Notice & Comment

Linke v. Freed: Weighing In On Public Official’s Social Media Sites

Many state and local officials host social media sites and use them to converse with followers on matters related to their governmental responsibilities, among other things.[1]  Not surprisingly, many choose to block from their sites certain members of the public they find disagreeable.[2] Being disagreeable, or at least in disagreement with such actions, blocked followers sometimes sue alleging that their exclusion violates the First Amendment.[3]  One of the most notable examples was a challenge to President Trump’s practice of blocking some users on his Twitter account, Knight First Amdt. Inst. v. Trump, 928 F. 3d 226 (2d Cir. 2019), vacated as moot sub nom, Biden v. Knight First Amdt. Inst., 593 U. S. ___, 141 S.Ct. 1220 (2021); see, Bernard Bell, A Little Blue Birdie Told Me: Knight First Amendment Institute v. Trump (SDNY), YALE J. ON REG.: NOTICE & COMMENT (July 11, 2018).  In Lindke v. Freed, — U.S. —, 144 S.Ct. 756 (March 15, 2024), considered with a companion case O’Connor-Radcliffe v. Garnier, — U.S. —, 144 S.Ct. 717 (March 15, 2024)(per curiam), the Supreme Court grappled with whether exercising such custodial control constituted “state action” for purposes of 42 U.S.C. §1983 (“section 1983”),[4] and, for that matter, the Fourteenth Amendment’s “state action” requirement.[5]

The Court framed the issue before it as a choice between two competing standards. The first, the Sixth Circuit’s “authority or duty” test, required plaintiffs to prove “state action” by establishing that either (1) the “text of state law requires an officeholder to maintain a social-media account,” (2) the defendant official “use[s] … state resources” or “government staff” to run the account, or (3) the “accoun[t] belong[s] to an office, rather than an individual officeholder.”  Slip op. at 4-5 (quoting Lindke v. Freed, 37 F.4th 1199, 1203-04 (2022).  The second was the Ninth Circuit’s less demanding “appearance and function” test, Garnier v. O’Connor-Ratcliff, 41 F. 4th 1158, 1170–1171 (9th Cir. 2022); accord, Knight First Amdt. Inst. v. Trump, supra, 928 F. 3d at 236, focusing on whether the social media site functions as a “tool of governance” “swathe[d] in the trappings of [the official’s] office,” see, Davison v. Randall, 912 F.3d 666, 680 (4th Cir. 2019).[6] The Court chose the former, doing so unanimously.  Speaking through Justice Amy Coney Barrett, the Justices elaborated upon their less constricted version of the “authority or duty” test in a way that provides some guidance to government officials.  But that further analysis over-simplifies some of the complexity of government officials’ roles and use of social media and, moreover, leaves a number of questions unresolved.  (Granted, the Court acknowledged that the resolution of such issues was a highly fact-specific endeavor. Slip op. at 8.)

This blogpost will briefly discuss the Court’s decision and then highlight some aspects of the opinion and the questions left unanswered.

I. The Case and the Decision

After his appointment as City Manager for Port Huron, Michigan, James Freed expanded the use of his “public” Facebook page to post about not only personal matters but also about matters related to his public office. Significantly, his posts discussed communications received from other city officials and solicited feedback from the public on issues of concern. Freed often responded to comments with his own reply postings, including replies to residents who had posed their own inquiries concerning community matters. He occasionally deleted comments he considered “derogatory” or “stupid.”  Freed’s posts about the COVID–19 pandemic, some purely personal but others job-related, prompted Kevin Lindