s ability to manage its workplaces.

To be sure, as Professor Post argues, a legitimate managerial domain exists, and government employers must be able to manage employees to achieve the institutional missions of government agencies and departments.113 113. Post, supra note 10, at 234–39, 247–57, 265–67. Even so, however, Canada’s example shows that categorical rules are not essential to safeguarding the legitimate managerial prerogatives of government employers. A more contextual, open-ended approach has not led Canadian courts to suffer from a deluge of government employment-related cases—or transformed the SCC into a kind of judicial Office of Personal Management or Civil Service Review Board. If the Canadian courts are capable of developing and deploying more open-ended, context-sensitive free speech rules, the federal courts in the United States should be capable of doing so as well.

Conclusion: Striking and Holding an Appropriate Constitutional Balance Between the Government as a Manager and the Constitutional Speech Rights of Government Employees Requires a Holistic and Context-Sensitive Approach—Which Canadian Free Speech Jurisprudence Accomplishes

Whistleblowing speech by government employees often provides the only means of providing We the People with the information essential to using ballots as a means of securing government accountability—and thus constitutes information essential to the process of democratic self-government. Accordingly, when fashioning doctrinal rules in this area, courts should consider not only the autonomy interests of the would-be speaker and the managerial needs of the government employer but also the value and importance of the information to the body politic. Unfortunately, courts both in the United States and Australia, at least to date, have failed to take into adequate account the voting public’s interest in hearing what government employees have to say about fraud, waste, abuse, mismanagement, and malfeasance within their government workplaces. If we require professional and financial martyrdom as the potential price for blowing the whistle on misconduct within a government agency, democratic deliberation—and ultimately democracy itself—will suffer greatly as a result.

Judges, across jurisdictions, have a pronounced tendency to defer reflexively to the government when it invokes managerial imperatives to justify keeping government employees tethered to a (very) short free speech leash. However, if a jurisdiction embraces the democratic self-government theory to justify the judicial protection of speech and other forms of expressive freedom (assembly, association, petition, and press), the courts must convey targeted protection on whistleblowing speech. This protection must also include ensuring that effective remedies exist, on a reliable basis, for government employees who blow the whistle and then suffer retaliation as a consequence—including reinstatement even if the whistleblower’s comments are highly unpopular with co-workers.114 114. See Krotoszynski & Roberts, supra note 33, at 964–65 (arguing that bona fide government employee whistleblowers “need affirmative reinstatement relief” that “ensure[s] the government restores both the tangible and intangible aspects of employment” but cautioning that ersatz or “pretextual whistleblowing” would provide a legitimate reason, under standard principles of equity, for a court to decline to reinstate a government employee who publicly disclosed government misconduct).

The SCC has developed a better approach to reconciling the legitimate managerial needs of the government with the constitutional speech rights of government employees. The U.S. Supreme Court could, if so inclined, take a page (or two) from the SCC and reform Pickering by making it more context-sensitive and less cabined by categorical rules that zero out completely government employees’ First Amendment rights. Democratic deliberation in the United States would benefit significantly from taking these jurisprudential steps.

Acknowledgments

The author wishes to thank, and acknowledge with gratitude, the Knight First Amendment Institute at Columbia University for inviting me to participate in its important symposium, Permission to Speak Freely? Managing Government Employee Speech in a Democracy, which considered in some detail various problems associated with protecting the speech rights of government employees. This essay reflects the benefit of the very helpful, constructive suggestions and ideas provided by the remarkably talented legal scholars who participated in this symposium. The University of Alabama Law School Foundation provided a generous summer research grant that supported my research and writing of this piece. Finally, the usual disclaimer applies: Any errors or omissions are the author’s sole responsibility.

© 2024, Ronald J. Krotoszynski, Jr.

Cite as: Ronald J. Krotoszynski, Jr., Reforming Pickering to Better Protect Government Employees’ Speech, 24-20 Knight First Amend. Inst. (Nov. 13, 2024), https://knightcolumbia.org/content/reforming-pickering-to-better-protect-government-employees-speech[https://perma.cc/97VE-5HMB].