Emory Law Journal

Campaign Speech Law with a Twist: When the Government Is the Speaker, Not the Regulator
Helen Norton Associate Dean for Academic Affairs and Associate Professor, University of Colorado Law School. Thanks for very thoughtful comments to Steven Andre, Rachel Arnow-Richman, Joseph Blocher, Richard Boldt, Alan Chen, Danielle Keats Citron, Caroline Mala Corbin, Roberto Corrada, Aya Gruber, Melissa Hart, Deborah Hellman, Jessie Hill, Leslie Gielow Jacobs, Sarah Krakoff, Scott Moss, Ofer Raban, Nantiya Ruan, Jana Singer, Catherine Smith, and the participants at University of Colorado and University of Maryland School of Law faculty workshops as well as at symposia at Case Western Reserve University School of Law, Loyola University of Chicago School of Law, and the University of Connecticut School of Law. Thanks also to Jordan Bunch, Evan Branigan, David Cline, Caitlin McHugh, Bridget Newman, and Chelsea Strautman Lassen for excellent research assistance, to Doris Cheung for endlessly patient technical assistance, to the Emory Law Journal staff for their outstanding editorial work, and to Dean David Getches and the University of Colorado Law School for financial support for this project.

Abstract

Although government entities frequently engage in issue-related campaign speech on a variety of contested ballot and legislative measures, this fact has been entirely overlooked in contemporary First Amendment debates over campaign speech law specifically and government speech more generally. The Supreme Court’s “campaign speech” and “government speech” dockets have focused to date on claims by private parties that the government has restricted or silenced their speech in violation of the First Amendment. In contrast, disputes over what this Article calls “governmental campaign speech” involve Free Speech Clause and other challenges by private parties who seek instead to silence the government’s speech on matters subject to vote by members of the public or their elected representatives.

This Article thus explores when, if ever, governmental campaign speech on contested ballot and legislative measures is sufficiently dangerous to justify a departure from the general rule that the government’s own speech is insulated from Free Speech Clause review. This inquiry invites important and challenging questions about both the nature of government and the nature of speech, valuably forcing us to think about how government does, and should, work—as well as how speech does, and should, work. To this end, this Article reexamines the constitutionality of governmental campaign speech by incorporating perspectives offered by the emerging—but so far entirely separate—constitutional debates over campaign finance reform and government speech.

This Article contends that government speech on issue campaigns generally furthers, rather than frustrates, key First Amendment interests. Transparently governmental campaign speech often provides great value to the public: it enhances political accountability by informing voters of their governments’ priorities and preferences, provides a valuable heuristic for those who do not or cannot evaluate the competing arguments for themselves, and adds to the marketplace of available ideas and arguments, especially (but not only) as a counter to expression from powerful, private sources.

The Article also identifies limits to its general proposition that the government’s campaign speech is constitutionally valuable. First, it emphasizes that the government should be permitted to assert the government speech defense to constitutional challenges to its campaign speech on contested ballot or legislative measures only when that speech is transparently governmental in origin—when the public can clearly identify the message’s governmental origins and thus hold the government politically accountable for its views. Second, it distinguishes government campaign speech that involves government’s endorsement of political candidates, concluding that governmental bodies’ campaign speech endorsing or opposing specific candidates raises greater threats to constitutional interests in preventing the self-perpetuation of incumbents and the entrenchment of political power. Finally, it highlights the availability of statutory and other nonconstitutional limits on government campaign speech, concluding that such constraints are constitutionally permissible yet often unwise as a policy matter in light of such expression’s great instrumental value to the public. It urges instead that policymakers carefully target such constraints to address specific instances of abusive government speech.

Introduction

Among the most contentious First Amendment controversies in recent decades have been those over federal, state, and local government efforts to regulate private parties’ campaign speech. This longstanding debate features vigorous disagreements over whether private entities’ campaign speech is sufficiently unique to justify, if not require, special First Amendment rules. 1 See Frederick Schauer & Richard H. Pildes, Electoral Exceptionalism and the First Amendment, 77 Tex. L. Rev. 1803 (1999) (exploring the possibility of First Amendment doctrine unique to expression concerning political campaigns and elections). Some argue that restrictions on private individuals’ or organizations’ campaign contributions or spending improve the quality of political speech by limiting its quantity, 2 See, e.g., Citizens United v. FEC, 130 S. Ct. 876, 903–04 (2010); Davis v. FEC, 554 U.S. 724, 741 (2008); McConnell v. FEC, 540 U.S. 93, 205 (2003), overruled in part by Citizens United, 130 S. Ct. 876; Austin v. Mich. State Chamber of Commerce, 494 U.S. 652, 658–59 (1990), overruled by Citizens United, 130 S. Ct. 876; Buckley v. Valeo, 424 U.S. 1, 48 (1976) (per curiam). while others maintain that such constraints on expression’s quantity undermine the First Amendment interests of both speakers and listeners. 3 See, e.g., Citizens United, 130 S. Ct. at 893; Davis, 554 U.S. at 736; McConnell, 540 U.S. at 134–40; Austin, 494 U.S. at 656; Buckley, 424 U.S. at 15.

Contemporary First Amendment controversies also include those over the Supreme Court’s “recently minted” government speech doctrine, 4 See Pleasant Grove City v. Summum, 129 S. Ct. 1125, 1139 (2009) (Stevens, J., concurring). which has focused to date on untangling competing claims by public and private entities to the same expression. 5For a more extensive discussion of the Supreme Court’s brief history with government speech, see Helen Norton & Danielle Keats Citron, Government Speech 2.0, 87 Denv. U. L. Rev. 899, 904–16 (2010). These disputes involve Free Speech Clause claims by private parties who argue that the government has impermissibly silenced, excluded, or punished their speech. The government, in turn, argues that it was instead speaking itself and thus entitled to prevent others from changing, joining, or otherwise garbling its own message. 6 See, e.g., Summum, 129 S. Ct. at 1129–30. Whether the contested speech is characterized as private or governmental in origin drives the applicable—and generally outcome-determinative—First Amendment analysis: although the Free Speech Clause constrains government’s power to regulate private speech, 7 E.g., Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 641 (1994) (recounting the First Amendment’s bar on government’s viewpoint-based discrimination against private speech); R.A.V. v. City of St. Paul, 505 U.S. 377, 382 (1992) (“The First Amendment generally prevents government from proscribing speech, or even expressive conduct, because of disapproval of the ideas expressed. Content-based regulations are presumptively invalid.” (citations omitted)); Police Dep’t of Chi. v. Mosley, 408 U.S. 92, 95 (1972) (“[A]bove all else, the First Amendment means that government has no power to restrict expression because of its message, its ideas, its subject matter, or its content.”). the Court has made clear that the government’s own speech is exempt from Free Speech Clause scrutiny. 8Under the Supreme Court’s government speech doctrine, the government’s own expression is insulated from Free Speech Clause challenges by plaintiffs who seek to alter or join that expression. See Summum, 129 S. Ct. at 1131 (“If [public entities] were engaging in their own expressive conduct, then the Free Speech Clause has no application. The Free Speech Clause restricts government regulation of private speech; it does not regulate government speech. A government entity has the right to ‘speak for itself.’ . . . Indeed, it is not easy to imagine how government could function if it lacked this freedom.” (citations omitted) (quoting Bd. of Regents v. Southworth, 529 U.S. 217, 229 (2000))); Johanns v. Livestock Mktg. Ass’n, 544 U.S. 550, 553 (2005) (explaining that the government’s own speech is “exempt” from Free Speech Clause scrutiny); see also Southworth, 529 U.S. at 235 (“When the government speaks, for instance to promote its own policies or to advance a particular idea, it is, in the end, accountable to the electorate and the political process for its advocacy. If the citizenry objects, newly elected officials later could espouse some different or contrary position.”).

Entirely overlooked in both sets of discussions, however, is the fact that the government frequently engages in issue-related campaign speech of its own on a variety of contested ballot and legislative measures. Indeed, such expression has generated considerable controversy at various times but has yet to be analyzed in light of contemporary First Amendment debates over the potential value and danger of private parties’ campaign speech (and governmental efforts to regulate it), as well as over the potential value and danger of government’s own speech more generally. The Court’s entirely separate “campaign speech” and “government speech” dockets have thus focused to date on claims by private parties that the government has impermissibly restricted or silenced their speech. In contrast, disputes over what I call “governmental campaign speech” involve Free Speech Clause and other challenges by private parties who seek instead to silence the government’s speech that expresses a position on matters subject to vote by members of the public or their elected representatives—pending ballot initiatives and referenda, 9A referendum generally enables citizens to enact or reject statutes or constitutional amendments proposed by a legislature, while an initiative enables citizens to draft proposals themselves and submit them directly to the populace for a vote. Ethan J. Leib, Can Direct Democracy Be Made Deliberative?, 54 Buff. L. Rev. 903, 904 (2006). as well as legislative measures. 10I acknowledge that many use the term “campaign speech” more narrowly to refer only to speech on matters subject to vote directly by the people themselvesballot measures and candidate elections. As discussed in more detail below, however, my definition of government speech on “issue campaigns” also includes the government’s speech on pending legislation subject to vote by the people’s representatives, because a number of critics urge that such efforts pose constitutional threats very similar to the threats they perceive posed by the government’s speech on ballot measures to be decided directly by the people. See infra notes 41–46, 60–67, and accompanying text.

For purposes of this Article, such “governmental campaign speech” means speech to the public (rather than to other government entities) that expresses the official view of a governmental branch or body, such as speech issued collectively in the form of a resolution or proclamation, or speech by an official empowered to speak for that governmental entity. It does not include speech by individual government officials expressing their own views in a personal, nongovernmental capacity—e.g., when a governor endorses a particular candidate for Senate on her own time and without any expenditure of government resources. Moreover, as discussed in greater detail in Part III, this Article distinguishes between governmental speech on issue campaigns—governmental speech that takes a position on a pending ballot or legislative measure—from governmental speech on candidate campaigns.

Governmental speech on issue campaigns takes a wide variety of forms and may be delivered by a broad range of government speakers. Examples include not only government officials’ statements and press releases critical or supportive of pending ballot or legislative measures but also government agencies’ reports and analyses, as well as flyers, pamphlets, newsletter articles, online postings, and print and broadcast advertisements communicating their views of such measures to the public.

Critics of such government expression (who include a variety of courts, policymakers, and commentators) argue that the government should refrain from taking sides in such contested policy contests. They maintain that the government’s campaign speech on ballot and legislative measures sufficiently differs from that of other participants in the political marketplace of ideas to require the government’s exclusion from that market, asserting more specifically that the government’s campaign speech undermines both speakers’ and listeners’ autonomy and equality interests. Controversies over such speech include those over the Eisenhower Administration’s advocacy on behalf of its proposed health care legislation, 11 See infra notes 78–79 and accompanying text. state human rights agencies’ communications in support of the Equal Rights Amendment, 12 See infra notes 33–34 and accompanying text. and local school boards’ expressive support for school bond measures. 13 See infra notes 24–28 and accompanying text. Such debates still rage today: the Fourth Circuit recently considered a First Amendment challenge to a public school board’s communications to potential voters in opposition to pending school voucher legislation, 14Page v. Lexington Cnty. Sch. Dist. One, 531 F.3d 275, 277–80, 287–88 (4th Cir. 2008). and a sharply divided Sixth Circuit recently rejected a similar challenge to a town’s use of public funds to express its position on ballot measures related to the local fire department’s financing and organization. 15Kidwell v. City of Union, 462 F.3d 620, 621–22, 626 (6th Cir. 2006).

This Article seeks to reexamine longstanding controversies over governmental campaign speech by drawing upon lessons from contemporary First Amendment debates over campaign finance reform and government speech more generally. In so doing, it explores when, if ever, governmental campaign speech is sufficiently dangerous to justify a departure from the general rule that the government’s own speech is insulated from Free Speech Clause review. 16I am indebted to Joseph Blocher for his observation that this is an analysis of the possibility of an exception to an exception: whether government campaign speech should be excepted (by subjecting it to Free Speech Clause scrutiny) from government speech doctrine more generally, which itself is an exception to traditional First Amendment doctrine (by exempting the government’s own expression from Free Speech Clause scrutiny).

To this end, Part I describes the spectrum of governmental campaign speech on pending ballot and legislative measures that critics have identified as potentially troubling. In so doing, it identifies two major strands of argument that appear in the various critiques of the government’s speech on issue campaigns. One strand involves a disagreement about the appropriate expressive role of government specifically, while the other involves a disagreement about the effects of campaign speech by powerful parties more generally. These disputes valuably force us to think about how government does, and should, work—as well as how speech does, and should, work.

Part II responds to those critiques. In so doing, it reviews and renews this debate through the lens of the Court’s emerging government speech doctrine, as well as its campaign finance reform jurisprudence. It concludes that transparently governmental campaign speech on certain contested political issues—campaign speech on pending ballot or legislative measures that is clearly identified to the public as governmental in source—furthers, rather than frustrates, key First Amendment values and thus should not trigger any unique constitutional suspicion. More specifically, such expression enhances political accountability by informing voters of their government’s priorities and preferences, provides a valuable heuristic for those who do not have the time or expertise to evaluate competing policy arguments for themselves, and adds to the marketplace of available ideas and arguments, especially (but not exclusively) as a counter to less accountable and nontransparent expression from powerful, private sources. Indeed, government expression’s value in this context may be particularly heightened in light of the current Supreme Court’s unwillingness in Citizens United v. FEC to permit limitations on campaign speech by corporate and other powerful, private interests. 17130 S. Ct. 876, 913 (2010).

Part III considers possible limits to this Article’s proposition that government speech on issue campaigns generally furthers key constitutional values and thus should not be understood to run afoul of the First Amendment. First, it emphasizes that government’s campaign speech on contested ballot or legislative measures should be considered consistent with the Free Speech Clause only when that speech is transparently governmental in origin—when the public can clearly identify the message’s governmental origins and thus hold the government politically accountable for its views. Second, it distinguishes government campaign speech regarding candidates, concluding that government entities’ speech endorsing or opposing specific candidates raises potentially greater threats to First Amendment interests in constraining the self-perpetuation of incumbents and the entrenchment of political power. Finally, it discusses the availability of statutory and other nonconstitutional limits on government speech on issue campaigns, concluding that those constraints are constitutionally permissible yet often unwise as a policy matter in light of government expression’s great instrumental value to the public. It urges instead that policymakers carefully target such constraints to address specific instances of abusive government speech.

I. The Perceived Constitutional Dangers of Governmental Campaign Speech

This Part briefly describes the concerns raised in longstanding debates over the dangers—or lack thereof—of government’s expressive participation in contested issue campaigns. 18As discussed in greater detail in Part III, this Article distinguishes government speech expressing a position on contested issue campaigns from government entities’ speech endorsing or opposing a specific candidate or party. Examples include those over state and local governments’ advocacy in support of or opposition to ballot measures to be voted on directly by the citizenry, as well as government entities’ advocacy in support of or opposition to measures to be voted on by a legislature.

A. Debates over Government Speech on Contested Ballot Measures

Federal and state courts have long displayed deep suspicion of the government’s issue-related campaign expression. As discussed in detail below, for years they routinely upheld plaintiffs’ challenges to the government’s political advocacy in support of or opposition to pending ballot measures. Moreover, several of the opinions in the Supreme Court’s most recent discussion of government speech assumed, without citation or explanation, the existence of limitations on the government’s power to engage in certain political or partisan advocacy. 19 See Pleasant Grove City v. Summum, 129 S. Ct. 1125, 1131–32 (2009) (“This does not mean that there are no restraints on government speech. . . . The involvement of public officials in advocacy may be limited by law, regulation, or practice.”); id. at 1139 (Stevens, J., concurring) (“Finally, recognizing permanent displays on public property as government speech will not give the government free license to communicate offensive or partisan messages. For even if the Free Speech Clause neither restricts nor protects government speech, government speakers are bound by the Constitution’s other proscriptions, including those supplied by the Establishment and Equal Protection Clauses. Together with the checks imposed by our democratic processes, these constitutional safeguards ensure that the effect of today’s decision will be limited.”).

Note, however, that these courts did not consistently identify a clear legal source of their discomfort with the government’s campaign speech. While many grounded their resistance in state and local government law limiting the powers of state or local governmental bodies, 20 See, e.g., Rees v. Carlisle, 153 P.3d 1131, 1137–39 (Haw. 2007) (concluding that a prosecutor’s use of public resources to urge voters to support an amendment to Hawaii’s constitution violates state law interpreted to permit a prosecutor to offer comments but not to engage in express advocacy); Citizens to Protect Pub. Funds v. Bd. of Educ., 98 A.2d 673, 676–77 (N.J. 1953) (determining that a school board’s use of public funds to print a booklet urging voters to “Vote Yes” on a bond referendum was not within the power implied to the board by its express power to operate the schools); Porter v. Tiffany, 502 P.2d 1385 (Or. Ct. App. 1972) (finding no statutory authorization for a city water and electric board’s expenditure of funds for materials advocating support of election measures regarding nuclear power program participation and delay of construction of a nuclear power plant). others pointed to the Free Speech Clause and the Guarantee Clause, as well as other unidentified constitutional sources. For example, the California Supreme Court has suggested that public agencies’ expenditures to advocate for or against pending ballot measures “raise potentially serious constitutional questions.” 21Stanson v. Mott, 551 P.2d 1, 9 (Cal. 1976) (in bank); accord Vargas v. City of Salinas, 205 P.3d 207, 226 (Cal. 2009) (discussing potential constitutional problems that may arise from the government’s use of public funds for speech in support of issue campaigns). More unequivocally, a Colorado federal court steeped its constitutional rejection of a school board’s power to expend funds to communicate with voters regarding a pending ballot measure in First Amendment rights to free speech and petition, 22 See Mountain States Legal Found. v. Denver Sch. Dist. # 1, 459 F. Supp. 357, 360 (D. Colo. 1978) (“[A] grant of express authority for a partisan use of public funds in an election of this type would violate the First Amendment to the United States Constitution . . . .”). as well as in Guarantee Clause concerns. 23 See id. at 361 (“When residents within a state seek to participate in this process by proposing an amendment to the state constitution, the expenditure of public funds in opposition to that effort violates a basic precept of this nation’s democratic process. Indeed, it would seem so contrary to the root philosophy of a republican form of government as might cause this Court to resort to [Article IV’s] guaranty clause . . . .”). Article IV, Section 4 of the Constitution—commonly known as the Guarantee Clause—provides that “[t]he United States shall guarantee to every State in this Union a Republican Form of Government.” U.S. Const. art. IV, § 4. Although the Supreme Court has long held that Guarantee Clause claims are not justiciable, e.g., Luther v. Borden, 48 U.S. (7 How.) 1, 42–43 (1849), a number of judges reviewing challenges to government campaign speech on contested ballot measures have suggested that this holding should be revisited, see, e.g., Kidwell v. City of Union, 462 F.3d 620, 635–36 & n.5 (6th Cir. 2006) (Martin, J., dissenting) (noting that, because “ordinary democratic controls are insufficient as a remedy in situations where governmental influence threatens to undermine the independent political process” and “[g]overnmental advocacy and campaign expenditures could arguably threaten to undermine free and fair elections, could be coercive, and could reasonably undermine the reliability and outcome of elections where the government acts as a participant,” “[p]erhaps it is time for the Supreme Court to reconsider its Guarantee Clause jurisprudence”).

Regardless of the specific constitutional source of their discomfort with governmental campaign speech, critics generally voice one or both of two objections. Some urge that the government should refrain from seeking to persuade the public on such matters because its governmental status means that its voice will inevitably coerce listeners’ beliefs. Others also (or instead) argue that the government’s voice threatens to drown out or otherwise unfairly disadvantage dissenting speakers. The following examples illustrate these courts’ concerns in more detail.

Among the earliest 24For other early examples of this trend, see Mines v. Del Valle, 257 P. 530, 536–37 (Cal. 1927) (in bank) (holding that the express municipal power to operate a public utility did not imply the power to appropriate funding to urge voters to approve a bond referendum to extend the utility), overruled by Stanson, 551 P.2d 1; and Elsenau v. City of Chicago, 165 N.E. 129, 131 (Ill. 1929) (invalidating as an unauthorized municipal function a city’s expenditures for advertisements that “did not purport to be an impartial statement of facts for the information of the voters, but that . . . was an attempt, partisan in its nature, to induce the voters to act favorably upon the bond issues submitted at the election”). Some early courts bucked this trend, emphasizing the instrumental value of the government’s expression on ballot campaigns. E.g., City Affairs Comm. v. Bd. of Comm’rs, 41 A.2d 798, 800 (N.J. Sup. Ct. 1945) (“We think municipalities may, within their discretion and in good faith, present their views for or against proposed legislation or referendum to the people of questions which in their judgment would adversely affect the interests of their residents.”), aff’d, 46 A.2d 425 (N.J. 1946). —and subsequently most influential—objections to the government’s campaign speech on contested ballot measures was that by Justice-to-be William Brennan when he served on the New Jersey Supreme Court. 25 Citizens to Protect Pub. Funds, 98 A.2d 673. A school board had appropriated a few hundred dollars in public funds for the printing and dissemination of an eighteen-page booklet that urged voters to support a bond referendum that would finance the expansion of several school buildings—an expansion, the board maintained, necessary to ensure adequate educational facilities for the town’s children. 26 Id. at 674. In dictum 27The court determined the issue to be moot because the election had already occurred. Id. at 676 (dictum) (“Plainly, then, any issues as to both the booklet and the radio broadcast are moot . . . . Nevertheless, the importance of the question makes appropriate our comment upon the actions taken.”). that proved persuasive to many later courts, Brennan characterized the government’s advocacy as fundamentally unfair to those with different views:

[T]he board made use of public funds to advocate one side only of the controversial question without affording the dissenters the opportunity by means of that financed medium to present their side, and thus imperilled the propriety of the entire expenditure. The public funds entrusted to the board belong equally to the proponents and opponents of the proposition, and the use of the funds to finance not the presentation of facts merely but also arguments to persuade the voters that only one side has merit, gives the dissenters just cause for complaint. The expenditure is then not within the implied power and is not lawful in the absence of express authority from the Legislature. . . .

. . . .

. . . We are persuaded, however, that simple fairness and justice to the rights of dissenters require that the use by public bodies of public funds for advocacy be restrained within those limits in the absence of a legislative grant in express terms of the broader power. 28 Id. at 677–78 (dictum).

Many other courts thereafter similarly characterized the government’s participation in such debates as fundamentally unfair to dissenters. 29 See, e.g., Palm Beach Cnty. v. Hudspeth, 540 So. 2d 147, 154 (Fla. Dist. Ct. App. 1989) (“If government, with its relatively vast financial resources, access to the media and technical know-how, undertakes a campaign to favor or oppose a measure placed on the ballot, then by so doing government undercuts the very fabric which the constitution weaves to prevent government from stifling the voice of the people.”); Anderson v. City of Boston, 380 N.E.2d 628, 639 (Mass. 1978) (“Fairness and the appearance of fairness are assured by a prohibition against using public tax revenues to advocate a position which certain taxpayers oppose.”); Stern v. Kramarsky, 375 N.Y.S.2d 235, 237 (Sup. Ct. Spec. Term 1975) (“It should be noted that by lending their support to the campaign underway for the passage of the Equal Rights Amendment, defendants not only provide certain promotional and advertising assistance, but they endow that campaign with all of the prestige and influence naturally arising from any endorsement of a governmental authority.”). A Florida state court offered a typical analysis when upholding a challenge to a county’s expenditure of funds to promote passage of a referendum that would have created a unified county health care district:

It is never in the public interest . . . to pick up the gauntlet and enter the fray. The funds collected from taxpayers theoretically belong to proponents and opponents of county action alike. To favor one side of any such issue by expending funds obtained from those who do not favor that issue turns government on its head and is the antithesis of the democratic process.

. . . The appropriate function of government in connection with an issue placed before the electorate is to enlighten, NOT to proselytize. 30 Hudspeth, 540 So. 2d at 154.

Most of these cases involved challenges to the government’s advocacy with respect to pending initiatives and referenda involving the financing of public schools 31 See, e.g., Campbell v. Joint Dist. 28-J, 704 F.2d 501 (10th Cir. 1983) (concluding that a school district’s expenditures urging the approval of a financing proposal were unauthorized under state statute); Choice-in-Educ. League v. L.A. Unified Sch. Dist., 21 Cal. Rptr. 2d 303 (Ct. App. 1993) (reversing the trial court’s grant of a preliminary injunction enjoining school board members from expressing opposition to a pending voucher proposal); Coffman v. Colo. Common Cause, 102 P.3d 999 (Colo. 2004) (en banc) (concluding that the secretary of state’s press releases opposing a statewide ballot initiative concerning school funding violated a state statute prohibiting public agencies from using more than fifty dollars in public monies to support or oppose ballot measures). and other government services. 32 See, e.g., Stanson v. Mott, 551 P.2d 1 (Cal. 1976) (in bank) (considering a challenge to a state agency’s expenditures to urge support for increased parks funding); Ameritel Inns, Inc. v. Greater Boise Auditorium Dist., 119 P.3d 624 (Idaho 2005) (concluding that a local government entity did not have statutory authorization to use public funds to support a ballot measure to approve the issuance of bonds to fund construction of the district’s convention center); Carter v. City of Las Cruces, 915 P.2d 336 (N.M. Ct. App. 1996) (considering a taxpayer’s challenge to city officials’ expenditures of public funds to support a ballot measure to empower the city to acquire a private electric utility). But government entities’ advocacy in support of ratification of the Equal Rights Amendment (ERA) triggered similar challenges and similar results. A New York state court, for example, expressed related objections when enjoining the state human rights agency from preparing flyers, pamphlets, and broadcast ads in support of a referendum to amend the state constitution to include the ERA. 33 Stern, 375 N.Y.S.2d at 239–40. For other examples of challenges to government speech that took sides on ballot measures outside of the government-financing context, see Rees v. Carlisle, 153 P.3d 1131, 1141 (Haw. 2007); and King County Council v. Public Disclosure Commission, 611 P.2d 1227 (Wash. 1980) (en banc). In so doing, the court suggested that such advocacy smacked of totalitarianism:

As a State agency supported by public funds they cannot advocate their favored position on any issue or for any candidates, as such. So long as they are an arm of the state government they must maintain a position of neutrality and impartiality.

It would be establishing a dangerous and untenable precedent to permit the government or any agency thereof, to use public funds to disseminate propaganda in favor of or against any issue or candidate. This may be done by totalitarian, dictatorial or autocratic governments but cannot be tolerated, directly or indirectly, in these democratic United States of America. This is true even if the position advocated is believed to be in the best interests of our country. 34 Stern, 375 N.Y.S.2d at 239.

To be sure, most of these challenges focused on government agencies’ expenditures of public funds for expressive purposes, arguing that such publicly financed government speech unfairly disadvantaged dissenters in light of the government’s potentially greater resources, prestige, and power. 35 See, e.g., Stanson, 551 P.2d at 9 (“A fundamental precept of this nation’s democratic electoral process is that the government may not ‘take sides’ in election contests or bestow an unfair advantage on one of several competing factions. A principal danger feared by our country’s founders lay in the possibility that the holders of governmental authority would use official power improperly to perpetuate themselves, or their allies, in office; the selective use of public funds in election campaigns, of course, raises the specter of just such an improper distortion of the democratic electoral process.” (citations omitted)). These concerns foreshadow and parallel (without ever acknowledging) similar concerns about the dangers posed by unfettered campaign spending by well-financed or otherwise-powerful, private parties.

But some critics object to any governmental expression in contested issue campaigns that deviates from neutrality, regardless of whether the government’s expression involves the use of public monies. Examples include challenges to school board members’ remarks in opposition to a pending school voucher initiative 36 Choice-in-Educ. League, 21 Cal. Rptr. 2d 303. Indeed, governments frequently express themselves on contested issues through resolutions and proclamations that require little, if anything, in the way of public expenditures. See Jacob E. Gersen & Eric A. Posner, Soft Law: Lessons from Congressional Practice, 61 Stan. L. Rev. 573, 584–86 (2008) (discussing congressional use of resolutions and other hortatory means to make declarations of policy). and county council members’ vote to endorse an antipornography measure. 37 King Cnty. Council, 611 P.2d at 1231. See generally Bruce W. Blakely, Comment, Public Utility Bill Inserts, Political Speech, and the First Amendment: A Constitutionally Mandated Right to Reply, 70 Calif. L. Rev. 1221, 1254 (1982) (“[I]t is the proper role of government to engage in informational speech. The state, however, may not employ a forum to engage in partisan expression. Government should not seek to influence initiatives or campaigns; nor are government editorials acceptable.”). Critics of such speech view the government’s sovereign role to require that it entirely refrain from taking sides in such debates because of the possibility that listeners might otherwise be particularly vulnerable to coercion. 38 See Stanson, 551 P.2d at 9–10. These critics object to the government’s campaign speech on any ballot measure that would be decided by the voters themselves, adopting the view that institutions of representative democracy should express no opinion on matters to be decided through direct democracy measures like ballot referenda or initiatives. 39 See id. (“[W]hile public agency ‘lobbying’ efforts undeniably involve the use of public funds to promote causes which some members of the public may not support, one of the primary functions of elected and appointed executive officials is, of course, to devise legislative proposals to attempt to implement the current administration’s policies. Since the legislative process contemplates that interested parties will attend legislative hearings to explain the potential benefits or detriments of proposed legislation, public agency lobbying, within the limits authorized by statute, in no way undermines or distorts the legislative process. By contrast, the use of the public treasury to mount an election campaign which attempts to influence the resolution of issues which our Constitution leave to the ‘free election’ of the people does present a serious threat to the integrity of the electoral process.” (citations omitted)); Note, The Constitutionality of Municipal Advocacy in Statewide Referendum Campaigns, 93 Harv. L. Rev. 535, 556–57 (1980) (“Cities are usually free to lobby before the legislature to prevent a measure from being put before the statewide electorate in referendum form. But once the state legislature decides to entrust the final legislative decision to the popular electorate, it explicitly removes the decision from the hands of state or municipal officeholders. Permitting those officials to use public funds to attempt to influence the outcome of that decision would partially return them to a role from which they have been excluded by constitutional design. Municipal governments should thus refrain from establishing an official political viewpoint during the time that the popular electorate, rather than its elected representatives, makes law.” (footnotes omitted)). Others modify this posture a bit, remaining sanguine about government entities’ advocacy in support of referenda they propose themselves, but objecting to such entities’ expressive opposition to ballot initiatives proposed by private citizens. 40 See, e.g., Ala. Libertarian Party v. City of Birmingham, 694 F. Supp. 814, 818 (N.D. Ala. 1988) (“While defendants might be forbidden to spend funds to support candidates, oppose initiative proposals, etc., they are not forbidden to publicize and seek public support for their own governmental proposals.”).

Some scholars have asserted parallel, and extremely broad, constitutional objections to the government’s departure from neutrality in contested policy debates. Robert Kamenshine, for example, proposed that courts should read the First Amendment to require “an implied prohibition against political establishment,” arguing that “participation by the government in the dissemination of political ideas poses a threat to open public debate that is distinct from government impairment of individual expression.” 41Robert D. Kamenshine, The First Amendment’s Implied Political Establishment Clause, 67 Calif. L. Rev. 1104, 1104 (1979). Professor Kamenshine downplayed the instrumental value—and highlighted the dangers—of the government’s policy advocacy in almost any context:

Some may argue that the government has a compelling interest in stimulating support for its programs. The argument is that in order to operate effectively, the government must be able to advocate the merits of its policies and programs. Institutional dissemination of this sort, however, conflicts with the goal of an open marketplace of ideas; it distorts public debate and thus raises serious establishment concerns. 42 Id. at 1116; accord id. at 1129, 1137 (“Under the prohibition of an implied political establishment clause, regulations that permit the dissemination of information to improve an agency’s image or to enhance support for its policies would be impermissible—both as to purpose and effect. . . . The political establishment clause would require balanced curricular and textbook treatment of the various viewpoints on race relations and women’s rights. It is possible that in some instances the government would design its textbooks and curriculum to present only a positive image of a particular group. Courts would have to prohibit such efforts, however, as a political establishment.” (footnote omitted)).

Edward Ziegler similarly viewed the government’s expressive participation in contested policy debates as so unavoidably coercive as to threaten democratic functions:

[A] characteristic distinguishing democratic from totalitarian government is that while a democracy attempts to facilitate and ascertain public opinion and establish policy in accordance therewith, an autocracy attempts to engineer public opinion in support of its decisions. . . . It is a truism that, if a governing structure based upon widespread genuine citizen opinions is to survive as a viable democracy, it must place legal restraints on the governments ability to manipulate the formulation and expression of that opinion.

. . . Although more subtle than censorship, official partisanship thorough [sic] the affirmative act of disseminating propaganda in support of a partisan viewpoint may pose as great or greater danger to political rights of free expression. 43Edward H. Ziegler, Jr., Government Speech and the Constitution: The Limits of Official Partisanship, 21 B.C. L. Rev. 578, 579–80 (1980) (footnote omitted). Professor Ziegler defined prohibited “official partisanship” for these purposes very broadly to include government speech on pending referenda, constitutional amendments, and legislative proposals. See id. at 581 n.12.

As an example of such objectionable partisanship, Professor Ziegler pointed to governmental efforts to urge support for ratification of the ERA. 44 Id. at 583.

In proposing to constrain government expression in an enormously wide range of settings, Professors Kamenshine and Ziegler staked out one of the far poles in the spectrum of views about the breadth of the government’s appropriate expressive function. 45Indeed, the views of Professors Kamenshine and Ziegler may reflect not only a more limited view of government’s appropriate expressive functions but also an assumption that the coercive effect of the government’s voice is especially great because people generally trust and defer to their government. That assumption may be less valid today in light of evidence that voters are significantly more suspicious of the government than they once were. See, e.g., Margaret Levi & Laura Stoker, Political Trust and Trustworthiness, 3 Ann. Rev. Pol. Sci. 475, 476–85 (2000) (surveying social science literature documenting declines in political trust). Indeed, much of this debate can be traced to longstanding, and continuing, battles over what it is that the government should do. As Steven D. Smith has observed:

What a government can properly say depends on the defined proper and essential role or function of the government. And issues of government speech are difficult—intractable, maybe—because there is no agreement about what governments purpose and function is or is not. In this respect, controversies about government speech are merely symptoms of a deeper disagreement about the proper domain and role of government. 46Steven D. Smith, Why Is Government Speech Problematic? The Unnecessary Problem, the Unnoticed Problem, and the Big Problem, 87 Denv. U. L. Rev. 945, 968 (2010).

By no means, however, have these debates been entirely one-sided, as other courts and commentators have demonstrated considerably greater comfort with the government’s campaign speech in at least some contexts. For example, Justice Brennan himself later embraced the value of the government’s campaign speech when he sat on the United States Supreme Court. 47City of Boston v. Anderson, 439 U.S. 1389 (Brennan, Circuit Justice 1978). There—years after repudiating the government’s persuasive efforts in support of contested ballot measures 48 See supra notes 25–28 and accompanying text. —he stayed a state court’s order that had enjoined the City of Boston from spending public funds to urge support of a ballot referendum on residential and commercial property tax rates. 49 Anderson, 439 U.S. 1389. This change of heart emerged shortly after the Supreme Court’s ruling in First National Bank of Boston v. Bellotti, which struck down, on First Amendment grounds, Massachusetts’s campaign finance law that had limited corporate campaign expenditures on pending ballot measures 50435 U.S. 765 (1978). —a ruling that newly empowered corporate participation in this debate and thus triggered Boston’s efforts to rebut such corporate speech. In determining that “the balance of the equities” justified a stay, 51 Anderson, 439 U.S. at 1390. Justice Brennan was among the first (and few) to see the link between government speech and campaign finance debates. He thus emphasized the value of the government’s voice in informing the voters on contested ballot measures—especially in countering powerful private speech:

In light of Bellotti, corporate industrial and commercial opponents of the referendum are free to finance their opposition. On the other hand, unless the stay is granted, the city is forever denied any opportunity to finance communication to the statewide electorate of its views in support of the referendum as required in the interests of all taxpayers, including residential property owners. 52 Id.

Still more controversies involving the ERA further illustrate competing views among the courts of the comparative costs and benefits of the government’s expressive participation in issue campaigns. Emphasizing the value as well as the inevitability of the government’s expressive participation in contested political debates, a California appellate court rejected a First Amendment challenge to a state women’s commission’s promotion of the ERA—and, indeed, rejected the possibility of the government’s expressive neutrality:

The root problem with plaintiffs free speech contention is that it proves too much. They offer no point of distinction between government speech addressing the status of women and government speech on any other topic; save that they deem the topic of womens status controversial. But, controversial or not, it is too late in the day to contend the economic and social status of women is not a legitimate topic of governmental concern. If the government, i.e., the Governor and legislative leaders, cannot appoint a commission to speak on the topic without implicating plaintiffs First Amendment rights it may not address any other controversial topics. If the government cannot address controversial topics it cannot govern. 53Miller v. Cal. Comm’n on the Status of Women, 198 Cal. Rptr. 877, 882 (Ct. App. 1984) (footnote omitted).

Other courts similarly upheld the expenditure of public funds for expressive purposes as central to the government’s role and found no constitutional barrier to the expression of views by institutions of representative democracy on matters to be decided by the voters themselves through direct democracy. 54 See, e.g., City Affairs Comm. v. Bd. of Comm’rs, 41 A.2d 798, 800 (N.J. Sup. Ct. 1945) (“We think municipalities may, within their discretion and in good faith, present their views for or against proposed legislation or referendum to the people of questions which in their judgment would adversely affect the interests of their residents. To accomplish this purpose we think they may incur expenditures by the publication of pamphlets, circulars, newspaper advertisements or radio addresses and that to do so is a proper governmental function.”), aff’d, 46 A.2d 245 (N.J. 1946). Indeed, illustrating a considerably more expansive view of the government’s appropriate functions, at least one court directly refuted a presumption of neutrality by characterizing the government as having a duty to share its views on such measures with the public: “The City and its officials not only have the right, but the duty, to determine the needs of its citizens and to provide funds to service those needs.” The same court expressed “a serious doubt as to whether governmental entities should be required to stand silently by while propositions which can impact on their tax structures, funds, services and programs are voted upon, even if initiated by others; unless participation is statutorily forbidden by an appropriate authority.” 55Ala. Libertarian Party v. City of Birmingham, 694 F. Supp. 814, 817, 819 n.10 (N.D. Ala. 1988); accord Steven Shiffrin, Government Speech, 27 UCLA L. Rev. 565, 613 (1980) (“[I]t is arguably the function, and perhaps the duty, of public officials to speak out on all issues of the day . . . . Governments, then, can justify subsidizing the speech of public officials, not to reelect them or others, but because there is a substantial interest in hearing what they have to say.”).

The majority opinion in Kidwell v. City of Union offers a recent example of this view—and is one of the first (and few) decisions to tie that view to the Supreme Court’s emerging government speech doctrine. 56462 F.3d 620 (6th Cir. 2006). There, a divided Sixth Circuit rejected a First Amendment challenge to a city’s expenditures of public funds to express its views on various initiatives related to the restructuring and funding of the town’s fire department: “To hold that [the city’s] advocacy converts its treasury to a public forum would severely limit the town’s ability to self-regulate and would be tantamount to a heckler’s veto, where the government could not speak for fear of opening its treasury to the public.” 57 Id. at 624. Demonstrating that the constitutional debate over the government’s campaign speech remains live, however, dissenting Judge Martin strongly objected to the government’s expressive participation in any ballot measure as unconstitutionally straying from a governmental duty of expressive neutrality: “I believe that the Constitution properly prohibits the government from having a horse in the race when it comes to elections. When government advocates on one side of an issue, the ultimate source of governing power is shifted away from the people and the threat of official doctrine exists.” 58 Id. at 635 (Martin, J., dissenting).

B. Debates over Government Speech on Contested Legislative Matters

As discussed above, most of the controversy to date has focused on challenges to public entities’ advocacy in support of or opposition to pending ballot measures to be decided by the voters themselves. But not all. Some critics of government speech also object to government entities’ speech to the public (as opposed to other government officials 59 See, e.g., Stanson v. Mott, 551 P.2d 1, 9 (Cal. 1976) (in bank) (noting that executive officials necessarily create legislative proposals to advance an administration’s policies). ) about issues to be decided by the legislature 60For a discussion of how voters considering ballot measures should be viewed as akin to legislators considering legislative measures, see Doe v. Reed, 130 S. Ct. 2811, 2833 (2010) (Scalia, J., concurring in the judgment). —again citing concerns about the appropriate role of government and about coercion and unfairness to those with different views. 61As explained supra note 10, this Article’s definition of “governmental campaign speech” includes government entities’ speech on legislative campaigns as well as ballot campaigns because some critics suggest that the government’s persuasive efforts with respect to matters subject to vote by legislators pose constitutional threats very similar to the threats they perceive posed by the government’s speech on matters subject to vote directly by the people themselves. The First Circuit, for example, expressed its 1973 rejection of the government’s power to take sides in a contested policy debate in constitutional terms, upholding a First Amendment challenge to the Boston School Board’s decision to send notices to all Boston parents urging them to join a march and rally in opposition to pending busing legislation. 62Bonner-Lyons v. Sch. Comm., 480 F.2d 442 (1st Cir. 1973). More recently, the Fourth Circuit considered a constitutional challenge to a local school board’s advocacy of a proposed school voucher bill pending before the state legislature. 63Page v. Lexington Cnty. Sch. Dist. One, 531 F.3d 275 (4th Cir. 2008). Among other things, the plaintiff there objected to the school board’s voter-directed speech on pending legislation, arguing for First Amendment limits on a government body’s advocacy to voters—rather than to legislators—on a matter to be decided by the state legislature. 64 Id. at 287. The Fourth Circuit rejected his constitutional claim. 65 Id. at 288.

Congress has similarly debated the propriety of executive branch officials’ and agencies’ advocacy to the public on matters to be decided by the national Legislature. Although those critics generally root the constitutional source of their discomfort in separation-of-powers terms rather than in the First Amendment, 66 See infra notes 80–82 and accompanying text. the debates in great part parallel those over the government’s speech on contested ballot measures. In both, controversy swirls not only over the appropriate role of government—and when, if ever, that role demands neutrality or silence—but also over whether and when government speech poses dangers of coercion and unfairness. 67This too is by no means a new debate over the scope of the government’s appropriate expressive role. Indeed, Jeffrey Tulis has exhaustively examined historical shifts in views about whether and when the President should speak directly to the public (rather than only to Congress) in support of or opposition to pending legislative matters. Jeffrey K. Tulis, The Rhetorical Presidency (1987). After documenting the rarity with which eighteenth- and nineteenth-century presidents spoke directly to the people on policy matters, id. at 61–87, Professor Tulis then describes the presidencies of Theodore Roosevelt and Woodrow Wilson as developing appeals to popular rhetoric as “a principal tool of presidential governance” for the first time, id. at 4. Of course, presidents now regularly appeal directly to the people at large, rather than to Congress, in soliciting support for legislative and other policy initiatives. I thank Jonathan Estin for directing me to this very helpful resource.

Indeed, congressional debates over efforts to regulate the government’s persuasive speech—termed by some as “propaganda”—in certain political contexts further illuminate the views of both critics and champions of such government campaign speech. Along these lines, David W. Guth has carefully documented longstanding controversies over the meaning and dangers of government “propaganda” that remain largely unresolved. 68 See David W. Guth, Black, White, and Shades of Gray: The Sixty-Year Debate over Propaganda Versus Public Diplomacy, 14 J. Promotion Mgmt. 309 (2008). One often-articulated, but very broad, view characterizes government “propaganda” as any effort by the government to persuade its public listeners. 69 See id. at 310 (noting that many use the term propaganda “as an umbrella covering all forms of persuasive communication, including advertising and public relations”). Although my focus here is on controversies over the government’s domestic “propaganda,” such definitional challenges play out in the international arena as well. See, e.g., Richard B. Collins, Propaganda for War and Transparency, 87 Denv. U. L. Rev. 819, 821–28 (2010) (discussing varying definitions of “war propaganda” prohibited by international law and prosecuted in the Tokyo and Nuremberg war crimes trials). Others instead characterize only the government’s covert, 70 See Guth, supra note 68, at 310 (noting that some distinguish between “revealed propaganda, messages that are overt in their effort to persuade, such as those in conventional advertising,” and “concealed propaganda, such as publicity generated from the distribution of news releases”). misleading, 71 E.g., Kelly Sarabyn, Prescribing Orthodoxy, 8 Cardozo Pub. L. Pol’y & Ethics J. 367, 369 (2010). or monopolistic 72 See Frederick Schauer, Is Government Speech a Problem?, 35 Stan. L. Rev. 373, 380 (1983) (reviewing Mark G. Yudof, When Government Speaks: Politics, Law, and Government Expression in America (1983)) (“Now it is true that propaganda is hardly a term of precise definition, but we usually use it when talking about totalitarian regimes whose propaganda is coupled with a rigid, if not absolute, suppression of any voice other than that of the state. The issues raised when the state is the only voice are significantly different from those raised when the state is one of many voices, especially when the other voices are unrestrained both in law and in fact.”). persuasive efforts as “propaganda.” More specifically, Garth Jowett and Victoria O’Donnell describe distinctions between what some call “white propaganda” (transparent and accurate persuasive communications by government); “black propaganda” (deceptive and nontransparent governmental efforts to persuade); and “gray propaganda” (where the source and accuracy of the communications remain unclear). 73 Garth S. Jowett & Victoria O’Donnell, Propaganda & Persuasion 17–20 (5th ed. 2012). Objections to government’s “propaganda,” however defined, may turn on the government’s target audience. For example, Guth describes a 1948 statute that “authorized the federal government to engage in a program of dissemination of truthful information to international audiences, while . . . prohibit[ing] the government from transmitting the same information to domestic publics.” Guth, supra note 68, at 312. Once again, these competing definitions of “propaganda” reveal differing levels of discomfort with government’s expressive participation in certain political contests.

In response to various concerns about the potential dangers of such expression, Congress has regulated government “propaganda” on a number of occasions. 74In 1942, for instance, Congress amended the Foreign Agents Registration Act to require agents of foreign principals who distribute “political propaganda” to ensure the material’s conspicuous labeling as such. Pub. L. No. 77-532, sec. 1, § 4(b), 56 Stat. 248, 255 (1942) (amended 1995). In enacting this requirement, Congress was spurred by concerns about the circulation of anonymous anti-American propaganda from Nazi sources in the years before and during World War II. See H.R. Rep. No. 75-1381, at 2 (1937) (“This required registration will publicize the nature of subversive or other similar activities of such foreign propagandists, so that the American people may know those who are engaged in this country by foreign agencies to spread doctrines alien to our democratic form of government, or propaganda for the purpose of influencing American public opinion on a political question. . . . We believe that the spotlight of pitiless publicity will serve as a deterrent to the spread of pernicious propaganda. We feel that our people are entitled to know the sources of any such efforts . . . .”). Decades later, a plaintiff who sought to exhibit Canadian films on acid rain and nuclear war challenged the statute on First Amendment grounds, arguing that the Act’s labeling requirements with respect to “propaganda” included a pejorative connotation that deterred him from exhibiting the films under such terms. Meese v. Keene, 481 U.S. 465, 467–68 (1987). After discussing the wide variety of available definitions of the term “propaganda,” id. at 477–78, the Supreme Court ultimately—and controversially—concluded that the labeling requirements did not burden speech (and thus posed no First Amendment problem) because Congress intended no pejorative connotation when using the term in the Act, id. at 484–85. Congress amended the statute in 1995 to replace the term “political propaganda” with “informational materials.” Lobbying Disclosure Act of 1995, Pub. L. No. 104-65, sec. 9, § 614(a)–(c), 109 Stat. 691, 700. For example, since 1951 each Congress has enacted an appropriations rider that entirely bars federal agencies from unauthorized expenditures to engage in “publicity or propaganda.” 75Jodie Morse, Note, Managing the News: The History and Constitutionality of the Government Spin Machine, 81 N.Y.U. L. Rev. 843, 853 (2006) (quoting 97 Cong. Rec. 4098 (1951)). Never, however, has Congress defined this statutory term. 76 See, e.g., Consolidated Appropriations Act, 2004, Pub. L. No. 108-199, 118 Stat. 3.

This proposal was initially spurred by members of Congress unhappy 77Powerful, private interests like the American Medical Association were unhappy as well. See Harry M. Marks, Revisiting “The Origins of Compulsory Drug Prescriptions,85 Am. J. Pub. Health 109, 112–13 (1995) (describing the American Medical Association’s hostility toward Oscar Ewing and his efforts). with the Eisenhower Administration’s efforts to generate public support for its proposed health care legislation. 78 See, e.g., Oscar R. Ewing, The Nation’s Health: A Report to the President 12–13 (1948) (“In the face of these and the other proofs given in this report that further rapid improvement in national health can be achieved only by concerted effort, and that the need for increased Federal action is imperative, I have reached the considered conclusion that more extensive and efficient Nation-wide planning is the only effective way to accomplish a significant betterment in national health.”); id. at 18 (“I am compelled to urge, as strongly as I know how, that the Congress enact, as President Truman has recommended, a system of Government prepayment health insurance in the terms in which it has been mapped out in [the report].”). In particular, the prohibition’s sponsors objected strenuously to, and characterized as potentially totalitarian, public speeches and other materials in support of the Administration’s bill by Oscar Ewing, the head of the U.S. Federal Security Administration (the federal agency that served as the precursor to today’s Department of Health and Human Services). 79 See Legislative Activities of Executive Agencies: Hearings Before the H. Select Comm. on Lobbying Activities, 81st Cong. 338–39 (1950) (statements of Charles A. Halleck & Clarence J. Brown, Members, H. Select Comm. on Lobbying Activities) (disparaging Ewing’s efforts in “sell[ing] that particular piece of legislation” as “mak[ing] speeches and spread[ing] the philosophy of socialism and Government dictatorship”). Ewing’s advocacy triggered a debate over the value and danger of executive branch speech that remains unresolved today.

More specifically, congressional supporters of the propaganda ban objected to executive branch officials’ advocacy to the public on an issue pending before the legislative branch as not only potentially coercive of public opinion but also offensive to the appropriate separation of powers. 80 See id. at 341 (statement of Clarence J. Brown, Member, H. Select Comm. on Lobbying Activities) (“[Y]ou have certain administrative responsibilities, but the Constitution does not give you any authority to attempt to make law. It restricts that authority to no one else in the world but to the Congress of the United States. The President does not have any authority to make law, nor should he have. He sometimes tries to exercise it, however.”). The original congressional debate over the proposed rider thus centered on whether a prohibition on undefined propaganda would interfere with the government’s responsibility to inform the public about its programs, 81 See 97 Cong. Rec. 4098 (1951) (statement of Rep. Sidney R. Yates) (“Would not the effect of the gentleman’s amendment in using the word ‘propaganda’ jeopardize publication by the Children’s Bureau of pamphlets pertaining to the training and growth of children?”); id. at 4099 (statement of Rep. John E. Fogarty) (“We do not even know what the gentleman calls propaganda. We do not know what he calls the right type of publicity or the wrong type of publicity. That is the fault I find with this amendment.”); id. at 4100 (statement of Rep. John E. Fogarty) (“Here you are limiting the amount of publicity and propaganda which may be issued by any agency of government in this bill and yet you do not define in the amendment what propaganda is or what publicity is.”). or instead would simply prevent the government from adopting what some characterized as coercive governance tactics. 82 See id. at 4099 (statement of Rep. George Meader) (“[The propaganda prohibition] is necessary to strengthen the Congress in the interest of formulating national policy by the people themselves. It is a corollary to that principle that public opinion ought not to be subjected to influence and direction by the executive agencies, the administrative branch of the Government, in the manner that it is today. In a democracy, where public opinion rules in the long run, the media of communication: the press, the radio, television, and the printed word, are very potent weapons in the control of the affairs of this country.”). The latter view prevailed, and the propaganda ban passed without any definition of the prohibited speech. The ban has remained in place as an appropriations rider for the many decades since its initial adoption.

Despite this longstanding ban, executive branch agencies and officials, of course, regularly continue to speak to the public about a wide variety of matters, legislative and otherwise. 83For discussion of the inevitability of executive speech on such matters, see, e.g., Leslie Gielow Jacobs, Bush, Obama and Beyond: Observations on the Prospect of Fact Checking Executive Department Threat Claims Before the Use of Force, 26 Const. Comment. 433 (2010); and Vasan Kesavan & J. Gregory Sidak, The Legislator-in-Chief, 44 Wm. & Mary L. Rev. 1 (2002). Indeed, the office charged with monitoring the ban’s enforcement—the Government Accountability Office (GAO)—has defined the ban narrowly and identified violations very rarely. 84As the agency charged generally with monitoring public expenditures, the GAO is the administrative body nominally responsible for attending to the “propaganda rider.” Morse, supra note 75, at 859; accord 1 Office of the Gen. Counsel, U.S. Gen. Accounting Office, Principles of Federal Appropriations Law 4-198 (3d ed. 2004). The GAO, however, has only a purely advisory function and no direct enforcement power. Morse, supra note 75, at 859. At most, the GAO makes findings and can refer its determinations to Congress or other governmental bodies for further investigation. Id. In so doing, the GAO has emphasized the value of executive branch expression:

Our decisions reflect societal values in favor of a robust exchange of information between the government and the public it serves. This includes the right to disseminate information in defense of an administrations point of view on policy matters.

Accordingly, as part of our efforts to strike the right balance, we have historically afforded agencies wide discretion in their informational activities. 85 U.S. Gen. Accounting Office, B-302504, Medicare Prescription Drug, Improvement, and Modernization Act of 2003—Use of Appropriated Funds for Flyer and Print and Television Advertisements 7 (2004) (citations omitted). Along these lines, the GAO has declined to characterize the government’s expressive activities as prohibited “propaganda” as long as they “are reasonably related to the agency’s duty to inform the public of agency actions, programs, and policies, or justify and rebut attacks upon its policies.” U.S. Gov’t Accountability Office, B-316443, Department of Defense—Retired Military Officers as Media Analysts 8 (2009); accord U.S. Gen. Accounting Office, supra, at 13 (“To restrict all materials that have some political content or express support for an Administration’s policies would significantly curtail the recognized and legitimate exercise of the Administration’s authority to inform the public of its policies, to justify its policies and to rebut attacks on its policies. It is important for the public to understand the philosophical underpinnings of the policies advanced by elected officials and their staff in order for the public to evaluate and form opinions on those policies.”).

More specifically, absent any statutory definition of prohibited government propaganda, the GAO has interpreted the ban to prohibit only a federal agency’s “self-aggrandizement” or “puffery” (i.e., materials that “emphasize the importance of the agency or one of its officials”); federal agency activities that are “purely partisan in nature” (i.e., that are “designed to aid a political candidate or party”); and “covert propaganda” (i.e., materials that do not disclose their governmental source). 86 U.S. Gov’t Accountability Office, B-319075, Department of Health And Human Services—Use of Appropriated Funds for “HealthReform.gov” Web Site and “State Your Support” Web Page 7–8 (2010). Emphasizing government expression’s great instrumental value to the public, moreover, the GAO has construed even those categories narrowly 87 See U.S. Gov’t Accountability Office, supra note 85, at 2 (“Application of the prohibition is necessarily balanced against an agency’s responsibility to inform the public about its activities and programs, explain its policies and priorities, and defend its policies, priorities, and point of view.”). and has been further reluctant to find violations other than in cases of covert government promotion of its programs or policies. 88 See, e.g., U.S. Gov’t Accountability Office, B-305368, Department of Education—Contract to Obtain Services of Armstrong Williams 14 (2005) (“Every agency has a legitimate interest in the ‘dissemination to the general public . . . of information reasonably necessary to the proper administration of the laws’ for which the agency is responsible. However, while we agree that the Department should disseminate information to the public on the NCLB Act, it must disclose its role.” (citations omitted) (quoting U.S. Gen. Accounting Office, B-106139, Appropriations—Limitations—Publicity and Propaganda Prohibition—Labor–Federal Security Appropriation Act, 1952 (1952))).

Recent controversies, however, have renewed attention to the propaganda ban, its definitional deficiencies, and its lack of enforcement. Examples include the Department of Education’s contract with newspaper columnists to produce op-eds supporting the Bush Administration’s “No Child Left Behind” initiative without disclosing the Department’s sponsorship, 89 See id. at 1. as well as the Administration’s briefing of and other close involvement with retired military personnel who then appeared on television as private military analysts offering their view of the war in Iraq and Afghanistan. 90David Barstow, Behind TV Analysts, Pentagon’s Hidden Hand, N.Y. Times (Apr. 20, 2008), http://www.nytimes.com/2008/04/20/us/20generals.html?_r=3&scp=1&sq=hand+of+pentagon&st=nyt&oref=slogin&oref=slogin. The GAO’s 2009 report found that the program did not violate the propaganda ban. See U.S. Gov’t Accountability Office, supra note 85, at 2 (“Clearly, DOD attempted to favorably influence public opinion with respect to the Administration’s war policies in Iraq and Afghanistan through the RMOs [Retired Military Officers]. However, as we discuss below, and based on the record before us in this case, we conclude that DOD’s public affairs outreach program to RMOs did not violate the prohibition. We found no evidence that DOD attempted to conceal from the public its outreach to RMOs or its role in providing RMOs with information, materials, access to department officials, travel, and luncheons. Moreover, we found no evidence that DOD contracted with or paid RMOs for positive commentary or analysis.”). These developments triggered new (but so far unsuccessful) congressional efforts not only to define prohibited government “propaganda” for the first time but also to define it very broadly. 91For example, legislation introduced in the Senate in 2005 would have prohibited, inter alia, any executive branch expression “designed to support or defeat legislation pending before Congress or any State legislature” other than testimony at legislative hearings. S. 266, 109th Cong. §§ 3–4 (2005). Indeed, although the public controversies themselves focused on allegedly covert governmental speech—persuasive speech sponsored by the government that did not disclose its governmental origins—congressional critics responded with the introduction of bills that proposed much more sweeping restrictions on executive branch speech on legislative matters. 92 See, e.g., S. 266.

This Part has documented longstanding debates over the comparative value and dangers of the government’s expressive participation in contested policy debates. The next Part joins this debate.

II. Evaluating the Potential Constitutional Dangers of Governmental Campaign Speech on Contested Policy Issues

This Part identifies and responds to two major strands of argument that appear in the various critiques of the government’s campaign speech on contested ballot and legislative measures. Critics offer both sets of arguments to support the claim that constitutional values are best protected by silencing the government’s speech in such contexts. 93Such a claim—that some speech should be restricted to facilitate free speech values—may seem novel from a First Amendment standpoint but, of course, is not without some precedent. For example, the debate over the First Amendment implications of campaign finance regulation in part turns on a debate over whether the First Amendment protects speech (in which case, government restrictions on corporations’ campaign speech are suspect) or certain speakers (in which case, corporate campaign speech might be permissibly regulated to prevent drowning out the speech of protected speakers). See Kathleen M. Sullivan, Two Concepts of Freedom of Speech, 124 Harv. L. Rev. 143, 155, 158 (2010) (discussing implications of the debate over whether the First Amendment protects speech or speakers). I suggest that one strand involves a disagreement about the appropriate expressive role of the government specifically, while the other involves a disagreement about the effects of campaign speech by powerful parties more generally. These disputes valuably force us to think about how the government does, and should, work—as well as how speech does, and should, work.

More specifically, the first critique focuses on the government as speaker, objecting to governmental deviations from neutrality or silence on certain matters. My response here draws in great part from debates over government speech more generally (i.e., government speech outside of the campaign context).

The second critique focuses instead on the effects of campaign speech by especially powerful speakers, governmental or otherwise. My discussion here draws in great part from ongoing debates over the effects of speech by private parties in the campaign context and whether the dangers of such speech or of its regulation pose the greater threat to First Amendment values. More specifically, Kathleen Sullivan has helpfully described the debate over the constitutionality of campaign finance reform as a clash between competing views of the First Amendment. 94Sullivan, supra note 93. One envisions free speech as “serving the interest of political liberty,” treating “with skepticism all government efforts at speech suppression that might skew the private ordering of ideas. And on this view, members of the public are trusted to make their own individual evaluations of speech, and government is forbidden to intervene for paternalistic or redistributive reasons.” 95 Id. at 145. The other approach to the First Amendment envisions free speech rights as “serv[ing] an overarching interest in political equality,” which primarily embraces an antidiscrimination principle that protects dissenters and other marginal speakers from disadvantage. 96 Id. at 144. Under this view, “politically disadvantaged speech prevails over regulation but regulation promoting political equality prevails over speech.” 97 Id. at 145.

With this as background, we can understand critics of government campaign speech as variously arguing that such expression offends both liberty and equality values implicated by the First Amendment’s Free Speech Clause. First, critics share a concern that the government’s campaign advocacy threatens a liberty-based conception of the First Amendment because they perceive it as inherently coercive of listeners’ beliefs due to the government’s sovereign status. 98 See supra notes 33–34 and accompanying text. Second, even if government persuasion falls short of actually coercing listeners, critics fear that the government’s voice remains sufficiently powerful to threaten First Amendment equality interests by drowning out or otherwise unfairly disadvantaging dissenting speakers. 99 See supra notes 25–29 and accompanying text.

After considering each of these critiques, this Part ultimately rejects both, concluding that the government’s expressive participation in contested issue campaigns generally furthers, rather than frustrates, key First Amendment values. 100Note that this Article does not maintain that the government has a First Amendment right to speak, but instead that government speech does not violate the Free Speech Clause because its expression furthers, rather than frustrates, key First Amendment values. As discussed in more detail below, that the government is not a First Amendment rights holder itself leaves open the possibility that its speech could be constrained through statute or other nonconstitutional means. See infra notes 209–12 and accompanying text.

A. Responding to Critiques of the Government as Speaker

A number of critics object to any persuasive efforts by government speakers on matters subject to vote by members of the public; some critics also object to government speech to the public on matters subject to vote by the people’s elected representatives. These objections are rooted in a specific theory of government and its appropriate role, rather than (as discussed later 101 See infra notes 139–41 and accompanying text. ) in a theory that speech by powerful parties more generally inflicts harm in the campaign context.

1. Government Persuasion Can Be Parsed from Government Coercion

Central to this critique is the notion that the government is unique among all speakers because of its potentially coercive power as sovereign—its power to apply “sufficient pressure . . . to compel a certain course of action.” 102Ekow N. Yankah, The Force of Law: The Role of Coercion in Legal Norms, 42 U. Rich. L. Rev. 1195, 1216 (2008); accord id. at 1199 (“Asserting that the law is coercive isolates a particular way it treats people; namely, the law imposes a non-voluntary normative system on people.”); id. at 1218 (“Coercion is normally claimed when one has been forced by another to act, or refrain from acting, against their will. Coercive pressure can overcome one’s will and make a particular course of action unreasonably costly. For example, where coercive pressure is applied to Bob, that pressure would render one or more of his options unreasonably costly. Coercive pressure in this respect makes a particular option unreasonable but not necessarily impossible.” (footnote omitted)). Recall that the values at the heart of the Free Speech Clause include an interest in protecting individual liberty from government coercion. 103The First Amendment’s primary purposes are most often identified to include protecting individual interests in autonomy and self-expression, furthering citizen participation in democratic self-government, and contributing to the discovery of truth and the development of knowledge. See, e.g., Ronald A. Cass, The Perils of Positive Thinking: Constitutional Interpretation and Negative First Amendment Theory, 34 UCLA L. Rev. 1405, 1411 (1987); Thomas I. Emerson, First Amendment Doctrine and the Burger Court, 68 Calif. L. Rev. 422, 423 (1980). Liberty-based arguments usually focus on whether the government is undermining individual autonomy and self-expression by punishing private speakers for their views. 104 See supra note 7. Such liberty-based arguments take different form in the context of governmental campaign speech, however, where critics seek to silence the government for fear that otherwise the government’s expressive efforts will coerce listeners’ beliefs and expression. In other words, as described above, some see the government’s persuasive voice on campaign matters as inherently and inevitably coercive—potentially even posing threats of totalitarianism. 105 See supra notes 30–34 and accompanying text; see also Brian C. Castello, Note, The Voice of Government as an Abridgement of First Amendment Rights of Speakers: Rethinking Meese v. Keene, 1989 Duke L.J. 654, 676 (“When the public perceives the government as having expert knowledge of an issue or unrivaled control over a matter, the government communication may command even greater respect and faith. Respect for the government, reliance on the accuracy of its statements, and fear of challenging its authoritative word might cause citizens to withdraw, to a certain extent, from their duties of self-governance. As a result, government would coerce, rather than earn, the majority’s consent.” (footnotes omitted)).

To be sure, governmental efforts at thought control would be anathema to the First Amendment. 106 See Martin H. Redish & Kevin Finnerty, What Did You Learn in School Today? Free Speech, Values Inculcation, and the Democratic–Educational Paradox, 88 Cornell L. Rev. 62, 67 (2002) (“[T]he notion of thought control is inconsistent with the concepts of free thought and mental autonomy that render the exercise of the free expression right meaningful.”); id. at 75 (“[A] foundational strategy of any sophisticated totalitarian society is to control the minds of its citizens, thereby destroying individual mental autonomy.”); Sarabyn, supra note 71, at 369 (“[Prescribing orthodoxy] fails to treat citizens as free and autonomous, and essentially appoints the state itself as the supreme judge of ideological truth, reproducing its ideological views in the citizenry by deploying coercion.”). But just as not all exercises of a sovereign’s power are coercive, the government’s persuasive expression by itself is not inevitably manipulative. 107 See Steven Lukes, Power: A Radical View 21 (2d ed. 2005) (describing power as including “coercion, influence, authority, force and manipulation,” and distinguishing “coercion,” which means that “A secures B’s compliance by the threat of deprivation where there is ‘a conflict over values or course of action between A and B’” from “influence,” which “exists where A, ‘without resorting to either a tacit or an overt threat of severe deprivation, causes [B] to change his course of action’” (alteration in original) (quoting Peter Bachrach & Morton S. Baratz, Power and Poverty: Theory and Practice 24, 30 (1970))); see also Schauer, supra note 72, at 384–85 (“If . . . we view the first amendment as primarily or exclusively protecting individual self-expression, self-realization, self-fulfillment, or something of that sort, then it is hard to see how government speech could be a first amendment problem. When we focus on communicators or listeners not instrumentally but as the end result of our concern, the fact that the government may be speaking is of little moment.” (footnote omitted)). In other words, as Abner Greene has explained, that the government represents—and, in fact, is—authority does not mean that its persuasive efforts are necessarily coercive. 108 See Abner S. Greene, Government of the Good, 53 Vand. L. Rev. 1, 45 (2000) (“No choice is made exogenously to the various authority structures in which one lives (e.g., government–citizen, parent–child, spouse–spouse, friend–friend, human being–God, etc.). We make choices under the influence of many whom we consider authorities . . . . [W]e must accept choice in a web of power relations. We must deem nonautonomous only those choices made under the influence of particularly overbearing power, and not simply because of deference to authority, be it of the state or of any other person or corporate entity.”).

Although he did not address the government’s campaign speech specifically, Professor Greene has thoughtfully demonstrated that government speech in general threatens individual liberty only in those rare instances when it is monopolistic or coercive. He explains coercion as “choice under a kind of pressure that allows us fairly to say, ‘she did not choose; she was compelled.’. . . Although there are difficult cases at the margin, we generally accept the distinction between coercion and persuasion, deeming action pursuant to persuasion a proper exercise of autonomy.” 109 Id. at 41–42. The Supreme Court has recognized the difference between coercive and persuasive speech in the private context as well—for example, by interpreting the First Amendment to permit statutory restrictions on private parties’ speech that rises to the level of regulable conduct because of its coercive effects. See, e.g., Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 457 (1978) (“[I]n-person solicitation may exert pressure and often demands an immediate response, without providing an opportunity for comparison or reflection.”); NLRB v. Gissel Packing Co., 395 U.S. 575, 618–20 (1969) (“If there is any implication that an employer may or may not take action solely on his own initiative for reasons unrelated to economic necessities and known only to him, the statement is . . . a threat of retaliation based on misrepresentation and coercion, and as such without the protection of the First Amendment. . . . [E]mployees, who are particularly sensitive to rumors of plant closings, take such hints as coercive threats rather than honest forecasts.” (footnote omitted)); Watts v. United States, 394 U.S. 705, 707 (1969) (per curiam) (“The Nation undoubtedly has a valid, even an overwhelming, interest in protecting the safety of its Chief Executive and in allowing him to perform his duties without interference from threats of physical violence.”). As he observes:

[O]ne of the basic principles underlying the strong protection of private speech is that the persuasive effect of speech is not a proper reason for regulation. Only when there is no time for counterargument, or when persuasion will not cure the original harm, do we permit regulation of speech.

We should think no differently about government speech. Assuming that dissent is open, and assuming no monopolization, we should consider even quite persuasive government speech to be just that, quite persuasive, and hold fast the distinction between persuasion and coercion. 110Greene, supra note 108, at 43 (footnote omitted).

To illustrate the difference between the government’s coercive and persuasive efforts, recall Wooley v. Maynard, where the Supreme Court held that New Hampshire could not require an objecting private speaker to display the state’s motto on his car’s license plate. 111430 U.S. 705, 713 (1977). For further examples of government coercion of individual expression, see Rutan v. Republican Party of Illinois, 497 U.S. 62, 74 (1990); and Elrod v. Burns, 427 U.S. 347, 355–56 (1976) (plurality opinion). As the Court held, punishing an individual for refusing to deliver the government’s chosen message constitutes governmental coercion, undermining a dissenting individual’s liberty by forcing her to express views with which she disagrees. 112 Wooley, 430 U.S. at 715. But Wooley also suggests the additional proposition that the government retains the power to express its own views and values, so long as it does not force others to join or share those views; indeed, the Wooley Court raised no quarrel with New Hampshire’s expressive choice to feature its motto “Live Free or Die” on the state’s license plates, only with its efforts to coerce dissenters to display the motto against their will. 113 See id. at 713; see also id. at 717 (“The State is seeking to communicate to others an official view as to proper appreciation of history, state pride, and individualism. Of course, the State may legitimately pursue such interests in any number of ways.”).

To further illustrate the distinction between government persuasion and government coercion, contrast the government’s various policy responses to the health risks created by smoking. The Surgeon General’s 1964 report on the dangers of tobacco offers an example of government persuasion, rather than coercion: the public was free to read it, or not; those who read it were free to change their smoking habits, or not. 114 See Pub. Health Serv., U.S. Dep’t of Health, Educ., & Welfare, Pub. No. 1103, Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service (1964). The public is similarly free to accept, reject, or entirely ignore government speech on contested issue campaigns. In contrast, the government’s requirement that tobacco companies post warnings on cigarette packages and advertisements 11515 U.S.C. § 1333 (2006). is coercive (albeit potentially still constitutional as a factual disclosure required of a commercial speaker to ensure adequate consumer information): tobacco companies face fines 116 Id. § 1338. and other sanctions 117 Id. § 1339. if they fail to comply. 118Although government coercion of private parties’ speech raises substantial First Amendment issues, such regulation is not always unconstitutional—especially in the context of commercial speech like cigarette labels and advertisements. See, e.g., Commonwealth Brands, Inc. v. United States, 678 F. Supp. 2d 512, 532 (W.D. Ky. 2010) (holding that such regulation of commercial speech satisfies the First Amendment because it is sufficiently tailored to meet a substantial government interest).

As another example, some expressive settings may offer the potential for coercing listeners’ beliefs or behavior. For instance, the government appears to be acting more as a regulator than as just another participant in the marketplace of ideas when it speaks without disclosing its message’s governmental origins: its deception as to the source of its message suggests an effort to mislead the public into giving the message greater credibility than it would otherwise. 119 See Gia B. Lee, Persuasion, Transparency, and Government Speech, 56 Hastings L.J. 983, 1009–15 (2005); Lawrence Lessig, The Regulation of Social Meaning, 62 U. Chi. L. Rev. 943 (1995). Because purposefully masking a message’s governmental source may improve its reception in certain circumstances, government manipulation of the public’s attitudes toward its views by deliberately obscuring its identity as the speaker smacks more of coercion than simple persuasion. 120 See Lee, supra note 119, at 984–88.

The government’s monopolistic speech to a captive audience also raises the possibility of coercing listeners’ beliefs or behavior, in that listeners may not be free to avoid, resist, or counter the government’s message. 121 See Caroline Mala Corbin, The First Amendment Right Against Compelled Listening, 89 B.U. L. Rev. 939 (2009) (suggesting a constitutional right against government-compelled listening, especially in contexts that raise captive-audience concerns like state-mandated abortion counseling or diversity training); Redish & Finnerty, supra note 106, at 99 (identifying free speech concerns with respect to government speech in public schools that seeks to indoctrinate “a captive audience of undeveloped and impressionable minds”). Indeed, concerns of this type may explain at least some of the initial controversy when, shortly after his inauguration, President Obama announced plans to deliver a speech to be broadcast to all public schools 122 See Joshua Rhett Miller, Critics Decry Obama’s ‘Indoctrination’ Plan for Students, FOXNews.com (Sept. 2, 2009), http://www.foxnews.com/politics/2009/09/02/critics-decry-obamas-indoctrination-plan-students/. —a controversy that largely evaporated upon disclosure of the speech’s apolitical content. 123 See Obama Urges Students to Work Hard, Stay in School, CNN.com (Sept. 8, 2009, 7:17 PM), http://www.cnn.com/2009/POLITICS/09/08/obama.school.speech/index.html. In evaluating whether and when government expression is coercive, Professors Redish and Finnerty have similarly emphasized the importance of context:

As a general matter, no one worries that the government is seeking to indoctrinate adults when, for instance, the President gives his State of the Union address or some administrative agency issues a report on an issue of national concern. The reasons are obvious: Adults are not a captive audience compelled to listen to the governments speech, and often another party presents competing speech as a countermessage. Moreover, although young minds are not fully developed and presumably are more susceptible to indoctrination because they lack the ability to think critically and evaluate messages, we presume adults are capable of analyzing the variety of messages they hear on a daily basis. 124Redish & Finnerty, supra note 106, at 90.

But contemporary debates over ballot measures or legislative proposals rarely, if ever, involve government’s monopolistic speech to a captive audience. Recall that governmental campaign speech generally takes the form of government officials’ statements and press releases critical or supportive of pending ballot or legislative measures, as well as government agencies’ reports and analyses, flyers, pamphlets, newsletter articles, website postings, and print and broadcast advertisements communicating their view of such measures to the public. In none of these contexts is the audience captive, nor is there any paucity of opportunity for dissent and counterargument. 125Indeed, monopolistic government speech is increasingly unlikely given changes in communications technology that enable more speakers to participate, expanding opportunities for dissent and counterspeech. See Schauer, supra note 72, at 380 (“Although the quantity of government speech is no doubt increasing, and technological advances may make it possible for the government to convey its messages more efficiently, we should not forget that concomitant increases and improvements in private sector communication probably match this development. There seems to be no evidence whatsoever that the proportion of government speech within the total universe of communication is increasing. In fact, it seems quite possible that it is decreasing.”).

In short, there remains a meaningful distinction between government coercion and government persuasion. 126Although I see a real distinction between the government’s persuasive and coercive speech, I am considerably more skeptical that another distinction proposed by some courts—permitting the government’s informational or factual but not persuasive expression—is meaningful. See, e.g., Harrison v. Rainey, 179 S.E.2d 923, 925 (Ga. 1971) (authorizing expenditures for a legal memorandum on the effects of, but not for advocacy materials regarding, a proposed state constitutional amendment); Rees v. Carlisle, 153 P.3d 1131, 1138–39 (Haw. 2007) (interpreting a statute to permit the government’s informational but not persuasive speech on a pending ballot measure); Stern v. Kramarsky, 375 N.Y.S.2d 235, 239–40 (Sup. Ct. Spec. Term 1975) (suggesting that the government’s “factual” speech on the ERA, but not its advocacy in support of the proposal, is legally permissible). Efforts to distinguish between “factual” and persuasive speech face often-insuperable challenges because the speaker’s most basic choices about her “factual” presentation can be intentionally and effectively persuasive, depending on quantity, order, contrast, tone, etc. See Martin H. Redish & Abby Marie Mollen, Understanding Post’s and Meiklejohn’s Mistakes: The Central Role of Adversary Democracy in the Theory of Free Expression, 103 Nw. U. L. Rev. 1303, 1338–39 (2009) (“[I]t is . . . impossible to separate information from opinion because the use of information is often a central element in the persuasive nature of opinion. . . . [I]nformation is often conveyed, not as an end in itself, but rather as part and parcel of an effort to employ the system of public discourse to achieve certain normative and personal goals.”). That the government acts as both regulator and as speaker does not mean that we cannot parse those roles for constitutional purposes. 127 See Alan K. Chen, Right Labels, Wrong Categories: Some Comments on Steven D. Smith’s Why Is Government Speech Problematic?, Denv. U. L. Rev. Online (Aug. 12, 2010, 9:14 AM), http://www.denverlawreview.org/government-speech/2010/8/12/right-labels-wrong-categories-some-comments-on-steven-d-smit.html (“First Amendment doctrine appropriately distinguishes between the[] two scenarios because the government’s own speech can rarely influence the public debate in the same qualitative or quantitative way as when it excludes private speakers’ ideas from the marketplace.”).

2. Expectations of the Governments Expressive Neutrality Are Unwise and Unrealistic

Not only is the government’s expressive neutrality unnecessary to prevent government coercion and protect individual autonomy, but expectations of such neutrality are also generally impracticable. Numerous scholars have directly (and, in my opinion, powerfully and effectively) challenged the notion that the government’s role as sovereign requires its neutrality on contested issues. Joseph Tussman, for example, observed that “[t]he danger in the careless use of notions of neutrality and non-partisanship is that the concern for fairness may be taken as requiring the relinquishing of commitment.” 128 Joseph Tussman, Government and the Mind 80 (1977). More recently, Abner Greene urged that “government in a liberal democracy not only may promote contested views of the good, but should do so, as well.” 129Greene, supra note 108, at 2. As Steven Smith noted, “[T]here is nothing inherently problematic about government speech, even on matters about which citizens energetically disagree,” 130Smith, supra note 46, at 952. and as Alan Chen observed, “[N]eutrality would forbid the government ever to adopt a policy, which is the function of government in the first instance.” 131Chen, supra note 127.

I join these commentators in rejecting the contentions of Professors Kamenshine, Ziegler, and other government speech critics 132 See supra notes 41–43 and accompanying text. that the government should—or indeed, could, as a practical matter—remain neutral on contested policy matters. Such a thin conception of the government’s appropriate functions is neither wise as a matter of policy nor realistic as a matter of observing and describing how democratic governments actually work. 133 See Nat’l Endowment for the Arts v. Finley, 524 U.S. 569, 598 (1998) (Scalia, J., concurring in the judgment) (“It is the very business of government to favor and disfavor points of view on (in modern times, at least) innumerable subjects—which is the main reason we have decided to elect those who run the government, rather than save money by making their posts hereditary.”). Indeed, the government’s noncoercive speech generally furthers the other values most often identified at the heart of the First Amendment: facilitating participation in democratic self-governance by informing voters of their government’s priorities and encouraging the discovery of truth and dissemination of knowledge by adding to the marketplace of ideas.

For these reasons, expectations of the government’s expressive neutrality are now a dead letter as a matter of constitutional law outside the campaign context because the Supreme Court’s emerging government speech doctrine acknowledges that the government inevitably has views that the Constitution permits it to express. Indeed, even those Justices skeptical of the majority’s approach to government speech nonetheless also reject the premise that the First Amendment bars government from spending money to articulate a view with which some (or even many) taxpayers disagree. 134Johanns v. Livestock Mktg. Ass’n, 544 U.S. 550, 559 (2005); id. at 574 (Souter, J., dissenting) (“The first point of certainty is the need to recognize the legitimacy of government’s power to speak despite objections by dissenters whose taxes or other exactions necessarily go in some measure to putting the offensive message forward to be heard. To govern, government has to say something, and a First Amendment heckler’s veto of any forced contribution to raising the government’s voice in the ‘marketplace of ideas’ would be out of the question.” (footnote omitted)). Recognizing that an effective government must take positions on a wide variety of matters, even the dissenters agreed that private speakers can constitutionally be compelled to pay for government speech from which they dissent. 135 See id. In considering this First Amendment challenge to a generic beef promotion campaign implemented by the Department of Agriculture and funded by taxes targeted at beef producers, the Johanns majority and dissenters differed vigorously, however, on the question whether government must identify itself as the source of that speech to successfully assert the government speech defense to the plaintiffs’ free speech claim. Compare id. at 562 (majority opinion) (“When . . . the government sets the overall message to be communicated and approves every word that is disseminated, it is not precluded from relying on the government-speech doctrine merely because it solicits assistance from nongovernmental sources in developing specific messages.”), with id. at 571–72 (Souter, J., dissenting) (“[I]f government relies on the government-speech doctrine . . . , it must make itself politically accountable by indicating that the content actually is a government message, not just the statement of one self-interested group the government is currently willing to invest with power.”).

Although this subpart has concluded that expectations of the government’s expressive neutrality are generally both unwise and unrealistic, the possibility remains that certain government speech in certain contexts threatens specific dangers that may justify its constraint. This leads to the next question of whether there are specific dangers posed not by the government’s deviation from expressive neutrality generally, but instead by the government’s deviation from expressive neutrality in the campaign context specifically. In considering this question, the remainder of this Part draws from lessons learned in parallel debates over when, if ever, private speakers’ persuasive efforts in the campaign context should be considered sufficiently threatening to First Amendment values to justify their regulation.

B. Responding to Critiques of Powerful Parties Campaign Speech

Some critics perceive governmental campaign speech on pending ballot or legislative measures as repugnant to an equality-based view of the Free Speech Clause. In other words, they see the government’s campaign speech as tilting the playing field to its advantage because the government’s greater power, prestige, and resources may ultimately change actual political outcomes in ways that are fundamentally unfair to dissenting speakers. They fear that this is especially, but not only, the case when government campaign speech is supported by public funds for printing, dissemination, and other costs. Note that this objection is a critique of the instrumental effects of campaign speech by powerful actors generally, where government is among those powerful actors. In other words, whereas the arguments described in the preceding subpart focused on the government as a unique speaker because of its sovereign status, the arguments explored in this subpart focus on campaign speech by powerful parties, without necessarily emphasizing their governmental or nongovernmental status.

Recall that the First Amendment is often understood to protect speech (especially political speech) not only to protect the individual liberty interests of potential speakers but also to serve certain instrumental goals by ensuring listeners’ access to views that will inform their decisions (and especially their political decisions). 136 See Alexander Meiklejohn, Free Speech and Its Relation to Self-Government 25 (1948) (“The final aim of the [town] meeting is the voting of wise decisions. The voters, therefore, must be made as wise as possible. The welfare of the community requires that those who decide issues shall understand them. They must know what they are voting about. And this, in turn, requires that so far as time allows, all facts and interests relevant to the problem shall be fully and fairly presented to the meeting. . . . What is essential is not that everyone shall speak, but that everything worth saying shall be said.”); Redish & Mollen, supra note 126, at 1307 (“[A] valid democratic theory of the First Amendment must protect all speech that allows individuals to discover their personal needs, interests, and goals—in government and in society at large—and to advocate and vote accordingly.”); see also Garrison v. Louisiana, 379 U.S. 64, 77 (1964) (recognizing “the paramount public interest in a free flow of information to the people concerning public officials, their servants”). As Professors Redish and Mollen explain:

[P]ublic opinion cannot be formed without the evaluation and ultimate acceptance of certain ideas over others, the listener participates in the formation of public opinion as much as the speaker does. Just as the speaker may benefit by contributing to public discourse, so too may listeners moral and intellectual horizons be expanded by the receipt of information and opinion. Their ability to function as active participants in a democracy is improved as a result. More importantly, governments decision to insulate citizens from information and opinion because of a paternalistic distrust of their ability to make wise choices is as threatening to core democratic values as the suppression of any speaker. 137Redish & Mollen, supra note 126, at 1337 (footnotes omitted).

To achieve these instrumental aims, First Amendment theory thus often assumes the “rational actor” model that the more speech from all sources, the better for listeners’ ultimate decision making. 138As Lyrissa Barnett Lidsky explains, “[T]he modern trend . . . has been to assume that audiences are savvy and sophisticated, capable of sorting through masses of information to discover truth, however provisional or contested. Indeed, these assumptions underpin two articles of faith in modern First Amendment theory: (1) audiences are capable of rationally evaluating the truth, quality, credibility, and usefulness of core speech without the aid of government intervention; and (2) more speech is better than less.” Lyrissa Barnett Lidsky, Nobody’s Fools: The Rational Audience as First Amendment Ideal, 2010 U. Ill. L. Rev. 799, 810–11. Under this view, one party’s speech does not threaten the equality interests of other speakers because all are free to compete in the marketplace of ideas.

On the other hand, many skeptics remain unconvinced that more speech inevitably improves listeners’ decision making. 139 See, e.g., James A. Gardner, Comment, Protecting the Rationality of Electoral Outcomes: A Challenge to First Amendment Doctrine, 51 U. Chi. L. Rev. 892, 892 (1984) (“A central though rarely articulated premise of many election laws and much democratic theory is that electoral outcomes should be rational rather than irrational—that they should reflect the true, reasoned, and informed choice of the people. Unfortunately, as political scientists have shown, people do not always vote rationally.”). This skepticism is illustrated by longstanding efforts to regulate private parties’ campaign speech, which are equality-focused efforts based in part on the premise that certain powerful speakers’ political expression will overwhelm or otherwise undermine the views of dissenting speakers. 140 See, e.g., J. Skelly Wright, Money and the Pollution of Politics: Is the First Amendment an Obstacle to Political Equality?, 82 Colum. L. Rev. 609, 640 (1982) (“When the forum is limited by physical, technological, or economic factors, the messages of some speakers must be limited if all points of view are to be heard, so that the audience may enjoy a full range of uninhibited debate.”).

This divide over the accuracy of the rational actor model (and thus over the comparative benefits and dangers of powerful parties’ campaign speech) in many ways mirrors that described by Daniel Ortiz in understanding voters either as “civic smart[ies]” (“individuals [who] make highly informed political choices” after carefully acquiring and sorting available information) or as “civic slob[s]” (passive and uniformed voters who do not engage in the same cognitive effort, instead relying largely on “images, feelings, and emotions”). 141Daniel R. Ortiz, The Engaged and the Inert: Theorizing Political Personality Under the First Amendment, 81 Va. L. Rev. 1, 4 (1995). Professor Ortiz himself finds both models incomplete. Ortiz, supra. Although one need not rely on the pejorative “slob” to conclude that many time-strapped voters rely on shortcuts to help inform their decisions, one’s view of the instrumental value and dangers of campaign speech (by any speaker, private or governmental) turns in great part on one’s assessment of how listeners process speech of this type.

For those who identify a—and perhaps the—primary First Amendment value as facilitating participation in democratic self-governance, 142 See, e.g., Robert Post, Participatory Democracy and Free Speech, 97 Va. L. Rev. 477, 482 (2011) (“[T]he best possible explanation of the shape of First Amendment doctrine is the value of democratic self-governance.”). the key question then becomes whether the government’s expressive participation in debates over ballot and legislative measures furthers or frustrates that aim. The remainder of this Part explores how the government’s campaign speech can help inform the decisions of both informed and comparatively uninformed voters without threatening the equality interests of other speakers.

1. Governmental Campaign Speech Can Provide Valuable Information for Comparatively Sophisticated Voters

Government expression can valuably further citizens’ capacities to participate in democratic self-governance by enabling them to identify and assess—positively or negatively—their government’s priorities and performance. 143 See, e.g., 2 Zechariah Chafee, Jr., Comm’n on Freedom of the Press, Government and Mass Communications 723 (1947) (“Now it is evident that government must itself talk and write and even listen.”); Thomas I. Emerson, The System of Freedom of Expression 698 (1970) (“Participation by the government in the system of freedom of expression is an essential feature of any democratic society. It enables the government to inform, explain, and persuade—measures especially crucial in a society that attempts to govern itself with a minimum use of force. Government participation also greatly enriches the system; it provides the facts, ideas, and expertise not available from other sources.”); Martin H. Redish & Daryl I. Kessler, Government Subsidies and Free Expression, 80 Minn. L. Rev. 543, 565 (1996) (“From the perspective of democratic theory, it is essential that these government employees inform the populace of the government’s policies and initiatives. Because the government informs the populace about its functioning through these subsidies, it facilitates self-government by providing members of the community with information and data on which to judge the performance of its political leaders. As a result, the electorate is better able to check elected officials and hold them accountable.”); Shiffrin, supra note 55, at 604 (describing government speech as providing the public with “the advantage of knowing the collective judgment of the legislature and of knowing the views of its representatives, which would in turn be useful for evaluating them”). For these reasons, as I have urged elsewhere, 144 See Helen Norton, Constraining Public Employee Speech: Government’s Control of Its Workers’ Speech to Protect Its Own Expression, 59 Duke L.J. 1 (2009); see also Norton & Citron, supra note 5. the primary value of government speech turns not on its popularity or even its wisdom, 145Recall, as just one of many examples, the insights into government policymaking provided to the public by the Pentagon Papers (a Department of Defense study that reviewed U.S. military and diplomatic policy in Indochina). See N.Y. Times Co. v. United States, 403 U.S. 713, 714 (1971) (per curiam) (rejecting the government’s efforts to stop publication of the Pentagon Papers). but instead on its transparency—when members of the public can actually identify the government as the message’s source, “maximiz[ing] prospects for meaningful credibility assessment and political accountability.” 146Helen Norton, The Measure of Government Speech: Identifying Expression’s Source, 88 B.U. L. Rev. 587, 632 (2008). Examples of transparency in this context include speeches made by or quotes attributed to specified government officials in their official capacity, as well as press releases, reports, pamphlets, online postings, and advertisements by identified government agencies. In this way, government expression provides great instrumental value because of what it offers its listeners: information that furthers the public’s ability to evaluate its government. 147 Id.

Moreover, the government’s voice may be particularly knowledgeable on some issues, potentially adding valuable new perspectives to the ideas and arguments available for voters’ consideration. 148 Id. at 589–90. Indeed, because many initiatives and referenda directly implicate government services and their financing, affected governmental bodies may have unique expertise to offer voters on the merits of such measures. Consider, as just one example, a series of three 2010 Colorado ballot measures generated by private parties that proposed to cut at least a billion dollars annually in state taxes (and thus government funding). 149 See Steve Raabe, Ritter Bashes Tax-Cut Initiatives, Denver Post, May 14, 2010, at A1. Supporters characterized the measure as “forc[ing] the government to operate more efficiently and cut bloated spending.” 150 Id. In response, the state’s governor joined business leaders to urge voters to reject the measures, arguing that they could have a devastating effect on the state economy that “would set Colorado back a generation.” 151 Id. (internal quotation mark omitted). More specifically, “[Governor] Ritter pledged to rally against the three ballot measures at every speech he gives until the vote in November,” 152Aldo Svaldi, Political Foes Turn Partners, Denver Post, May 24, 2010, at A18. “call[ing] them ‘three of the most backward-thinking ballot measures this state has ever seen.’” 153Editorial, Candidates Weigh In on Right Side of Ballot Items, Denver Post, Feb. 28, 2010, at D3. The point is not that the government’s views are necessarily correct, but instead that they may provide value to voters by exposing the views of those who would be charged with implementing the measure if enacted.

Those who are inclined to characterize listeners as “civic smarties” should thus be especially comfortable with the addition of the government’s voice to the marketplace of ideas in contested policy debates because the government’s campaign expression contributes to the information available to voters for consideration. 154The Supreme Court has noted on several occasions the informational value of private parties’ campaign speech with respect to contested ballot measures when rejecting limitations on such speech. See Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290, 299 (1981) (“Whatever may be the state interest . . . in regulating and limiting contributions to or expenditures of a candidate[,] . . . there is no significant state or public interest in curtailing debate and discussion of a ballot measure.”); First Nat’l Bank of Bos. v. Bellotti, 435 U.S. 765, 777 (1978) (“The inherent worth of the speech in terms of its capacity for informing the public does not depend upon the identity of its source, whether corporation, association, union, or individual.”).

2. Governmental Campaign Speech Can Provide a Valuable Heuristic for Comparatively Uninformed Voters

The government’s campaign speech may also offer great benefit even to those inclined to adopt the “civic slob” model when describing voters’ behavior. Transparently governmental speech provides a valuable heuristic, or cognitive shortcut, for those with neither the time nor the expertise to analyze the competing arguments themselves. Indeed, a significant body of evidence suggests that reliance on heuristics may enable comparatively uninformed voters to vote as “competently”—to vote in a way consistent with their own policy preferences—as comparatively well-informed voters. 155 See, e.g., James N. Druckman, Does Political Information Matter?, 22 Pol. Comm. 515, 515 (2005) (summarizing a study finding that “citizens can compensate for a lack of political information by using shortcuts to make the same decisions they would have made if they had that information”—“that many poorly informed voters (who lacked knowledge about the initiatives’ details) used elite endorsements (e.g., from interest groups) to emulate the behavior of well-informed . . . voters”); Michael S. Kang, Democratizing Direct Democracy: Restoring Voter Competence Through Heuristic Cues and “Disclosure Plus, 50 UCLA L. Rev. 1141, 1161 (2003) (“[R]eliance on heuristic cues is a learned practice based on past success and accuracy. Voting behavior in candidate elections, when heuristic cues are readily available, is relatively rational, consistent, and well-ordered, whereas in issue elections, particularly when heuristic cues are difficult to find, voting behavior appears more random, irrational, and contradictory. Without heuristic cues, voters in direct democracy are more confused, [and] money is more influential . . . . Even if they do not cure voter confusion in every instance, voters armed with heuristic cues will be much more likely to vote competently in the face of complexity than will voters without them.” (footnote omitted)). For a more critical assessment of heuristics’ value to voters, see Molly J. Walker Wilson, Behavioral Decision Theory and Implications for the Supreme Court’s Campaign Finance Jurisprudence, 31 Cardozo L. Rev. 679, 703–06 (2010).

As Michael Kang argues specifically with respect to ballot measures:

[T]he source of voter confusion in direct democracy is not political ignorance or heavy campaign spending, as commonly alleged, but the scarcity of “heuristic cues” . . . . [S]trengthening heuristic cues in direct democracy offers the best means of rehabilitating voter competence pragmatically, at low cost, without trying to force voters to adjust the way they think about politics. 156Kang, supra note 155, at 1141; accord id. at 1164–65 (“[M]uch of the electorate is rationally ignorant and unlikely to become more engaged in a way that cures concerns about voter confusion. . . . Heuristic cues quickly put uninformed voters on roughly equal footing with better-informed voters, even if they do not transform civic slackers into infallible or perfectly informed voters.”).

Among the most effective of these heuristics is knowledge of the opinions of trusted—or distrusted—third parties, who might include experts, community leaders, and government speakers. As a general illustration of government as comparatively trusted speaker, recall the Surgeon General’s report on the dangers of tobacco, which responded to the tobacco industry’s well-financed speech. 157 See, e.g., Pub. Health Serv., supra note 114 (describing the adverse health effects of smoking). This dynamic proves accurate in the campaign context as well, as Professor Kang explains:

Underlying policy specifics are less useful to the average citizen than knowing the synthesized opinion of a trusted leader. As a result, when they know what an identifiable politician thinks about an issue, people report a great deal more certainty about their own attitude, compared to when they based their attitude purely on policy information.

. . . .

A normative endorsement of heuristic reasoning thus flows from a realistic acknowledgment of the central role that political elites play in American politics. It is a wishful endeavor to pray that citizens can become better individual democrats, without also considering the powerful function of politicians, activists, interest groups, and other elites. . . . In short, citizens depend on political elites to gather political information and synthesize deliberative judgments for them. 158Kang, supra note 155, at 1161–62.

Providing voters access to the government’s transparently sourced views on often-confusing ballot measures thus provides the public with a potentially valuable heuristic. 159 See id. at 1169 (“People have plentiful access to political information, but otherwise busy and semi-interested voters need information disseminated to them in a way that requires them to take no affirmative steps or do anything more than they otherwise would do. Unless campaign finance information is delivered to voters at virtually no cost to them, such information may never reach them and will always remain too difficult for voters to acquire. Voters must be made aware of such heuristic cues to take advantage of them.”).

Again, government’s views are by no means necessarily wise, popular, or persuasive; instead, they simply provide considerable heuristic value to voters who know whether their values align with those of the government speaker. 160 See Schauer, supra note 72, at 385 (“In the face of almost completely unrestrained criticism of government from all quarters, . . . views about the ability of government to use speech to falsify consent seem to require such a negative view of popular competence as to call into question the very reasons for considering democracy or majoritarianism to be any good at all.”). Whether the government’s views persuade or dissuade any particular voter depends on whether that voter views the government speaker with trust or distrust. 161 See, e.g., June Fessenden-Raden et al., Providing Risk Information in Communities: Factors Influencing What Is Heard and Accepted, Sci. Tech. & Hum. Values, Summer/Fall 1987, at 94, 96 (finding that the trust people have in political institutions varies, as does their trust in the information provided by such governments); Schauer, supra note 72, at 381 (“[A]ntigovernment biases may be so great, particularly with reference to the veracity of political leaders, that much government speech may encounter a public strongly predisposed to disbelief.”); Leigh Contreras, Comment, Contemplating the Dilemma of Government as Speaker: Judicially Identified Limits on Government Speech in the Context of Carter v. City of Las Cruces, 27 N.M. L. Rev. 517, 539–40 (1997) (“[V]oters may tend to defer to the government’s judgment—although the assumption that government has an especially influential effect on the people depends on the public’s reverence and respect for government.” (footnote omitted)). Indeed, some number of voters will always disagree with the government.

3. Governmental Campaign Speech Can Provide a Valuable Response to Campaign Speech from Powerful, Private Sources

Government speech, moreover, can be especially important to an equality-based conception of free speech when its voice counters that of powerful, private speakers. 162Abner Greene has made this point about government speech more generally, outside of the specific campaign context that is the focus of this Article. See Greene, supra note 108, at 9, 11 (“[G]overnment speech often makes a distinctive contribution to public debate. If the government’s point of view were simply corroborative of private points of view, the affirmative argument for government speech would weaken as the concerns about government power rise. . . . Moreover, government persuasion on a contested matter plays an important role in countering private power. As one locus of power in society, government can check agglomerations of private power, just as checks on government ensure that it be only one voice among many. Additionally, even in a contested arena government speech can help foster debate, fleshing out views, and leading toward a more educated citizenry and a better chance of reaching the right answer.” (footnote omitted)). This remains especially true in the specific context of often-confusing ballot measures, where the government’s expression not infrequently counters that of powerful, private interests. Indeed, government expression’s ability to further the First Amendment’s democracy-enhancing purposes may now be particularly heightened in light of the current Supreme Court’s unwillingness to permit limitation on the quantity of political speech from corporate and other private interests. In Citizens United v. FEC, the Court held that the First Amendment prohibits efforts to regulate corporations’ independent expenditures to support or oppose candidates for office. 163130 S. Ct. 876, 908–09 (2010). Emphatically rejecting equality-based arguments that such regulations justifiably protect the marketplace of ideas from distortions due to the disproportionate volume of speech from wealthy speakers, 164 Id. at 904–05. the majority refused to characterize certain speech as unusually dangerous based on its (corporate) source. 165 Id. at 905. For arguments that Citizens United was wrongly decided even if one focuses on a liberty-based view of the First Amendment, see Molly J. Walker Wilson, Too Much of a Good Thing: Campaign Speech After Citizens United, 31 Cardozo L. Rev. 2365 (2010).

Now that the quantity of corporate political speech is largely unregulated, government expression that counters that of private interests may prove especially valuable to both informed and comparatively uninformed voters. For example, in the 2008 election cycle, the governor of Colorado engaged in a “statewide ‘Truth Tour’” aimed at countering the arguments by powerful, private opponents of a ballot measure to end a tax credit for the oil and gas industry. 166Gayle Perez, Ritter: Amendment 58 Foes Using Scare Tactics, Pueblo Chieftain (Oct. 10, 2008), http://www.chieftain.com/news/local/ritter-amendment-foes-using-scare-tactics/article_e7c5ae2f-0d70-56e4-a4cb-26096f2209ee.html (“Ritter said the amendment will not raise taxes, but simply removes the tax credit that oil and gas companies have been receiving from the state for more than three decades. Of the money the state will keep from those credits, at least 60 percent will be devoted to providing financial aid to low-income resident[] students who attend a state public college or university. The remaining money will be used for wildlife habitat protection, renewable energy grants and to local communities impacted by oil and gas industry.”). The governor’s advocacy came in response to assertions by the amendment’s opponents, who included oil and gas companies and the Denver Metro Chamber of Commerce, that terminating the tax credit would “drive oil and gas companies to other states. That move will negatively impact Colorado’s economy, taking away jobs and adversely affecting those cities and towns that rely on the industry.” 167Cari Merrill, Dems Make Case for 58, Fort Collins Coloradoan, Sept. 27, 2008, at A11. Again, the point is not that the government’s views are necessarily correct, but instead that they may provide value by responding to speech from powerful, private parties that might otherwise not face effective rebuttal. 168 See, e.g., David Michaels, Doubt Is Their Product: How Industry’s Assault on Science Threatens Your Health 85–90, 201 (2008) (documenting efforts by tobacco companies to contest claims of the health hazards of cigarettes and by oil companies to cast doubt on scientists’ claims on the role of fossil fuels in contributing to climate change).

Along these lines, the government’s campaign speech may also be especially important given the frequent use of ballot measures by powerful, private parties to seek to restrict minority rights. Consider, as just one of many examples, the California Real Estate Association’s 1964 proposed ballot measure to repeal the state’s fair housing law that prohibited racial discrimination in the sale or lease of property. 169 See California: Proposition 14, Time, Sept. 25, 1964, at 23, 23, available at http://www.time.com/time/magazine/article/0,9171,876158-1,00.html. Opposed to any “open housing” laws, 170 Id. the real estate association started with one hundred thousand dollars in “seed money.” 171Totton J. Anderson & Eugene C. Lee, The 1964 Election in California, 18 W. Pol. Q. 451, 470 (1965). Opponents of the measure included not only a wide variety of civil rights and other private groups but also California Governor Pat Brown 172 Id. at 470–71. and Senator Pierre Salinger. 173 California Proposition 14, supra note 169, at 23.

Of course, government speakers are not monolithic in their views on this or any other issue. Three decades later, California Governor Pete Wilson and University of California Regent Ward Connerly were among the governmental speakers that supported Proposition 209, a ballot proposition that amended the state constitution to prohibit state and local governments’ affirmative action programs in California. 174Pete Wilson et al., Argument in Favor of Proposition 209, in California Ballot Pamphlet: General Election, November 5, 1996, at 32 (1996), available at http://vote96.sos.ca.gov/Vote96/html/BP/209yesarg.htm. Again, the point is simply that the government’s voice adds to those available to voters collecting information about pending ballot measures.

Not only are ballot measures often the subject of campaign speech by powerful and well-financed private parties, 175 See Kang, supra note 155, at 1148 (“Many direct democracy elections, particularly on economic measures, attract spectacular disparities in campaign spending between opposing and supporting sides. Indeed, the consensus from empirical research is that spending advantages are nearly outcome-determinative when aimed at defeating a ballot measure.”). but the identity of such powerful, private parties may not be clear, obscuring the public’s efforts to assess such speakers’ self-interest and credibility:

Interest groups strategically obscure their involvement when they believe identification would hurt their campaigns. Many industry groups form political committees to conduct campaign activities under nondescript names like Californians for Paycheck Protection (religious conservatives supporting limitations on labor union political activity), Alliance to Revitalize California (Silicon Valley executives supporting a tort reform measure), and Californians for Affordable and Reliable Electrical Service (industry opponents of utility regulation). 176 Id. at 1158–59.

Similarly, advocates for statewide initiatives seeking to ban all forms of affirmative action named themselves the “American Civil Rights Institute,” an identifier that at least some voters may have found confusing. 177 See Chris Chambers Goodman, (M)Ad Men: Using Persuasion Factors in Media Advertisements to Prevent a “Tyranny of the Majority” on Ballot Propositions, 32 Hastings Comm. & Ent. L.J. 247, 257 (2010) (“The [organizations’] names are specifically designed, in some[] cases to obfuscate, and these names of the supporters can be very influential in the outcome of the ballot measure.”); see also Bruce E. Cain, Commentary, Garrett’s Temptation, 85 Va. L. Rev. 1589, 1592–93 (1999) (“Many of the groups that succeed in getting initiatives on the ballot have primarily economic motives—e.g., insurance companies, lawyers, and teachers’ unions. To make matters worse from an informational perspective, these groups often adopt false, generic labels such as ‘Committee for a Just America’ or ‘Campaign for Consumer Justice.’ . . . In an ideal world, information heuristics would operate like warning labels on consumer items. In reality, those who sponsor initiatives know that labels can be important, and they choose labels strategically in order to create images they think voters will receive best.”).

For these reasons, voters often cannot be sure of the source of private campaign speech and are thus deprived of a key cue to the expression’s credibility. These dynamics invite a wide range of policy proposals designed to enhance transparency in the campaign speech context. 178 See Gilda R. Daniels, Voter Deception, 43 Ind. L. Rev. 343, 381–86 (2010) (proposing legislative measures to address voter deception); Goodman, supra note 177, at 294–301 (offering a series of proposals to require greater transparency of the source of private speech in direct-democracy campaigns). Indeed, the Citizens United majority itself invited legislatures to pass transparency-forcing requirements with respect to private campaign speech. 179 See Citizens United v. FEC, 130 S. Ct. 876, 886 (2010) (“The Government may regulate corporate political speech through disclaimer and disclosure requirements, but it may not suppress that speech altogether.”). Whether Congress or state legislatures do so, however, is a matter of political will subject to jurisdictional and temporal variation. 180 See David G. Savage, Corporate Campaign Spending Still Murky, L.A. Times, Oct. 27, 2010, at A1 (“Because of loopholes in tax laws and a weak enforcement policy at the Federal Election Commission, corporations and wealthy donors have been able to spend huge sums on campaign ads, confident the public will not know who they are, election law experts say.”); Editorial, The Secret Election, N.Y. Times, Sept. 19, 2010, at WK8 (noting Congress’s failure to pass disclosure requirements in the immediate aftermath of Citizens United). And even when enacted, such requirements still remain vulnerable to constitutional challenge because the Court has held that disclosure requirements are subject to “exacting” scrutiny. 181Buckley v. Valeo, 424 U.S. 1, 64 (1976) (per curiam). The government’s voice on ballot issues may thus prove especially valuable to voters—both informed and uninformed—by responding to private power that sometimes operates in nontransparent ways.

Not only is private speech often nontransparent, but it is unaccountable to the public in other ways. For example, courts frequently interpret the First Amendment to protect false political speech by private speakers. 182 See Rickert v. Washington, 168 P.3d 826 (Wash. 2007) (en banc) (striking down an amended statute that prohibited sponsoring with actual malice a political advertisement containing a false statement of material fact about a candidate for public office); State ex rel. Pub. Disclosure Comm’n v. 119 Vote No! Comm., 957 P.2d 691 (Wash. 1998) (en banc) (striking down a state statute that prohibited the sponsorship of political advertisements containing false statements of material fact). But see Vanasco v. Schwartz, 401 F. Supp. 87 (E.D.N.Y. 1975) (characterizing calculated falsehoods during political campaigns as unprotected by the First Amendment), aff’d mem., 423 U.S. 1041 (1976). The government’s counterspeech again may serve a valuable checking function. 183 See Frederick Schauer, Facts and the First Amendment, 57 UCLA L. Rev. 897, 917–18 (2010) (“[N]either First Amendment theory, nor history, nor doctrine significantly restrict[s] the government’s ability to attempt to correct widespread public factual inaccuracy . . . .”). I leave for another day the question of whether the Constitution prohibits government falsehoods. For all these reasons, government speech can be understood to further an equality-based conception of free speech when its voice counters that of powerful, private speakers.

To be sure, some who take an equality-based view of the First Amendment may view government speakers—like corporations or other wealthy, private speakers—as nonetheless threatening to the interests of less powerful speakers. For example, they may argue that the government’s inherent power, prestige, and especially resources tilt the playing field such that dissenting speakers cannot fairly compete.

Of course, the government will not inevitably speak in opposition to powerful, private interests; indeed, it is often aligned with them. But closer examination reveals a number of checks that limit the dangers to other speakers’ equality interests posed by government speech. For example, the government is by no means monolithic. 184 See Yudof, supra note 72, at 114–16 (describing how federalism, separation of powers, and the variety of administrative agencies considerably fragment political power, thus undermining the possibility of monolithic government speech). Instead, it comprises a large number and range of potential government speakers—both individual and institutional—with various interests. This suggests the possibility of diverse views even among government speakers, at least some of which may diverge from the views of powerful, private speakers. Indeed, that government speech is by no means monolithic both adds to its informational value and detracts from its potential danger. Different branches of the government can and do disagree in a way that helps inform voters. Such disagreements occur both horizontally (for example, when the executive disagrees with the legislature) and vertically (when federal and state governments disagree). For example, U.S. Attorney General Eric Holder expressed the U.S. Department of Justice’s opposition to a proposed California ballot measure to legalize marijuana on the grounds that it would “greatly complicate federal drug enforcement efforts to the detriment of our citizens.” 185Letter from Eric H. Holder, Jr., Att’y Gen., to Former Administrators of the Drug Enforcement Agency (Oct. 13, 2010) (on file with author). This prompted counter-speech by the measure’s proponents who assailed what they characterized as a defense of a failed war on drugs. 186 John Hoeffel, Holder Vows Fight over Prop. 19, L.A. Times, Oct. 16, 2010, at AA1.

Such expressive tensions inform and spur debate, while lessening the chance that government speech is monopolistic. Moreover, the collective-action problems confronted by institutional government speakers further undermine the notion of government as monolithic, and potentially monopolistic, speaker. Indeed, such problems may encourage such speakers to be more deliberative as a result. And, of course, government speakers remain politically accountable to the populace in ways untrue of private speakers. 187 Compare Joseph Blocher, Viewpoint Neutrality and Government Speech, 52 B.C. L. Rev. 695, 717–18 (2011) (“So long as the government identifies itself when speaking, the public can hold government speakers politically accountable and curb their excesses.”), with Steven Shiffrin, Government Speech and the Falsification of Consent, 96 Harv. L. Rev. 1745, 1752–53 (1983) (reviewing Yudof, supra note 72) (“[C]orporations seek profit without concern for externalities and use enormous wealth in ways bearing no necessary or even likely relationship to the beliefs of their shareholders . . . .” (footnote omitted)).

To be sure, public entities and officials may have considerable self-interest in the outcome of contested ballot measures and legislative proposals. For example, such votes may determine the level of funding available to government. The government’s campaign speech can also be self-interested in ways other than purely financial—as may be the case with government speech opposing term limits or government speech otherwise aligned with what the government speaker sees as its political advantage. But as Professors Redish and Mollen persuasively explain in another context, a speaker’s self-interested motivation does not necessarily negate the value of that information to the listener. 188Redish & Mollen, supra note 126, at 1317. Although most speakers are self-interested in some way, 189 See id. at 1341 (“Speakers do not always speak solely to contribute to public discourse or solely for narrow personal economic gain.”). such motivation rarely justifies limiting their speech (although it may well justify listeners’ skepticism). What the listener needs, ideally, is an understanding of the speaker’s self-interest when evaluating her speech. 190 See First Nat’l Bank of Bos. v. Bellotti, 435 U.S. 765, 791–92 (1978) (“[T]he people in our democracy are entrusted with the responsibility for judging and evaluating the relative merits of conflicting arguments. They may consider, in making their judgment, the source and credibility of the advocate.” (footnote omitted)). As discussed above, such an understanding may be more readily available with respect to government, as opposed to private, speech: voters can assess the government speaker’s self-interest and hold her accountable for it. 191 See Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290, 298 (1981) (recognizing a value to voters of knowing “the identity of those whose money supports or opposes a given ballot measure”); Buckley v. Valeo, 424 U.S. 1, 67 (1976) (per curiam) (“The sources of a candidate’s financial support . . . alert the voter to the interests to which a candidate is most likely to be responsive and thus facilitate predictions of future performance in office.”).

In short, government speech on issue campaigns generally furthers, rather than frustrates, key constitutional interests. Transparently governmental campaign speech often provides great value to the public. It enhances political accountability by informing voters of their governments’ priorities and preferences, provides a valuable heuristic for comparatively uninformed voters, and enriches the marketplace of available ideas and arguments, especially (but not only) as a counter to expression from powerful, private sources.

III. Considering Possible Limitations on Governmental Campaign Speech

The preceding Part concluded that the government does not violate the Constitution simply by taking sides in contested ballot or legislative campaigns. To be sure, however, I do not claim that the government’s campaign speech is (or should be) entirely free from constraint. 192 See Schauer, supra note 72, at 384–85 (“[That the First Amendment does not constrain government speech] does not mean that government communication is not a constitutional problem under some other clause of the Constitution, or that it is not a political or moral problem. . . . When government misuses its power to communicate we do have a problem, but this does not mean that we have a first amendment problem.” (footnote omitted)).

For example, even though the preceding Part concluded that the government’s campaign speech on ballot and legislative measures is consistent with the Free Speech Clause, other constitutional constraints—such as the Establishment Clause—remain in play as independent checks on government speech generally, including, but not limited to, the government’s campaign speech. 193 See supra note 19. The remainder of this Part briefly explores other possible limitations.

A. The Transparency Requirement as a Constitutional Limitation on Government Speech Generally

Recall that the previous Part’s emphasis on the instrumental value of the government’s campaign speech referred only to transparently governmental speech—expression that the public can identify as governmental in source and thus hold the government accountable for it. 194 See supra notes 143–46 and accompanying text. I have urged previously that the government should be permitted to assert the government speech defense to free speech challenges only when it can establish that it expressly claimed the speech as its own when it authorized the communication and that onlookers understood the speech to be the government’s at the time of its delivery. 195 See Norton, supra note 146, at 599 (“[T]he government can establish its entitlement to the government speech defense only when it establishes itself as the source of that expression both as a formal and as a functional matter. In other words, government must expressly claim the speech as its own when it authorizes or creates a communication and onlookers must understand the message to be the government’s at the time of its delivery.”). For other commentators’ thoughtful discussions of these and related issues, see Randall P. Bezanson & William G. Buss, The Many Faces of Government Speech, 86 Iowa L. Rev. 1377 (2001); Caroline Mala Corbin, Mixed Speech: When Speech Is Both Private and Governmental, 83 N.Y.U. L. Rev. 605 (2008); Leslie Gielow Jacobs, Who’s Talking? Disentangling Government and Private Speech, 36 U. Mich. J.L. Reform 35 (2002); and Gia B. Lee, Persuasion, Transparency, and Government Speech, 56 Hastings L.J. 983 (2005). The Supreme Court, however, has yet to impose such a constitutional requirement in the context of government speech more generally 196 See Johanns v. Livestock Mktg. Ass’n, 544 U.S. 550, 563–64 (2005). —a doctrinal failure that I have criticized elsewhere. 197 See, e.g., Norton & Citron, supra note 5. Such a transparency requirement would be especially valuable in the context of the government’s campaign speech on contested policy matters because it would maximize the public’s ability to engage in meaningful political accountability measures as well as in undeceived assessments of the message’s credibility. 198Private speakers, unlike the government, have autonomy interests protected by the First Amendment. For this reason, the Court has interpreted the First Amendment to protect private speakers’ anonymity under certain circumstances. See, e.g., McIntyre v. Ohio Elections Comm’n, 514 U.S. 334, 357 (1995) (striking down Ohio’s ban on the distribution of unsigned political leaflets).

B. Constitutional Limitations on Government Speech on Candidate Campaigns

So far this Article has focused on the government’s speech on issue campaigns—government speech that advocates a position on a contested ballot measure to be decided by the voters or on contested legislation to be considered by another governmental body. In contrast, the use of official government resources to engage in campaign speech endorsing or opposing specific candidates raises potentially greater threats to First Amendment interests in restraining the self-perpetuation of incumbents and preventing the entrenchment of political power. 199Recall that this Article focuses only on speech that represents the views of a government entity or branch, as opposed to individual government officials’ speech expressing their own views on their own time and at their own expense, as individuals do not relinquish their First Amendment rights to express personal beliefs upon taking office. See, e.g., Colo. Taxpayers Union, Inc. v. Romer, 750 F. Supp. 1041, 1045 (D. Colo. 1990); see also Republican Party of Minn. v. White, 536 U.S. 765, 788 (2002) (holding that the First Amendment protects judges’ speech as candidates for judicial elections). This subpart proposes a constitutional limitation on the government’s expressive participation in candidate, as opposed to issue, campaigns. Here I focus on the First Amendment as the primary (but not necessarily only 200Other commentators have plausibly identified other potential constitutional sources of similar principles, including equal protection and substantive due process. See Erwin Chemerinsky, Protecting the Democratic Process: Voter Standing to Challenge Abuses of Incumbency, 49 Ohio St. L.J. 773, 778 (1988) (“[T]he equal protection clause of the fourteenth amendment is violated if the government acts to aid only the incumbent. . . . Those who support challengers have their votes diluted by abuse of incumbency in exactly the same way the malapportionment or stuffing of the ballot box lessens the effectiveness of an individual’s vote.”); Greene, supra note 108, at 38 (noting that efforts to entrench incumbents and thwart challengers “violate[] one of the two key principles of the famous footnote four [i]n Carolene Products, and should be deemed invalid” (footnote omitted) (citing United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938))). ) constitutional source of such a limitation, rooted in an understanding of that provision as informing and empowering the people’s meaningful self-governance—their free choice about whom to elect. Such an understanding of the First Amendment thus supports a principle prohibiting governmental campaign speech that directly seeks to entrench political power. 201 See Shiffrin, supra note 55, at 602 (“Citizens are entitled to a government that is neutral in the process of selecting candidates. Whether or not the concept of self-government is ‘central’ to the first amendment, it is undeniably an important first amendment value, and the integrity of the democratic process could rightly be questioned if government officially intervened in the political process to favor particular candidates. Whether or not the intervention was powerful, it would ipso facto disturb the first amendment equality principle.” (footnotes omitted)).

To be sure, identifying impermissibly self-perpetuating government speech presents significant challenges. Much, and probably most, government speech on any issue seems inextricable from a government speaker’s self-interest in reelection. For just one of many contemporary examples, consider the controversy over federal highway signs that noted that certain construction projects were funded by the federal stimulus package. 202 See Colorado Will Keep Stimulus Highway Signs, Street (Sept. 28, 2009, 10:53 AM), http://www.thestreet.com/story/10604159/1/colorado-will-keep-stimulus-highway-signs.html. Critics charged that the signs amounted to “political boosterism,” while the federal government as speaker emphasized the signs’ informational function: “taxpayers should know how stimulus dollars are being spent.” 203 Id.

Indeed, government speakers are often—and unavoidably—motivated both by public-minded and self-interested purposes. 204 See, e.g., Chafee, supra note 143, at 764–65 (“Some of these quandaries seem to me unavoidable if we are to have any sort of adequate information service. Effective presentation of any recent achievement of the government, no matter how completely it is accepted by everybody, cannot help benefiting the party and the officials who made that achievement possible. . . . In spite of the risks that men who know exactly what they want and are acquainted with the latest techniques for manipulating public opinion will dispose of large sums for their personal or departmental advantage, we may be wise to run those risks for the sake of the values of public information . . . in enabling citizens to govern themselves more intelligently.”); Shiffrin, supra note 55, at 603 (“Non-partisan aspects such as informing the populace of government policy and explaining that policy are also necessarily partisan because incumbent candidates almost invariably claim that their reelection is justified by their link to the government policy they explain and defend.”). But unless one is willing to prohibit government speech on issue campaigns as a constitutional matter—and for the reasons explained in the preceding Part, I am not—perhaps the best we can do is to root out government speech expressly geared to influencing results in current, contested candidate elections on the premise that official government speech most directly connected to candidate electioneering is generally more dangerous and less valuable than government speech on issue campaigns. Given that most government speech inevitably includes some mix of self-interested and public-spirited motivations, this may be more a difference of degree than a difference in kind. Yet I contend that the difference still remains meaningful. Thomas Emerson, for example, proposed a similar test:

The government’s right of expression does not extend to any sphere that is outside the governmental function. This might not seem to be much of a limitation; the governmental function certainly covers an extensive area. Nevertheless the principle does impose some limits. Thus the government would not be empowered to engage in expression in direct support of a particular candidate for office. It is not the function of the government to get itself reelected. 205 Emerson, supra note 143, at 699.

Along these lines, I too prefer a constitutional principle that permits transparently governmental speech on issue, but not candidate, campaigns.

Steven Shiffrin, in contrast, proposes a different principle. Identifying government expression’s potential threats to the equality interests of dissenting speakers as a greater danger than that of self-perpetuation, he suggests instead that the Constitution should be understood to limit the volume or means of the government’s expressive expenditures in both candidate and issue campaigns. 206 See Shiffrin, supra note 55, at 617. For example, he proposes that the “government should not be permitted to send mail to its citizens stating its views or adequate provision for opposing views must be made.” 207 Id.

Although I appreciate Professor Shiffrin’s thoughtful discussion of the benefits as well as the dangers of the government’s campaign speech, I find a constitutional line that turns on the amount of funds spent on the government’s expressive purposes as especially arbitrary to draw (and instead more appropriate for possible statutory limitations discussed below than for a constitutional principle). 208For a discussion of possible statutory limitations on the government’s expressive expenditures, see infra notes 210–12 and accompanying text. Indeed, such a line seems particularly problematic given that all government speech requires some expenditure of public resources, even if only in the form of government workers’ time and opportunity costs. Moreover, the instrumental value of the government’s campaign speech turns in large part on its actual ability to reach and thus inform listeners, an ability that often requires money—e.g., money to fund studies on the implications of a proposal; to publish and disseminate reports, flyers, and pamphlets; and perhaps to buy advertising. As Alan Chen observes:

To be sure, the government will have a greater chance of persuading people to agree with its position than a grass roots political organization with few resources. But so will Microsoft. Or, after the past Supreme Court term, groups such as Citizens United. Unless we start requiring something akin to the fairness doctrine whenever the government engages in speech, which would be both impracticable and unpalatable, transparency is about the best we can hope for in an imperfect doctrinal world. 209Chen, supra note 127 (endnotes omitted).

For these reasons, I propose a constitutional limitation on the government’s power to participate expressively in candidate campaigns, while rejecting constitutional limitations on the government’s expenditure of resources for expressive purposes more generally. As the next section explains, however, legislatures remain free to limit such expenditures as a statutory matter.

C. Statutory Limitations on Government Speech on Issue Campaigns

Even if the government’s speech on contested issue campaigns does not violate the Constitution, as this Article asserts, nonconstitutional solutions remain available to policymakers interested in constraining the government’s campaign speech of all types. 210As discussed supra note 100, the fact that the government is not a First Amendment rights holder itself leaves open the possibility that its speech could be constrained through statute and other policy measures.

1. Statutory Limitations on the Governments Expenditures for Expression Relating to Issue Campaigns

As discussed above, I see no constitutional bar to the government’s ability to spend public money for expressive purposes, even on matters that divide its constituents. Indeed, the government does so all the time, as the Supreme Court has recognized outside of the campaign context. 211 See supra note 134 and accompanying text. But for those who remain concerned that the government’s access to public funds for expressive purposes will permit it to drown out the speech of others, and thus monopolize the marketplace of ideas, statutory options remain available, as legislatures could cap (or entirely prohibit) expenditures for the government’s campaign speech on ballot measures. Indeed, many state statutes already prohibit any state employee or officer from using state facilities or funds in “the promotion of or opposition to a ballot proposition.” 212Wash. Rev. Code Ann. § 42.52.180(1) (West Supp. 2010); accord Colo. Rev. Stat. § 1-45-117(1)(a)(I) (2010) (prohibiting public entities from expending any public monies from any source for contributions to a campaign for elected office, or to urge electors to vote in favor of or against any ballot issue or referred measure); Okla. Stat. Ann. tit. 26, § 16-119 (West 1997) (prohibiting state officials from direct[ing] or authoriz[ing] the expenditure of any public funds under [their] care, except as specifically authorized by law, to be used either in support of, or in opposition to, any measure which is being referred to a vote of the people by means of the initiative or referendum, or which citizens of this state are attempting to have referred to a vote of the people by the initiative or referendum). Some of these laws, however, permit public officials to express their views on such measures in certain situations. See, e.g., Conn. Gen. Stat. Ann. § 9-369b(a) (West 2009) (explaining that its prohibition does not apply to a written, printed or typed summary of an officials views on a proposal or question, which is prepared for any news medium); Fla. Stat. Ann. § 106.113(3) (West Supp. 2011) (explaining that its prohibition does not preclude an elected official of the local government from expressing an opinion on any issue at any time); Wash. Rev. Code Ann. § 42.52.180(2)(a) (explaining that its prohibition does not apply to members of an elected body expressing their collective opinion on a ballot measure so long as required notice is provided and members of the public are given approximately the same opportunity to voice opposing viewpoints).

2. Statutory Limitations on the Governments Power to Deviate from Expressive Neutrality in Issue Campaigns

Although this Article concludes that the Constitution does not require the government’s expressive neutrality with respect to issue campaigns, nonconstitutional constraints remain available here as well. For example, in his groundbreaking work on government speech, Mark Yudof suggested that state legislatures consider amending the default presumption as to whether issue-related campaign speech by state or local government agencies is authorized under state statutes that define and limit the powers of such bodies. 213 See Yudof, supra note 72, at 47 (“The greatest threat of government domination and distortion of majoritarian processes emanates from executive bodies and officers. The greatest hope of restraining that power lies with the legislative branches of government. If a legislative body determines that particular government expression threatens democratic processes, the courts should not second-guess that decision.” (footnote omitted)). In other words, he proposed a statutory strategy to limit the enumerated powers of certain government actors to exclude certain expressive activities.

Indeed, legislatures not uncommonly impose more targeted statutory limits on government actors’ deviation from expressive neutrality where such deviation is considered especially dangerous. Consider, as just one example, state conflict-of-interest laws that prohibit public officials from advocating the passage or failure of matters in which they have, or may be perceived to have, a conflict of interest. 214 See Nev. Comm’n on Ethics v. Carrigan, 131 S. Ct. 2343 (2011) (rejecting a legislator’s First Amendment challenge to a Nevada law prohibiting public officials from voting or advocating with respect to matters in which the legislator has an actual or perceived conflict of interest and detailing the long history of such statutes).

* * *

Just because such nonconstitutional options are available to policymakers, however, does not mean that they are necessarily wise. Indeed, because I find government speech with respect to issue campaigns of such great instrumental value to the public, I remain uncomfortable with blanket withdrawals of the government’s expressive authority even as a policy matter. Vastly preferable, in my opinion, are policy approaches that target specific government speakers in response to evidence of abuse in specific contexts, rather than relying on broad and unsupported conclusions about the danger of government speech generally. 215 See Yudof, supra note 72, at 111–38 (urging legislative bodies to consider the dangers of government speech in deciding what types of government expression to authorize and to prohibit); Jacobs, supra note 83 (proposing a variety of statutory checks on Article II speakers). In short, policymakers should proceed with caution, preferring the scalpel to the bludgeon when choosing policy tools for addressing governmental misuse of its expressive power.

As one example of such a targeted policy response to specific abuses of the government’s expressive power, recall recent controversies over the federal government’s allegedly covert speech—executive branch speech that did not make clear its governmental source. 216 See supra notes 88–89 and accompanying text. Congress could—and, in my opinion, should—encourage greater executive branch transparency by amending the longstanding propaganda ban specifically to define such covert speech as prohibited and to provide meaningful enforcement mechanisms to punish and deter such speech.

As another example, consider policymakers’ concerns about efforts by the White House Office of Political Affairs to use government resources to fund expression to reelect incumbents. In response, a House committee recommended elimination of that office as well as amendments to the Hatch Act to create meaningful penalties for violations of that Act. 217 Comm. on Oversight & Gov’t Reform, U.S. House of Representatives, The Activities of the White House Office of Political Affairs, at ii (2008). This too exemplifies an appropriately calibrated policy response targeted to specific instances of government abuse of its expressive power.

Conclusion

This Article seeks to reexamine longstanding controversies over the government’s campaign speech by considering them in light of lessons from contemporary constitutional debates over campaign finance reform and government speech more generally. This inquiry invites important and challenging questions about both the nature of government and the nature of speech. When is government speech on contested public policy debates most valuable in facilitating participation in democratic self-governance and contributing to the dissemination of knowledge—and when, if ever, does such government expression endanger key constitutional values by drowning out others’ political speech or by entrenching incumbent political power? Can government ever simply participate in the marketplace of ideas in contested issue campaigns, or does its participation in such contests instead frustrate political competition? Is there any reason to depart from the traditional assumption that more speech is better than less when the government seeks to add its voice to the available body of expression?

This Article concludes that government speech on issue campaigns generally furthers, rather than frustrates, key constitutional values. More specifically, it finds that transparently governmental campaign speech on contested ballot and legislative measures is rarely, if ever, more harmful to First Amendment liberty or equality interests than that of any other speaker because in such contexts the government is acting as a participant in the marketplace of ideas, rather than as a regulator. Indeed, such government campaign speech is often of great value to the public. It enhances political accountability by informing voters of their governments’ priorities and preferences, provides a valuable heuristic for those who do not have the time or expertise to evaluate the competing arguments for themselves, and sometimes provides a counter to expression from powerful, private sources that often operate in nontransparent ways.

To be sure, however, the government’s campaign speech should not be entirely free from constraint. First, the government’s campaign speech on contested ballot or legislative measures should be considered consistent with constitutional values only when that speech is transparently governmental in origin—when the public can clearly identify the message’s governmental origins and thus hold the government politically accountable for its views. Second, government entities’ campaign speech endorsing or opposing specific candidates raises distinct constitutional threats to First Amendment interests in constraining incumbents’ self-perpetuation and entrenchment. Finally, statutory and other nonconstitutional limits on the government’s campaign speech remain available. The substantial instrumental value of the government’s issue-related campaign speech to the public, however, should counsel policymakers to proceed with caution, focusing their efforts to target identified abuses of the government’s expressive power, rather than relying on broader and unsupported conclusions about the danger of government campaign speech generally.

Footnotes

Associate Dean for Academic Affairs and Associate Professor, University of Colorado Law School. Thanks for very thoughtful comments to Steven Andre, Rachel Arnow-Richman, Joseph Blocher, Richard Boldt, Alan Chen, Danielle Keats Citron, Caroline Mala Corbin, Roberto Corrada, Aya Gruber, Melissa Hart, Deborah Hellman, Jessie Hill, Leslie Gielow Jacobs, Sarah Krakoff, Scott Moss, Ofer Raban, Nantiya Ruan, Jana Singer, Catherine Smith, and the participants at University of Colorado and University of Maryland School of Law faculty workshops as well as at symposia at Case Western Reserve University School of Law, Loyola University of Chicago School of Law, and the University of Connecticut School of Law. Thanks also to Jordan Bunch, Evan Branigan, David Cline, Caitlin McHugh, Bridget Newman, and Chelsea Strautman Lassen for excellent research assistance, to Doris Cheung for endlessly patient technical assistance, to the Emory Law Journal staff for their outstanding editorial work, and to Dean David Getches and the University of Colorado Law School for financial support for this project.

1 See Frederick Schauer & Richard H. Pildes, Electoral Exceptionalism and the First Amendment, 77 Tex. L. Rev. 1803 (1999) (exploring the possibility of First Amendment doctrine unique to expression concerning political campaigns and elections).

2 See, e.g., Citizens United v. FEC, 130 S. Ct. 876, 903–04 (2010); Davis v. FEC, 554 U.S. 724, 741 (2008); McConnell v. FEC, 540 U.S. 93, 205 (2003), overruled in part by Citizens United, 130 S. Ct. 876; Austin v. Mich. State Chamber of Commerce, 494 U.S. 652, 658–59 (1990), overruled by Citizens United, 130 S. Ct. 876; Buckley v. Valeo, 424 U.S. 1, 48 (1976) (per curiam).

3 See, e.g., Citizens United, 130 S. Ct. at 893; Davis, 554 U.S. at 736; McConnell, 540 U.S. at 134–40; Austin, 494 U.S. at 656; Buckley, 424 U.S. at 15.

4 See Pleasant Grove City v. Summum, 129 S. Ct. 1125, 1139 (2009) (Stevens, J., concurring).

5For a more extensive discussion of the Supreme Court’s brief history with government speech, see Helen Norton & Danielle Keats Citron, Government Speech 2.0, 87 Denv. U. L. Rev. 899, 904–16 (2010).

6 See, e.g., Summum, 129 S. Ct. at 1129–30.

7 E.g., Turner Broad. Sys., Inc. v. FCC, 512 U.S. 622, 641 (1994) (recounting the First Amendment’s bar on government’s viewpoint-based discrimination against private speech); R.A.V. v. City of St. Paul, 505 U.S. 377, 382 (1992) (“The First Amendment generally prevents government from proscribing speech, or even expressive conduct, because of disapproval of the ideas expressed. Content-based regulations are presumptively invalid.” (citations omitted)); Police Dep’t of Chi. v. Mosley, 408 U.S. 92, 95 (1972) (“[A]bove all else, the First Amendment means that government has no power to restrict expression because of its message, its ideas, its subject matter, or its content.”).

8Under the Supreme Court’s government speech doctrine, the government’s own expression is insulated from Free Speech Clause challenges by plaintiffs who seek to alter or join that expression. See Summum, 129 S. Ct. at 1131 (“If [public entities] were engaging in their own expressive conduct, then the Free Speech Clause has no application. The Free Speech Clause restricts government regulation of private speech; it does not regulate government speech. A government entity has the right to ‘speak for itself.’ . . . Indeed, it is not easy to imagine how government could function if it lacked this freedom.” (citations omitted) (quoting Bd. of Regents v. Southworth, 529 U.S. 217, 229 (2000))); Johanns v. Livestock Mktg. Ass’n, 544 U.S. 550, 553 (2005) (explaining that the government’s own speech is “exempt” from Free Speech Clause scrutiny); see also Southworth, 529 U.S. at 235 (“When the government speaks, for instance to promote its own policies or to advance a particular idea, it is, in the end, accountable to the electorate and the political process for its advocacy. If the citizenry objects, newly elected officials later could espouse some different or contrary position.”).

9A referendum generally enables citizens to enact or reject statutes or constitutional amendments proposed by a legislature, while an initiative enables citizens to draft proposals themselves and submit them directly to the populace for a vote. Ethan J. Leib, Can Direct Democracy Be Made Deliberative?, 54 Buff. L. Rev. 903, 904 (2006).

10I acknowledge that many use the term “campaign speech” more narrowly to refer only to speech on matters subject to vote directly by the people themselvesballot measures and candidate elections. As discussed in more detail below, however, my definition of government speech on “issue campaigns” also includes the government’s speech on pending legislation subject to vote by the people’s representatives, because a number of critics urge that such efforts pose constitutional threats very similar to the threats they perceive posed by the government’s speech on ballot measures to be decided directly by the people. See infra notes 41–46, 60–67, and accompanying text.

11 See infra notes 78–79 and accompanying text.

12 See infra notes 33–34 and accompanying text.

13 See infra notes 24–28 and accompanying text.

14Page v. Lexington Cnty. Sch. Dist. One, 531 F.3d 275, 277–80, 287–88 (4th Cir. 2008).

15Kidwell v. City of Union, 462 F.3d 620, 621–22, 626 (6th Cir. 2006).

16I am indebted to Joseph Blocher for his observation that this is an analysis of the possibility of an exception to an exception: whether government campaign speech should be excepted (by subjecting it to Free Speech Clause scrutiny) from government speech doctrine more generally, which itself is an exception to traditional First Amendment doctrine (by exempting the government’s own expression from Free Speech Clause scrutiny).

17130 S. Ct. 876, 913 (2010).

18As discussed in greater detail in Part III, this Article distinguishes government speech expressing a position on contested issue campaigns from government entities’ speech endorsing or opposing a specific candidate or party.

19 See Pleasant Grove City v. Summum, 129 S. Ct. 1125, 1131–32 (2009) (“This does not mean that there are no restraints on government speech. . . . The involvement of public officials in advocacy may be limited by law, regulation, or practice.”); id. at 1139 (Stevens, J., concurring) (“Finally, recognizing permanent displays on public property as government speech will not give the government free license to communicate offensive or partisan messages. For even if the Free Speech Clause neither restricts nor protects government speech, government speakers are bound by the Constitution’s other proscriptions, including those supplied by the Establishment and Equal Protection Clauses. Together with the checks imposed by our democratic processes, these constitutional safeguards ensure that the effect of today’s decision will be limited.”).

20 See, e.g., Rees v. Carlisle, 153 P.3d 1131, 1137–39 (Haw. 2007) (concluding that a prosecutor’s use of public resources to urge voters to support an amendment to Hawaii’s constitution violates state law interpreted to permit a prosecutor to offer comments but not to engage in express advocacy); Citizens to Protect Pub. Funds v. Bd. of Educ., 98 A.2d 673, 676–77 (N.J. 1953) (determining that a school board’s use of public funds to print a booklet urging voters to “Vote Yes” on a bond referendum was not within the power implied to the board by its express power to operate the schools); Porter v. Tiffany, 502 P.2d 1385 (Or. Ct. App. 1972) (finding no statutory authorization for a city water and electric board’s expenditure of funds for materials advocating support of election measures regarding nuclear power program participation and delay of construction of a nuclear power plant).

21Stanson v. Mott, 551 P.2d 1, 9 (Cal. 1976) (in bank); accord Vargas v. City of Salinas, 205 P.3d 207, 226 (Cal. 2009) (discussing potential constitutional problems that may arise from the government’s use of public funds for speech in support of issue campaigns).

22 See Mountain States Legal Found. v. Denver Sch. Dist. # 1, 459 F. Supp. 357, 360 (D. Colo. 1978) (“[A] grant of express authority for a partisan use of public funds in an election of this type would violate the First Amendment to the United States Constitution . . . .”).

23 See id. at 361 (“When residents within a state seek to participate in this process by proposing an amendment to the state constitution, the expenditure of public funds in opposition to that effort violates a basic precept of this nation’s democratic process. Indeed, it would seem so contrary to the root philosophy of a republican form of government as might cause this Court to resort to [Article IV’s] guaranty clause . . . .”). Article IV, Section 4 of the Constitution—commonly known as the Guarantee Clause—provides that “[t]he United States shall guarantee to every State in this Union a Republican Form of Government.” U.S. Const. art. IV, § 4. Although the Supreme Court has long held that Guarantee Clause claims are not justiciable, e.g., Luther v. Borden, 48 U.S. (7 How.) 1, 42–43 (1849), a number of judges reviewing challenges to government campaign speech on contested ballot measures have suggested that this holding should be revisited, see, e.g., Kidwell v. City of Union, 462 F.3d 620, 635–36 & n.5 (6th Cir. 2006) (Martin, J., dissenting) (noting that, because “ordinary democratic controls are insufficient as a remedy in situations where governmental influence threatens to undermine the independent political process” and “[g]overnmental advocacy and campaign expenditures could arguably threaten to undermine free and fair elections, could be coercive, and could reasonably undermine the reliability and outcome of elections where the government acts as a participant,” “[p]erhaps it is time for the Supreme Court to reconsider its Guarantee Clause jurisprudence”).

24For other early examples of this trend, see Mines v. Del Valle, 257 P. 530, 536–37 (Cal. 1927) (in bank) (holding that the express municipal power to operate a public utility did not imply the power to appropriate funding to urge voters to approve a bond referendum to extend the utility), overruled by Stanson, 551 P.2d 1; and Elsenau v. City of Chicago, 165 N.E. 129, 131 (Ill. 1929) (invalidating as an unauthorized municipal function a city’s expenditures for advertisements that “did not purport to be an impartial statement of facts for the information of the voters, but that . . . was an attempt, partisan in its nature, to induce the voters to act favorably upon the bond issues submitted at the election”). Some early courts bucked this trend, emphasizing the instrumental value of the government’s expression on ballot campaigns. E.g., City Affairs Comm. v. Bd. of Comm’rs, 41 A.2d 798, 800 (N.J. Sup. Ct. 1945) (“We think municipalities may, within their discretion and in good faith, present their views for or against proposed legislation or referendum to the people of questions which in their judgment would adversely affect the interests of their residents.”), aff’d, 46 A.2d 425 (N.J. 1946).

25 Citizens to Protect Pub. Funds, 98 A.2d 673.

26 Id. at 674.

27The court determined the issue to be moot because the election had already occurred. Id. at 676 (dictum) (“Plainly, then, any issues as to both the booklet and the radio broadcast are moot . . . . Nevertheless, the importance of the question makes appropriate our comment upon the actions taken.”).

28 Id. at 677–78 (dictum).

29 See, e.g., Palm Beach Cnty. v. Hudspeth, 540 So. 2d 147, 154 (Fla. Dist. Ct. App. 1989) (“If government, with its relatively vast financial resources, access to the media and technical know-how, undertakes a campaign to favor or oppose a measure placed on the ballot, then by so doing government undercuts the very fabric which the constitution weaves to prevent government from stifling the voice of the people.”); Anderson v. City of Boston, 380 N.E.2d 628, 639 (Mass. 1978) (“Fairness and the appearance of fairness are assured by a prohibition against using public tax revenues to advocate a position which certain taxpayers oppose.”); Stern v. Kramarsky, 375 N.Y.S.2d 235, 237 (Sup. Ct. Spec. Term 1975) (“It should be noted that by lending their support to the campaign underway for the passage of the Equal Rights Amendment, defendants not only provide certain promotional and advertising assistance, but they endow that campaign with all of the prestige and influence naturally arising from any endorsement of a governmental authority.”).

30 Hudspeth, 540 So. 2d at 154.

31 See, e.g., Campbell v. Joint Dist. 28-J, 704 F.2d 501 (10th Cir. 1983) (concluding that a school district’s expenditures urging the approval of a financing proposal were unauthorized under state statute); Choice-in-Educ. League v. L.A. Unified Sch. Dist., 21 Cal. Rptr. 2d 303 (Ct. App. 1993) (reversing the trial court’s grant of a preliminary injunction enjoining school board members from expressing opposition to a pending voucher proposal); Coffman v. Colo. Common Cause, 102 P.3d 999 (Colo. 2004) (en banc) (concluding that the secretary of state’s press releases opposing a statewide ballot initiative concerning school funding violated a state statute prohibiting public agencies from using more than fifty dollars in public monies to support or oppose ballot measures).

32 See, e.g., Stanson v. Mott, 551 P.2d 1 (Cal. 1976) (in bank) (considering a challenge to a state agency’s expenditures to urge support for increased parks funding); Ameritel Inns, Inc. v. Greater Boise Auditorium Dist., 119 P.3d 624 (Idaho 2005) (concluding that a local government entity did not have statutory authorization to use public funds to support a ballot measure to approve the issuance of bonds to fund construction of the district’s convention center); Carter v. City of Las Cruces, 915 P.2d 336 (N.M. Ct. App. 1996) (considering a taxpayer’s challenge to city officials’ expenditures of public funds to support a ballot measure to empower the city to acquire a private electric utility).

33 Stern, 375 N.Y.S.2d at 239–40. For other examples of challenges to government speech that took sides on ballot measures outside of the government-financing context, see Rees v. Carlisle, 153 P.3d 1131, 1141 (Haw. 2007); and King County Council v. Public Disclosure Commission, 611 P.2d 1227 (Wash. 1980) (en banc).

34 Stern, 375 N.Y.S.2d at 239.

35 See, e.g., Stanson, 551 P.2d at 9 (“A fundamental precept of this nation’s democratic electoral process is that the government may not ‘take sides’ in election contests or bestow an unfair advantage on one of several competing factions. A principal danger feared by our country’s founders lay in the possibility that the holders of governmental authority would use official power improperly to perpetuate themselves, or their allies, in office; the selective use of public funds in election campaigns, of course, raises the specter of just such an improper distortion of the democratic electoral process.” (citations omitted)).

36 Choice-in-Educ. League, 21 Cal. Rptr. 2d 303. Indeed, governments frequently express themselves on contested issues through resolutions and proclamations that require little, if anything, in the way of public expenditures. See Jacob E. Gersen & Eric A. Posner, Soft Law: Lessons from Congressional Practice, 61 Stan. L. Rev. 573, 584–86 (2008) (discussing congressional use of resolutions and other hortatory means to make declarations of policy).

37 King Cnty. Council, 611 P.2d at 1231. See generally Bruce W. Blakely, Comment, Public Utility Bill Inserts, Political Speech, and the First Amendment: A Constitutionally Mandated Right to Reply, 70 Calif. L. Rev. 1221, 1254 (1982) (“[I]t is the proper role of government to engage in informational speech. The state, however, may not employ a forum to engage in partisan expression. Government should not seek to influence initiatives or campaigns; nor are government editorials acceptable.”).

38 See Stanson, 551 P.2d at 9–10.

39 See id. (“[W]hile public agency ‘lobbying’ efforts undeniably involve the use of public funds to promote causes which some members of the public may not support, one of the primary functions of elected and appointed executive officials is, of course, to devise legislative proposals to attempt to implement the current administration’s policies. Since the legislative process contemplates that interested parties will attend legislative hearings to explain the potential benefits or detriments of proposed legislation, public agency lobbying, within the limits authorized by statute, in no way undermines or distorts the legislative process. By contrast, the use of the public treasury to mount an election campaign which attempts to influence the resolution of issues which our Constitution leave to the ‘free election’ of the people does present a serious threat to the integrity of the electoral process.” (citations omitted)); Note, The Constitutionality of Municipal Advocacy in Statewide Referendum Campaigns, 93 Harv. L. Rev. 535, 556–57 (1980) (“Cities are usually free to lobby before the legislature to prevent a measure from being put before the statewide electorate in referendum form. But once the state legislature decides to entrust the final legislative decision to the popular electorate, it explicitly removes the decision from the hands of state or municipal officeholders. Permitting those officials to use public funds to attempt to influence the outcome of that decision would partially return them to a role from which they have been excluded by constitutional design. Municipal governments should thus refrain from establishing an official political viewpoint during the time that the popular electorate, rather than its elected representatives, makes law.” (footnotes omitted)).

40 See, e.g., Ala. Libertarian Party v. City of Birmingham, 694 F. Supp. 814, 818 (N.D. Ala. 1988) (“While defendants might be forbidden to spend funds to support candidates, oppose initiative proposals, etc., they are not forbidden to publicize and seek public support for their own governmental proposals.”).

41Robert D. Kamenshine, The First Amendment’s Implied Political Establishment Clause, 67 Calif. L. Rev. 1104, 1104 (1979).

42 Id. at 1116; accord id. at 1129, 1137 (“Under the prohibition of an implied political establishment clause, regulations that permit the dissemination of information to improve an agency’s image or to enhance support for its policies would be impermissible—both as to purpose and effect. . . . The political establishment clause would require balanced curricular and textbook treatment of the various viewpoints on race relations and women’s rights. It is possible that in some instances the government would design its textbooks and curriculum to present only a positive image of a particular group. Courts would have to prohibit such efforts, however, as a political establishment.” (footnote omitted)).

43Edward H. Ziegler, Jr., Government Speech and the Constitution: The Limits of Official Partisanship, 21 B.C. L. Rev. 578, 579–80 (1980) (footnote omitted). Professor Ziegler defined prohibited “official partisanship” for these purposes very broadly to include government speech on pending referenda, constitutional amendments, and legislative proposals. See id. at 581 n.12.

44 Id. at 583.

45Indeed, the views of Professors Kamenshine and Ziegler may reflect not only a more limited view of government’s appropriate expressive functions but also an assumption that the coercive effect of the government’s voice is especially great because people generally trust and defer to their government. That assumption may be less valid today in light of evidence that voters are significantly more suspicious of the government than they once were. See, e.g., Margaret Levi & Laura Stoker, Political Trust and Trustworthiness, 3 Ann. Rev. Pol. Sci. 475, 476–85 (2000) (surveying social science literature documenting declines in political trust).

46Steven D. Smith, Why Is Government Speech Problematic? The Unnecessary Problem, the Unnoticed Problem, and the Big Problem, 87 Denv. U. L. Rev. 945, 968 (2010).

47City of Boston v. Anderson, 439 U.S. 1389 (Brennan, Circuit Justice 1978).

48 See supra notes 25–28 and accompanying text.

49 Anderson, 439 U.S. 1389.

50435 U.S. 765 (1978).

51 Anderson, 439 U.S. at 1390.

52 Id.

53Miller v. Cal. Comm’n on the Status of Women, 198 Cal. Rptr. 877, 882 (Ct. App. 1984) (footnote omitted).

54 See, e.g., City Affairs Comm. v. Bd. of Comm’rs, 41 A.2d 798, 800 (N.J. Sup. Ct. 1945) (“We think municipalities may, within their discretion and in good faith, present their views for or against proposed legislation or referendum to the people of questions which in their judgment would adversely affect the interests of their residents. To accomplish this purpose we think they may incur expenditures by the publication of pamphlets, circulars, newspaper advertisements or radio addresses and that to do so is a proper governmental function.”), aff’d, 46 A.2d 245 (N.J. 1946).

55Ala. Libertarian Party v. City of Birmingham, 694 F. Supp. 814, 817, 819 n.10 (N.D. Ala. 1988); accord Steven Shiffrin, Government Speech, 27 UCLA L. Rev. 565, 613 (1980) (“[I]t is arguably the function, and perhaps the duty, of public officials to speak out on all issues of the day . . . . Governments, then, can justify subsidizing the speech of public officials, not to reelect them or others, but because there is a substantial interest in hearing what they have to say.”).

56462 F.3d 620 (6th Cir. 2006).

57 Id. at 624.

58 Id. at 635 (Martin, J., dissenting).

59 See, e.g., Stanson v. Mott, 551 P.2d 1, 9 (Cal. 1976) (in bank) (noting that executive officials necessarily create legislative proposals to advance an administration’s policies).

60For a discussion of how voters considering ballot measures should be viewed as akin to legislators considering legislative measures, see Doe v. Reed, 130 S. Ct. 2811, 2833 (2010) (Scalia, J., concurring in the judgment).

61As explained supra note 10, this Article’s definition of “governmental campaign speech” includes government entities’ speech on legislative campaigns as well as ballot campaigns because some critics suggest that the government’s persuasive efforts with respect to matters subject to vote by legislators pose constitutional threats very similar to the threats they perceive posed by the government’s speech on matters subject to vote directly by the people themselves.

62Bonner-Lyons v. Sch. Comm., 480 F.2d 442 (1st Cir. 1973).

63Page v. Lexington Cnty. Sch. Dist. One, 531 F.3d 275 (4th Cir. 2008).

64 Id. at 287.

65 Id. at 288.

66 See infra notes 80–82 and accompanying text.

67This too is by no means a new debate over the scope of the government’s appropriate expressive role. Indeed, Jeffrey Tulis has exhaustively examined historical shifts in views about whether and when the President should speak directly to the public (rather than only to Congress) in support of or opposition to pending legislative matters. Jeffrey K. Tulis, The Rhetorical Presidency (1987). After documenting the rarity with which eighteenth- and nineteenth-century presidents spoke directly to the people on policy matters, id. at 61–87, Professor Tulis then describes the presidencies of Theodore Roosevelt and Woodrow Wilson as developing appeals to popular rhetoric as “a principal tool of presidential governance” for the first time, id. at 4. Of course, presidents now regularly appeal directly to the people at large, rather than to Congress, in soliciting support for legislative and other policy initiatives. I thank Jonathan Estin for directing me to this very helpful resource.

68 See David W. Guth, Black, White, and Shades of Gray: The Sixty-Year Debate over Propaganda Versus Public Diplomacy, 14 J. Promotion Mgmt. 309 (2008).

69 See id. at 310 (noting that many use the term propaganda “as an umbrella covering all forms of persuasive communication, including advertising and public relations”). Although my focus here is on controversies over the government’s domestic “propaganda,” such definitional challenges play out in the international arena as well. See, e.g., Richard B. Collins, Propaganda for War and Transparency, 87 Denv. U. L. Rev. 819, 821–28 (2010) (discussing varying definitions of “war propaganda” prohibited by international law and prosecuted in the Tokyo and Nuremberg war crimes trials).

70 See Guth, supra note 68, at 310 (noting that some distinguish between “revealed propaganda, messages that are overt in their effort to persuade, such as those in conventional advertising,” and “concealed propaganda, such as publicity generated from the distribution of news releases”).

71 E.g., Kelly Sarabyn, Prescribing Orthodoxy, 8 Cardozo Pub. L. Pol’y & Ethics J. 367, 369 (2010).

72 See Frederick Schauer, Is Government Speech a Problem?, 35 Stan. L. Rev. 373, 380 (1983) (reviewing Mark G. Yudof, When Government Speaks: Politics, Law, and Government Expression in America (1983)) (“Now it is true that propaganda is hardly a term of precise definition, but we usually use it when talking about totalitarian regimes whose propaganda is coupled with a rigid, if not absolute, suppression of any voice other than that of the state. The issues raised when the state is the only voice are significantly different from those raised when the state is one of many voices, especially when the other voices are unrestrained both in law and in fact.”).

73 Garth S. Jowett & Victoria O’Donnell, Propaganda & Persuasion 17–20 (5th ed. 2012). Objections to government’s “propaganda,” however defined, may turn on the government’s target audience. For example, Guth describes a 1948 statute that “authorized the federal government to engage in a program of dissemination of truthful information to international audiences, while . . . prohibit[ing] the government from transmitting the same information to domestic publics.” Guth, supra note 68, at 312.

74In 1942, for instance, Congress amended the Foreign Agents Registration Act to require agents of foreign principals who distribute “political propaganda” to ensure the material’s conspicuous labeling as such. Pub. L. No. 77-532, sec. 1, § 4(b), 56 Stat. 248, 255 (1942) (amended 1995). In enacting this requirement, Congress was spurred by concerns about the circulation of anonymous anti-American propaganda from Nazi sources in the years before and during World War II. See H.R. Rep. No. 75-1381, at 2 (1937) (“This required registration will publicize the nature of subversive or other similar activities of such foreign propagandists, so that the American people may know those who are engaged in this country by foreign agencies to spread doctrines alien to our democratic form of government, or propaganda for the purpose of influencing American public opinion on a political question. . . . We believe that the spotlight of pitiless publicity will serve as a deterrent to the spread of pernicious propaganda. We feel that our people are entitled to know the sources of any such efforts . . . .”). Decades later, a plaintiff who sought to exhibit Canadian films on acid rain and nuclear war challenged the statute on First Amendment grounds, arguing that the Act’s labeling requirements with respect to “propaganda” included a pejorative connotation that deterred him from exhibiting the films under such terms. Meese v. Keene, 481 U.S. 465, 467–68 (1987). After discussing the wide variety of available definitions of the term “propaganda,” id. at 477–78, the Supreme Court ultimately—and controversially—concluded that the labeling requirements did not burden speech (and thus posed no First Amendment problem) because Congress intended no pejorative connotation when using the term in the Act, id. at 484–85. Congress amended the statute in 1995 to replace the term “political propaganda” with “informational materials.” Lobbying Disclosure Act of 1995, Pub. L. No. 104-65, sec. 9, § 614(a)–(c), 109 Stat. 691, 700.

75Jodie Morse, Note, Managing the News: The History and Constitutionality of the Government Spin Machine, 81 N.Y.U. L. Rev. 843, 853 (2006) (quoting 97 Cong. Rec. 4098 (1951)).

76 See, e.g., Consolidated Appropriations Act, 2004, Pub. L. No. 108-199, 118 Stat. 3.

77Powerful, private interests like the American Medical Association were unhappy as well. See Harry M. Marks, Revisiting “The Origins of Compulsory Drug Prescriptions,85 Am. J. Pub. Health 109, 112–13 (1995) (describing the American Medical Association’s hostility toward Oscar Ewing and his efforts).

78 See, e.g., Oscar R. Ewing, The Nation’s Health: A Report to the President 12–13 (1948) (“In the face of these and the other proofs given in this report that further rapid improvement in national health can be achieved only by concerted effort, and that the need for increased Federal action is imperative, I have reached the considered conclusion that more extensive and efficient Nation-wide planning is the only effective way to accomplish a significant betterment in national health.”); id. at 18 (“I am compelled to urge, as strongly as I know how, that the Congress enact, as President Truman has recommended, a system of Government prepayment health insurance in the terms in which it has been mapped out in [the report].”).

79 See Legislative Activities of Executive Agencies: Hearings Before the H. Select Comm. on Lobbying Activities, 81st Cong. 338–39 (1950) (statements of Charles A. Halleck & Clarence J. Brown, Members, H. Select Comm. on Lobbying Activities) (disparaging Ewing’s efforts in “sell[ing] that particular piece of legislation” as “mak[ing] speeches and spread[ing] the philosophy of socialism and Government dictatorship”).

80 See id. at 341 (statement of Clarence J. Brown, Member, H. Select Comm. on Lobbying Activities) (“[Y]ou have certain administrative responsibilities, but the Constitution does not give you any authority to attempt to make law. It restricts that authority to no one else in the world but to the Congress of the United States. The President does not have any authority to make law, nor should he have. He sometimes tries to exercise it, however.”).

81 See 97 Cong. Rec. 4098 (1951) (statement of Rep. Sidney R. Yates) (“Would not the effect of the gentleman’s amendment in using the word ‘propaganda’ jeopardize publication by the Children’s Bureau of pamphlets pertaining to the training and growth of children?”); id. at 4099 (statement of Rep. John E. Fogarty) (“We do not even know what the gentleman calls propaganda. We do not know what he calls the right type of publicity or the wrong type of publicity. That is the fault I find with this amendment.”); id. at 4100 (statement of Rep. John E. Fogarty) (“Here you are limiting the amount of publicity and propaganda which may be issued by any agency of government in this bill and yet you do not define in the amendment what propaganda is or what publicity is.”).

82 See id. at 4099 (statement of Rep. George Meader) (“[The propaganda prohibition] is necessary to strengthen the Congress in the interest of formulating national policy by the people themselves. It is a corollary to that principle that public opinion ought not to be subjected to influence and direction by the executive agencies, the administrative branch of the Government, in the manner that it is today. In a democracy, where public opinion rules in the long run, the media of communication: the press, the radio, television, and the printed word, are very potent weapons in the control of the affairs of this country.”).

83For discussion of the inevitability of executive speech on such matters, see, e.g., Leslie Gielow Jacobs, Bush, Obama and Beyond: Observations on the Prospect of Fact Checking Executive Department Threat Claims Before the Use of Force, 26 Const. Comment. 433 (2010); and Vasan Kesavan & J. Gregory Sidak, The Legislator-in-Chief, 44 Wm. & Mary L. Rev. 1 (2002).

84As the agency charged generally with monitoring public expenditures, the GAO is the administrative body nominally responsible for attending to the “propaganda rider.” Morse, supra note 75, at 859; accord 1 Office of the Gen. Counsel, U.S. Gen. Accounting Office, Principles of Federal Appropriations Law 4-198 (3d ed. 2004). The GAO, however, has only a purely advisory function and no direct enforcement power. Morse, supra note 75, at 859. At most, the GAO makes findings and can refer its determinations to Congress or other governmental bodies for further investigation. Id.

85 U.S. Gen. Accounting Office, B-302504, Medicare Prescription Drug, Improvement, and Modernization Act of 2003—Use of Appropriated Funds for Flyer and Print and Television Advertisements 7 (2004) (citations omitted). Along these lines, the GAO has declined to characterize the government’s expressive activities as prohibited “propaganda” as long as they “are reasonably related to the agency’s duty to inform the public of agency actions, programs, and policies, or justify and rebut attacks upon its policies.” U.S. Gov’t Accountability Office, B-316443, Department of Defense—Retired Military Officers as Media Analysts 8 (2009); accord U.S. Gen. Accounting Office, supra, at 13 (“To restrict all materials that have some political content or express support for an Administration’s policies would significantly curtail the recognized and legitimate exercise of the Administration’s authority to inform the public of its policies, to justify its policies and to rebut attacks on its policies. It is important for the public to understand the philosophical underpinnings of the policies advanced by elected officials and their staff in order for the public to evaluate and form opinions on those policies.”).

86 U.S. Gov’t Accountability Office, B-319075, Department of Health And Human Services—Use of Appropriated Funds for “HealthReform.gov” Web Site and “State Your Support” Web Page 7–8 (2010).

87 See U.S. Gov’t Accountability Office, supra note 85, at 2 (“Application of the prohibition is necessarily balanced against an agency’s responsibility to inform the public about its activities and programs, explain its policies and priorities, and defend its policies, priorities, and point of view.”).

88 See, e.g., U.S. Gov’t Accountability Office, B-305368, Department of Education—Contract to Obtain Services of Armstrong Williams 14 (2005) (“Every agency has a legitimate interest in the ‘dissemination to the general public . . . of information reasonably necessary to the proper administration of the laws’ for which the agency is responsible. However, while we agree that the Department should disseminate information to the public on the NCLB Act, it must disclose its role.” (citations omitted) (quoting U.S. Gen. Accounting Office, B-106139, Appropriations—Limitations—Publicity and Propaganda Prohibition—Labor–Federal Security Appropriation Act, 1952 (1952))).

89 See id. at 1.

90David Barstow, Behind TV Analysts, Pentagon’s Hidden Hand, N.Y. Times (Apr. 20, 2008), http://www.nytimes.com/2008/04/20/us/20generals.html?_r=3&scp=1&sq=hand+of+pentagon&st=nyt&oref=slogin&oref=slogin. The GAO’s 2009 report found that the program did not violate the propaganda ban. See U.S. Gov’t Accountability Office, supra note 85, at 2 (“Clearly, DOD attempted to favorably influence public opinion with respect to the Administration’s war policies in Iraq and Afghanistan through the RMOs [Retired Military Officers]. However, as we discuss below, and based on the record before us in this case, we conclude that DOD’s public affairs outreach program to RMOs did not violate the prohibition. We found no evidence that DOD attempted to conceal from the public its outreach to RMOs or its role in providing RMOs with information, materials, access to department officials, travel, and luncheons. Moreover, we found no evidence that DOD contracted with or paid RMOs for positive commentary or analysis.”).

91For example, legislation introduced in the Senate in 2005 would have prohibited, inter alia, any executive branch expression “designed to support or defeat legislation pending before Congress or any State legislature” other than testimony at legislative hearings. S. 266, 109th Cong. §§ 3–4 (2005).

92 See, e.g., S. 266.

93Such a claim—that some speech should be restricted to facilitate free speech values—may seem novel from a First Amendment standpoint but, of course, is not without some precedent. For example, the debate over the First Amendment implications of campaign finance regulation in part turns on a debate over whether the First Amendment protects speech (in which case, government restrictions on corporations’ campaign speech are suspect) or certain speakers (in which case, corporate campaign speech might be permissibly regulated to prevent drowning out the speech of protected speakers). See Kathleen M. Sullivan, Two Concepts of Freedom of Speech, 124 Harv. L. Rev. 143, 155, 158 (2010) (discussing implications of the debate over whether the First Amendment protects speech or speakers).

94Sullivan, supra note 93.

95 Id. at 145.

96 Id. at 144.

97 Id. at 145.

98 See supra notes 33–34 and accompanying text.

99 See supra notes 25–29 and accompanying text.

100Note that this Article does not maintain that the government has a First Amendment right to speak, but instead that government speech does not violate the Free Speech Clause because its expression furthers, rather than frustrates, key First Amendment values. As discussed in more detail below, that the government is not a First Amendment rights holder itself leaves open the possibility that its speech could be constrained through statute or other nonconstitutional means. See infra notes 209–12 and accompanying text.

101 See infra notes 139–41 and accompanying text.

102Ekow N. Yankah, The Force of Law: The Role of Coercion in Legal Norms, 42 U. Rich. L. Rev. 1195, 1216 (2008); accord id. at 1199 (“Asserting that the law is coercive isolates a particular way it treats people; namely, the law imposes a non-voluntary normative system on people.”); id. at 1218 (“Coercion is normally claimed when one has been forced by another to act, or refrain from acting, against their will. Coercive pressure can overcome one’s will and make a particular course of action unreasonably costly. For example, where coercive pressure is applied to Bob, that pressure would render one or more of his options unreasonably costly. Coercive pressure in this respect makes a particular option unreasonable but not necessarily impossible.” (footnote omitted)).

103The First Amendment’s primary purposes are most often identified to include protecting individual interests in autonomy and self-expression, furthering citizen participation in democratic self-government, and contributing to the discovery of truth and the development of knowledge. See, e.g., Ronald A. Cass, The Perils of Positive Thinking: Constitutional Interpretation and Negative First Amendment Theory, 34 UCLA L. Rev. 1405, 1411 (1987); Thomas I. Emerson, First Amendment Doctrine and the Burger Court, 68 Calif. L. Rev. 422, 423 (1980).

104 See supra note 7.

105 See supra notes 30–34 and accompanying text; see also Brian C. Castello, Note, The Voice of Government as an Abridgement of First Amendment Rights of Speakers: Rethinking Meese v. Keene, 1989 Duke L.J. 654, 676 (“When the public perceives the government as having expert knowledge of an issue or unrivaled control over a matter, the government communication may command even greater respect and faith. Respect for the government, reliance on the accuracy of its statements, and fear of challenging its authoritative word might cause citizens to withdraw, to a certain extent, from their duties of self-governance. As a result, government would coerce, rather than earn, the majority’s consent.” (footnotes omitted)).

106 See Martin H. Redish & Kevin Finnerty, What Did You Learn in School Today? Free Speech, Values Inculcation, and the Democratic–Educational Paradox, 88 Cornell L. Rev. 62, 67 (2002) (“[T]he notion of thought control is inconsistent with the concepts of free thought and mental autonomy that render the exercise of the free expression right meaningful.”); id. at 75 (“[A] foundational strategy of any sophisticated totalitarian society is to control the minds of its citizens, thereby destroying individual mental autonomy.”); Sarabyn, supra note 71, at 369 (“[Prescribing orthodoxy] fails to treat citizens as free and autonomous, and essentially appoints the state itself as the supreme judge of ideological truth, reproducing its ideological views in the citizenry by deploying coercion.”).

107 See Steven Lukes, Power: A Radical View 21 (2d ed. 2005) (describing power as including “coercion, influence, authority, force and manipulation,” and distinguishing “coercion,” which means that “A secures B’s compliance by the threat of deprivation where there is ‘a conflict over values or course of action between A and B’” from “influence,” which “exists where A, ‘without resorting to either a tacit or an overt threat of severe deprivation, causes [B] to change his course of action’” (alteration in original) (quoting Peter Bachrach & Morton S. Baratz, Power and Poverty: Theory and Practice 24, 30 (1970))); see also Schauer, supra note 72, at 384–85 (“If . . . we view the first amendment as primarily or exclusively protecting individual self-expression, self-realization, self-fulfillment, or something of that sort, then it is hard to see how government speech could be a first amendment problem. When we focus on communicators or listeners not instrumentally but as the end result of our concern, the fact that the government may be speaking is of little moment.” (footnote omitted)).

108 See Abner S. Greene, Government of the Good, 53 Vand. L. Rev. 1, 45 (2000) (“No choice is made exogenously to the various authority structures in which one lives (e.g., government–citizen, parent–child, spouse–spouse, friend–friend, human being–God, etc.). We make choices under the influence of many whom we consider authorities . . . . [W]e must accept choice in a web of power relations. We must deem nonautonomous only those choices made under the influence of particularly overbearing power, and not simply because of deference to authority, be it of the state or of any other person or corporate entity.”).

109 Id. at 41–42. The Supreme Court has recognized the difference between coercive and persuasive speech in the private context as well—for example, by interpreting the First Amendment to permit statutory restrictions on private parties’ speech that rises to the level of regulable conduct because of its coercive effects. See, e.g., Ohralik v. Ohio State Bar Ass’n, 436 U.S. 447, 457 (1978) (“[I]n-person solicitation may exert pressure and often demands an immediate response, without providing an opportunity for comparison or reflection.”); NLRB v. Gissel Packing Co., 395 U.S. 575, 618–20 (1969) (“If there is any implication that an employer may or may not take action solely on his own initiative for reasons unrelated to economic necessities and known only to him, the statement is . . . a threat of retaliation based on misrepresentation and coercion, and as such without the protection of the First Amendment. . . . [E]mployees, who are particularly sensitive to rumors of plant closings, take such hints as coercive threats rather than honest forecasts.” (footnote omitted)); Watts v. United States, 394 U.S. 705, 707 (1969) (per curiam) (“The Nation undoubtedly has a valid, even an overwhelming, interest in protecting the safety of its Chief Executive and in allowing him to perform his duties without interference from threats of physical violence.”).

110Greene, supra note 108, at 43 (footnote omitted).

111430 U.S. 705, 713 (1977). For further examples of government coercion of individual expression, see Rutan v. Republican Party of Illinois, 497 U.S. 62, 74 (1990); and Elrod v. Burns, 427 U.S. 347, 355–56 (1976) (plurality opinion).

112 Wooley, 430 U.S. at 715.

113 See id. at 713; see also id. at 717 (“The State is seeking to communicate to others an official view as to proper appreciation of history, state pride, and individualism. Of course, the State may legitimately pursue such interests in any number of ways.”).

114 See Pub. Health Serv., U.S. Dep’t of Health, Educ., & Welfare, Pub. No. 1103, Smoking and Health: Report of the Advisory Committee to the Surgeon General of the Public Health Service (1964).

11515 U.S.C. § 1333 (2006).

116 Id. § 1338.

117 Id. § 1339.

118Although government coercion of private parties’ speech raises substantial First Amendment issues, such regulation is not always unconstitutional—especially in the context of commercial speech like cigarette labels and advertisements. See, e.g., Commonwealth Brands, Inc. v. United States, 678 F. Supp. 2d 512, 532 (W.D. Ky. 2010) (holding that such regulation of commercial speech satisfies the First Amendment because it is sufficiently tailored to meet a substantial government interest).

119 See Gia B. Lee, Persuasion, Transparency, and Government Speech, 56 Hastings L.J. 983, 1009–15 (2005); Lawrence Lessig, The Regulation of Social Meaning, 62 U. Chi. L. Rev. 943 (1995).

120 See Lee, supra note 119, at 984–88.

121 See Caroline Mala Corbin, The First Amendment Right Against Compelled Listening, 89 B.U. L. Rev. 939 (2009) (suggesting a constitutional right against government-compelled listening, especially in contexts that raise captive-audience concerns like state-mandated abortion counseling or diversity training); Redish & Finnerty, supra note 106, at 99 (identifying free speech concerns with respect to government speech in public schools that seeks to indoctrinate “a captive audience of undeveloped and impressionable minds”).

122 See Joshua Rhett Miller, Critics Decry Obama’s ‘Indoctrination’ Plan for Students, FOXNews.com (Sept. 2, 2009), http://www.foxnews.com/politics/2009/09/02/critics-decry-obamas-indoctrination-plan-students/.

123 See Obama Urges Students to Work Hard, Stay in School, CNN.com (Sept. 8, 2009, 7:17 PM), http://www.cnn.com/2009/POLITICS/09/08/obama.school.speech/index.html.

124Redish & Finnerty, supra note 106, at 90.

125Indeed, monopolistic government speech is increasingly unlikely given changes in communications technology that enable more speakers to participate, expanding opportunities for dissent and counterspeech. See Schauer, supra note 72, at 380 (“Although the quantity of government speech is no doubt increasing, and technological advances may make it possible for the government to convey its messages more efficiently, we should not forget that concomitant increases and improvements in private sector communication probably match this development. There seems to be no evidence whatsoever that the proportion of government speech within the total universe of communication is increasing. In fact, it seems quite possible that it is decreasing.”).

126Although I see a real distinction between the government’s persuasive and coercive speech, I am considerably more skeptical that another distinction proposed by some courts—permitting the government’s informational or factual but not persuasive expression—is meaningful. See, e.g., Harrison v. Rainey, 179 S.E.2d 923, 925 (Ga. 1971) (authorizing expenditures for a legal memorandum on the effects of, but not for advocacy materials regarding, a proposed state constitutional amendment); Rees v. Carlisle, 153 P.3d 1131, 1138–39 (Haw. 2007) (interpreting a statute to permit the government’s informational but not persuasive speech on a pending ballot measure); Stern v. Kramarsky, 375 N.Y.S.2d 235, 239–40 (Sup. Ct. Spec. Term 1975) (suggesting that the government’s “factual” speech on the ERA, but not its advocacy in support of the proposal, is legally permissible). Efforts to distinguish between “factual” and persuasive speech face often-insuperable challenges because the speaker’s most basic choices about her “factual” presentation can be intentionally and effectively persuasive, depending on quantity, order, contrast, tone, etc. See Martin H. Redish & Abby Marie Mollen, Understanding Post’s and Meiklejohn’s Mistakes: The Central Role of Adversary Democracy in the Theory of Free Expression, 103 Nw. U. L. Rev. 1303, 1338–39 (2009) (“[I]t is . . . impossible to separate information from opinion because the use of information is often a central element in the persuasive nature of opinion. . . . [I]nformation is often conveyed, not as an end in itself, but rather as part and parcel of an effort to employ the system of public discourse to achieve certain normative and personal goals.”).

127 See Alan K. Chen, Right Labels, Wrong Categories: Some Comments on Steven D. Smith’s Why Is Government Speech Problematic?, Denv. U. L. Rev. Online (Aug. 12, 2010, 9:14 AM), http://www.denverlawreview.org/government-speech/2010/8/12/right-labels-wrong-categories-some-comments-on-steven-d-smit.html (“First Amendment doctrine appropriately distinguishes between the[] two scenarios because the government’s own speech can rarely influence the public debate in the same qualitative or quantitative way as when it excludes private speakers’ ideas from the marketplace.”).

128 Joseph Tussman, Government and the Mind 80 (1977).

129Greene, supra note 108, at 2.

130Smith, supra note 46, at 952.

131Chen, supra note 127.

132 See supra notes 41–43 and accompanying text.

133 See Nat’l Endowment for the Arts v. Finley, 524 U.S. 569, 598 (1998) (Scalia, J., concurring in the judgment) (“It is the very business of government to favor and disfavor points of view on (in modern times, at least) innumerable subjects—which is the main reason we have decided to elect those who run the government, rather than save money by making their posts hereditary.”).

134Johanns v. Livestock Mktg. Ass’n, 544 U.S. 550, 559 (2005); id. at 574 (Souter, J., dissenting) (“The first point of certainty is the need to recognize the legitimacy of government’s power to speak despite objections by dissenters whose taxes or other exactions necessarily go in some measure to putting the offensive message forward to be heard. To govern, government has to say something, and a First Amendment heckler’s veto of any forced contribution to raising the government’s voice in the ‘marketplace of ideas’ would be out of the question.” (footnote omitted)).

135 See id. In considering this First Amendment challenge to a generic beef promotion campaign implemented by the Department of Agriculture and funded by taxes targeted at beef producers, the Johanns majority and dissenters differed vigorously, however, on the question whether government must identify itself as the source of that speech to successfully assert the government speech defense to the plaintiffs’ free speech claim. Compare id. at 562 (majority opinion) (“When . . . the government sets the overall message to be communicated and approves every word that is disseminated, it is not precluded from relying on the government-speech doctrine merely because it solicits assistance from nongovernmental sources in developing specific messages.”), with id. at 571–72 (Souter, J., dissenting) (“[I]f government relies on the government-speech doctrine . . . , it must make itself politically accountable by indicating that the content actually is a government message, not just the statement of one self-interested group the government is currently willing to invest with power.”).

136 See Alexander Meiklejohn, Free Speech and Its Relation to Self-Government 25 (1948) (“The final aim of the [town] meeting is the voting of wise decisions. The voters, therefore, must be made as wise as possible. The welfare of the community requires that those who decide issues shall understand them. They must know what they are voting about. And this, in turn, requires that so far as time allows, all facts and interests relevant to the problem shall be fully and fairly presented to the meeting. . . . What is essential is not that everyone shall speak, but that everything worth saying shall be said.”); Redish & Mollen, supra note 126, at 1307 (“[A] valid democratic theory of the First Amendment must protect all speech that allows individuals to discover their personal needs, interests, and goals—in government and in society at large—and to advocate and vote accordingly.”); see also Garrison v. Louisiana, 379 U.S. 64, 77 (1964) (recognizing “the paramount public interest in a free flow of information to the people concerning public officials, their servants”).

137Redish & Mollen, supra note 126, at 1337 (footnotes omitted).

138As Lyrissa Barnett Lidsky explains, “[T]he modern trend . . . has been to assume that audiences are savvy and sophisticated, capable of sorting through masses of information to discover truth, however provisional or contested. Indeed, these assumptions underpin two articles of faith in modern First Amendment theory: (1) audiences are capable of rationally evaluating the truth, quality, credibility, and usefulness of core speech without the aid of government intervention; and (2) more speech is better than less.” Lyrissa Barnett Lidsky, Nobody’s Fools: The Rational Audience as First Amendment Ideal, 2010 U. Ill. L. Rev. 799, 810–11.

139 See, e.g., James A. Gardner, Comment, Protecting the Rationality of Electoral Outcomes: A Challenge to First Amendment Doctrine, 51 U. Chi. L. Rev. 892, 892 (1984) (“A central though rarely articulated premise of many election laws and much democratic theory is that electoral outcomes should be rational rather than irrational—that they should reflect the true, reasoned, and informed choice of the people. Unfortunately, as political scientists have shown, people do not always vote rationally.”).

140 See, e.g., J. Skelly Wright, Money and the Pollution of Politics: Is the First Amendment an Obstacle to Political Equality?, 82 Colum. L. Rev. 609, 640 (1982) (“When the forum is limited by physical, technological, or economic factors, the messages of some speakers must be limited if all points of view are to be heard, so that the audience may enjoy a full range of uninhibited debate.”).

141Daniel R. Ortiz, The Engaged and the Inert: Theorizing Political Personality Under the First Amendment, 81 Va. L. Rev. 1, 4 (1995). Professor Ortiz himself finds both models incomplete. Ortiz, supra.

142 See, e.g., Robert Post, Participatory Democracy and Free Speech, 97 Va. L. Rev. 477, 482 (2011) (“[T]he best possible explanation of the shape of First Amendment doctrine is the value of democratic self-governance.”).

143 See, e.g., 2 Zechariah Chafee, Jr., Comm’n on Freedom of the Press, Government and Mass Communications 723 (1947) (“Now it is evident that government must itself talk and write and even listen.”); Thomas I. Emerson, The System of Freedom of Expression 698 (1970) (“Participation by the government in the system of freedom of expression is an essential feature of any democratic society. It enables the government to inform, explain, and persuade—measures especially crucial in a society that attempts to govern itself with a minimum use of force. Government participation also greatly enriches the system; it provides the facts, ideas, and expertise not available from other sources.”); Martin H. Redish & Daryl I. Kessler, Government Subsidies and Free Expression, 80 Minn. L. Rev. 543, 565 (1996) (“From the perspective of democratic theory, it is essential that these government employees inform the populace of the government’s policies and initiatives. Because the government informs the populace about its functioning through these subsidies, it facilitates self-government by providing members of the community with information and data on which to judge the performance of its political leaders. As a result, the electorate is better able to check elected officials and hold them accountable.”); Shiffrin, supra note 55, at 604 (describing government speech as providing the public with “the advantage of knowing the collective judgment of the legislature and of knowing the views of its representatives, which would in turn be useful for evaluating them”).

144 See Helen Norton, Constraining Public Employee Speech: Government’s Control of Its Workers’ Speech to Protect Its Own Expression, 59 Duke L.J. 1 (2009); see also Norton & Citron, supra note 5.

145Recall, as just one of many examples, the insights into government policymaking provided to the public by the Pentagon Papers (a Department of Defense study that reviewed U.S. military and diplomatic policy in Indochina). See N.Y. Times Co. v. United States, 403 U.S. 713, 714 (1971) (per curiam) (rejecting the government’s efforts to stop publication of the Pentagon Papers).

146Helen Norton, The Measure of Government Speech: Identifying Expression’s Source, 88 B.U. L. Rev. 587, 632 (2008).

147 Id.

148 Id. at 589–90.

149 See Steve Raabe, Ritter Bashes Tax-Cut Initiatives, Denver Post, May 14, 2010, at A1.

150 Id.

151 Id. (internal quotation mark omitted).

152Aldo Svaldi, Political Foes Turn Partners, Denver Post, May 24, 2010, at A18.

153Editorial, Candidates Weigh In on Right Side of Ballot Items, Denver Post, Feb. 28, 2010, at D3.

154The Supreme Court has noted on several occasions the informational value of private parties’ campaign speech with respect to contested ballot measures when rejecting limitations on such speech. See Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290, 299 (1981) (“Whatever may be the state interest . . . in regulating and limiting contributions to or expenditures of a candidate[,] . . . there is no significant state or public interest in curtailing debate and discussion of a ballot measure.”); First Nat’l Bank of Bos. v. Bellotti, 435 U.S. 765, 777 (1978) (“The inherent worth of the speech in terms of its capacity for informing the public does not depend upon the identity of its source, whether corporation, association, union, or individual.”).

155 See, e.g., James N. Druckman, Does Political Information Matter?, 22 Pol. Comm. 515, 515 (2005) (summarizing a study finding that “citizens can compensate for a lack of political information by using shortcuts to make the same decisions they would have made if they had that information”—“that many poorly informed voters (who lacked knowledge about the initiatives’ details) used elite endorsements (e.g., from interest groups) to emulate the behavior of well-informed . . . voters”); Michael S. Kang, Democratizing Direct Democracy: Restoring Voter Competence Through Heuristic Cues and “Disclosure Plus, 50 UCLA L. Rev. 1141, 1161 (2003) (“[R]eliance on heuristic cues is a learned practice based on past success and accuracy. Voting behavior in candidate elections, when heuristic cues are readily available, is relatively rational, consistent, and well-ordered, whereas in issue elections, particularly when heuristic cues are difficult to find, voting behavior appears more random, irrational, and contradictory. Without heuristic cues, voters in direct democracy are more confused, [and] money is more influential . . . . Even if they do not cure voter confusion in every instance, voters armed with heuristic cues will be much more likely to vote competently in the face of complexity than will voters without them.” (footnote omitted)). For a more critical assessment of heuristics’ value to voters, see Molly J. Walker Wilson, Behavioral Decision Theory and Implications for the Supreme Court’s Campaign Finance Jurisprudence, 31 Cardozo L. Rev. 679, 703–06 (2010).

156Kang, supra note 155, at 1141; accord id. at 1164–65 (“[M]uch of the electorate is rationally ignorant and unlikely to become more engaged in a way that cures concerns about voter confusion. . . . Heuristic cues quickly put uninformed voters on roughly equal footing with better-informed voters, even if they do not transform civic slackers into infallible or perfectly informed voters.”).

157 See, e.g., Pub. Health Serv., supra note 114 (describing the adverse health effects of smoking).

158Kang, supra note 155, at 1161–62.

159 See id. at 1169 (“People have plentiful access to political information, but otherwise busy and semi-interested voters need information disseminated to them in a way that requires them to take no affirmative steps or do anything more than they otherwise would do. Unless campaign finance information is delivered to voters at virtually no cost to them, such information may never reach them and will always remain too difficult for voters to acquire. Voters must be made aware of such heuristic cues to take advantage of them.”).

160 See Schauer, supra note 72, at 385 (“In the face of almost completely unrestrained criticism of government from all quarters, . . . views about the ability of government to use speech to falsify consent seem to require such a negative view of popular competence as to call into question the very reasons for considering democracy or majoritarianism to be any good at all.”).

161 See, e.g., June Fessenden-Raden et al., Providing Risk Information in Communities: Factors Influencing What Is Heard and Accepted, Sci. Tech. & Hum. Values, Summer/Fall 1987, at 94, 96 (finding that the trust people have in political institutions varies, as does their trust in the information provided by such governments); Schauer, supra note 72, at 381 (“[A]ntigovernment biases may be so great, particularly with reference to the veracity of political leaders, that much government speech may encounter a public strongly predisposed to disbelief.”); Leigh Contreras, Comment, Contemplating the Dilemma of Government as Speaker: Judicially Identified Limits on Government Speech in the Context of Carter v. City of Las Cruces, 27 N.M. L. Rev. 517, 539–40 (1997) (“[V]oters may tend to defer to the government’s judgment—although the assumption that government has an especially influential effect on the people depends on the public’s reverence and respect for government.” (footnote omitted)).

162Abner Greene has made this point about government speech more generally, outside of the specific campaign context that is the focus of this Article. See Greene, supra note 108, at 9, 11 (“[G]overnment speech often makes a distinctive contribution to public debate. If the government’s point of view were simply corroborative of private points of view, the affirmative argument for government speech would weaken as the concerns about government power rise. . . . Moreover, government persuasion on a contested matter plays an important role in countering private power. As one locus of power in society, government can check agglomerations of private power, just as checks on government ensure that it be only one voice among many. Additionally, even in a contested arena government speech can help foster debate, fleshing out views, and leading toward a more educated citizenry and a better chance of reaching the right answer.” (footnote omitted)).

163130 S. Ct. 876, 908–09 (2010).

164 Id. at 904–05.

165 Id. at 905. For arguments that Citizens United was wrongly decided even if one focuses on a liberty-based view of the First Amendment, see Molly J. Walker Wilson, Too Much of a Good Thing: Campaign Speech After Citizens United, 31 Cardozo L. Rev. 2365 (2010).

166Gayle Perez, Ritter: Amendment 58 Foes Using Scare Tactics, Pueblo Chieftain (Oct. 10, 2008), http://www.chieftain.com/news/local/ritter-amendment-foes-using-scare-tactics/article_e7c5ae2f-0d70-56e4-a4cb-26096f2209ee.html (“Ritter said the amendment will not raise taxes, but simply removes the tax credit that oil and gas companies have been receiving from the state for more than three decades. Of the money the state will keep from those credits, at least 60 percent will be devoted to providing financial aid to low-income resident[] students who attend a state public college or university. The remaining money will be used for wildlife habitat protection, renewable energy grants and to local communities impacted by oil and gas industry.”).

167Cari Merrill, Dems Make Case for 58, Fort Collins Coloradoan, Sept. 27, 2008, at A11.

168 See, e.g., David Michaels, Doubt Is Their Product: How Industry’s Assault on Science Threatens Your Health 85–90, 201 (2008) (documenting efforts by tobacco companies to contest claims of the health hazards of cigarettes and by oil companies to cast doubt on scientists’ claims on the role of fossil fuels in contributing to climate change).

169 See California: Proposition 14, Time, Sept. 25, 1964, at 23, 23, available at http://www.time.com/time/magazine/article/0,9171,876158-1,00.html.

170 Id.

171Totton J. Anderson & Eugene C. Lee, The 1964 Election in California, 18 W. Pol. Q. 451, 470 (1965).

172 Id. at 470–71.

173 California Proposition 14, supra note 169, at 23.

174Pete Wilson et al., Argument in Favor of Proposition 209, in California Ballot Pamphlet: General Election, November 5, 1996, at 32 (1996), available at http://vote96.sos.ca.gov/Vote96/html/BP/209yesarg.htm.

175 See Kang, supra note 155, at 1148 (“Many direct democracy elections, particularly on economic measures, attract spectacular disparities in campaign spending between opposing and supporting sides. Indeed, the consensus from empirical research is that spending advantages are nearly outcome-determinative when aimed at defeating a ballot measure.”).

176 Id. at 1158–59.

177 See Chris Chambers Goodman, (M)Ad Men: Using Persuasion Factors in Media Advertisements to Prevent a “Tyranny of the Majority” on Ballot Propositions, 32 Hastings Comm. & Ent. L.J. 247, 257 (2010) (“The [organizations’] names are specifically designed, in some[] cases to obfuscate, and these names of the supporters can be very influential in the outcome of the ballot measure.”); see also Bruce E. Cain, Commentary, Garrett’s Temptation, 85 Va. L. Rev. 1589, 1592–93 (1999) (“Many of the groups that succeed in getting initiatives on the ballot have primarily economic motives—e.g., insurance companies, lawyers, and teachers’ unions. To make matters worse from an informational perspective, these groups often adopt false, generic labels such as ‘Committee for a Just America’ or ‘Campaign for Consumer Justice.’ . . . In an ideal world, information heuristics would operate like warning labels on consumer items. In reality, those who sponsor initiatives know that labels can be important, and they choose labels strategically in order to create images they think voters will receive best.”).

178 See Gilda R. Daniels, Voter Deception, 43 Ind. L. Rev. 343, 381–86 (2010) (proposing legislative measures to address voter deception); Goodman, supra note 177, at 294–301 (offering a series of proposals to require greater transparency of the source of private speech in direct-democracy campaigns).

179 See Citizens United v. FEC, 130 S. Ct. 876, 886 (2010) (“The Government may regulate corporate political speech through disclaimer and disclosure requirements, but it may not suppress that speech altogether.”).

180 See David G. Savage, Corporate Campaign Spending Still Murky, L.A. Times, Oct. 27, 2010, at A1 (“Because of loopholes in tax laws and a weak enforcement policy at the Federal Election Commission, corporations and wealthy donors have been able to spend huge sums on campaign ads, confident the public will not know who they are, election law experts say.”); Editorial, The Secret Election, N.Y. Times, Sept. 19, 2010, at WK8 (noting Congress’s failure to pass disclosure requirements in the immediate aftermath of Citizens United).

181Buckley v. Valeo, 424 U.S. 1, 64 (1976) (per curiam).

182 See Rickert v. Washington, 168 P.3d 826 (Wash. 2007) (en banc) (striking down an amended statute that prohibited sponsoring with actual malice a political advertisement containing a false statement of material fact about a candidate for public office); State ex rel. Pub. Disclosure Comm’n v. 119 Vote No! Comm., 957 P.2d 691 (Wash. 1998) (en banc) (striking down a state statute that prohibited the sponsorship of political advertisements containing false statements of material fact). But see Vanasco v. Schwartz, 401 F. Supp. 87 (E.D.N.Y. 1975) (characterizing calculated falsehoods during political campaigns as unprotected by the First Amendment), aff’d mem., 423 U.S. 1041 (1976).

183 See Frederick Schauer, Facts and the First Amendment, 57 UCLA L. Rev. 897, 917–18 (2010) (“[N]either First Amendment theory, nor history, nor doctrine significantly restrict[s] the government’s ability to attempt to correct widespread public factual inaccuracy . . . .”). I leave for another day the question of whether the Constitution prohibits government falsehoods.

184 See Yudof, supra note 72, at 114–16 (describing how federalism, separation of powers, and the variety of administrative agencies considerably fragment political power, thus undermining the possibility of monolithic government speech).

185Letter from Eric H. Holder, Jr., Att’y Gen., to Former Administrators of the Drug Enforcement Agency (Oct. 13, 2010) (on file with author).

186 John Hoeffel, Holder Vows Fight over Prop. 19, L.A. Times, Oct. 16, 2010, at AA1.

187 Compare Joseph Blocher, Viewpoint Neutrality and Government Speech, 52 B.C. L. Rev. 695, 717–18 (2011) (“So long as the government identifies itself when speaking, the public can hold government speakers politically accountable and curb their excesses.”), with Steven Shiffrin, Government Speech and the Falsification of Consent, 96 Harv. L. Rev. 1745, 1752–53 (1983) (reviewing Yudof, supra note 72) (“[C]orporations seek profit without concern for externalities and use enormous wealth in ways bearing no necessary or even likely relationship to the beliefs of their shareholders . . . .” (footnote omitted)).

188Redish & Mollen, supra note 126, at 1317.

189 See id. at 1341 (“Speakers do not always speak solely to contribute to public discourse or solely for narrow personal economic gain.”).

190 See First Nat’l Bank of Bos. v. Bellotti, 435 U.S. 765, 791–92 (1978) (“[T]he people in our democracy are entrusted with the responsibility for judging and evaluating the relative merits of conflicting arguments. They may consider, in making their judgment, the source and credibility of the advocate.” (footnote omitted)).

191 See Citizens Against Rent Control v. City of Berkeley, 454 U.S. 290, 298 (1981) (recognizing a value to voters of knowing “the identity of those whose money supports or opposes a given ballot measure”); Buckley v. Valeo, 424 U.S. 1, 67 (1976) (per curiam) (“The sources of a candidate’s financial support . . . alert the voter to the interests to which a candidate is most likely to be responsive and thus facilitate predictions of future performance in office.”).

192 See Schauer, supra note 72, at 384–85 (“[That the First Amendment does not constrain government speech] does not mean that government communication is not a constitutional problem under some other clause of the Constitution, or that it is not a political or moral problem. . . . When government misuses its power to communicate we do have a problem, but this does not mean that we have a first amendment problem.” (footnote omitted)).

193 See supra note 19.

194 See supra notes 143–46 and accompanying text.

195 See Norton, supra note 146, at 599 (“[T]he government can establish its entitlement to the government speech defense only when it establishes itself as the source of that expression both as a formal and as a functional matter. In other words, government must expressly claim the speech as its own when it authorizes or creates a communication and onlookers must understand the message to be the government’s at the time of its delivery.”). For other commentators’ thoughtful discussions of these and related issues, see Randall P. Bezanson & William G. Buss, The Many Faces of Government Speech, 86 Iowa L. Rev. 1377 (2001); Caroline Mala Corbin, Mixed Speech: When Speech Is Both Private and Governmental, 83 N.Y.U. L. Rev. 605 (2008); Leslie Gielow Jacobs, Who’s Talking? Disentangling Government and Private Speech, 36 U. Mich. J.L. Reform 35 (2002); and Gia B. Lee, Persuasion, Transparency, and Government Speech, 56 Hastings L.J. 983 (2005).

196 See Johanns v. Livestock Mktg. Ass’n, 544 U.S. 550, 563–64 (2005).

197 See, e.g., Norton & Citron, supra note 5.

198Private speakers, unlike the government, have autonomy interests protected by the First Amendment. For this reason, the Court has interpreted the First Amendment to protect private speakers’ anonymity under certain circumstances. See, e.g., McIntyre v. Ohio Elections Comm’n, 514 U.S. 334, 357 (1995) (striking down Ohio’s ban on the distribution of unsigned political leaflets).

199Recall that this Article focuses only on speech that represents the views of a government entity or branch, as opposed to individual government officials’ speech expressing their own views on their own time and at their own expense, as individuals do not relinquish their First Amendment rights to express personal beliefs upon taking office. See, e.g., Colo. Taxpayers Union, Inc. v. Romer, 750 F. Supp. 1041, 1045 (D. Colo. 1990); see also Republican Party of Minn. v. White, 536 U.S. 765, 788 (2002) (holding that the First Amendment protects judges’ speech as candidates for judicial elections).

200Other commentators have plausibly identified other potential constitutional sources of similar principles, including equal protection and substantive due process. See Erwin Chemerinsky, Protecting the Democratic Process: Voter Standing to Challenge Abuses of Incumbency, 49 Ohio St. L.J. 773, 778 (1988) (“[T]he equal protection clause of the fourteenth amendment is violated if the government acts to aid only the incumbent. . . . Those who support challengers have their votes diluted by abuse of incumbency in exactly the same way the malapportionment or stuffing of the ballot box lessens the effectiveness of an individual’s vote.”); Greene, supra note 108, at 38 (noting that efforts to entrench incumbents and thwart challengers “violate[] one of the two key principles of the famous footnote four [i]n Carolene Products, and should be deemed invalid” (footnote omitted) (citing United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938))).

201 See Shiffrin, supra note 55, at 602 (“Citizens are entitled to a government that is neutral in the process of selecting candidates. Whether or not the concept of self-government is ‘central’ to the first amendment, it is undeniably an important first amendment value, and the integrity of the democratic process could rightly be questioned if government officially intervened in the political process to favor particular candidates. Whether or not the intervention was powerful, it would ipso facto disturb the first amendment equality principle.” (footnotes omitted)).

202 See Colorado Will Keep Stimulus Highway Signs, Street (Sept. 28, 2009, 10:53 AM), http://www.thestreet.com/story/10604159/1/colorado-will-keep-stimulus-highway-signs.html.

203 Id.

204 See, e.g., Chafee, supra note 143, at 764–65 (“Some of these quandaries seem to me unavoidable if we are to have any sort of adequate information service. Effective presentation of any recent achievement of the government, no matter how completely it is accepted by everybody, cannot help benefiting the party and the officials who made that achievement possible. . . . In spite of the risks that men who know exactly what they want and are acquainted with the latest techniques for manipulating public opinion will dispose of large sums for their personal or departmental advantage, we may be wise to run those risks for the sake of the values of public information . . . in enabling citizens to govern themselves more intelligently.”); Shiffrin, supra note 55, at 603 (“Non-partisan aspects such as informing the populace of government policy and explaining that policy are also necessarily partisan because incumbent candidates almost invariably claim that their reelection is justified by their link to the government policy they explain and defend.”).

205 Emerson, supra note 143, at 699.

206 See Shiffrin, supra note 55, at 617.

207 Id.

208For a discussion of possible statutory limitations on the government’s expressive expenditures, see infra notes 210–12 and accompanying text.

209Chen, supra note 127 (endnotes omitted).

210As discussed supra note 100, the fact that the government is not a First Amendment rights holder itself leaves open the possibility that its speech could be constrained through statute and other policy measures.

211 See supra note 134 and accompanying text.

212Wash. Rev. Code Ann. § 42.52.180(1) (West Supp. 2010); accord Colo. Rev. Stat. § 1-45-117(1)(a)(I) (2010) (prohibiting public entities from expending any public monies from any source for contributions to a campaign for elected office, or to urge electors to vote in favor of or against any ballot issue or referred measure); Okla. Stat. Ann. tit. 26, § 16-119 (West 1997) (prohibiting state officials from direct[ing] or authoriz[ing] the expenditure of any public funds under [their] care, except as specifically authorized by law, to be used either in support of, or in opposition to, any measure which is being referred to a vote of the people by means of the initiative or referendum, or which citizens of this state are attempting to have referred to a vote of the people by the initiative or referendum). Some of these laws, however, permit public officials to express their views on such measures in certain situations. See, e.g., Conn. Gen. Stat. Ann. § 9-369b(a) (West 2009) (explaining that its prohibition does not apply to a written, printed or typed summary of an officials views on a proposal or question, which is prepared for any news medium); Fla. Stat. Ann. § 106.113(3) (West Supp. 2011) (explaining that its prohibition does not preclude an elected official of the local government from expressing an opinion on any issue at any time); Wash. Rev. Code Ann. § 42.52.180(2)(a) (explaining that its prohibition does not apply to members of an elected body expressing their collective opinion on a ballot measure so long as required notice is provided and members of the public are given approximately the same opportunity to voice opposing viewpoints).

213 See Yudof, supra note 72, at 47 (“The greatest threat of government domination and distortion of majoritarian processes emanates from executive bodies and officers. The greatest hope of restraining that power lies with the legislative branches of government. If a legislative body determines that particular government expression threatens democratic processes, the courts should not second-guess that decision.” (footnote omitted)).

214 See Nev. Comm’n on Ethics v. Carrigan, 131 S. Ct. 2343 (2011) (rejecting a legislator’s First Amendment challenge to a Nevada law prohibiting public officials from voting or advocating with respect to matters in which the legislator has an actual or perceived conflict of interest and detailing the long history of such statutes).

215 See Yudof, supra note 72, at 111–38 (urging legislative bodies to consider the dangers of government speech in deciding what types of government expression to authorize and to prohibit); Jacobs, supra note 83 (proposing a variety of statutory checks on Article II speakers).

216 See supra notes 88–89 and accompanying text.

217 Comm. on Oversight & Gov’t Reform, U.S. House of Representatives, The Activities of the White House Office of Political Affairs, at ii (2008).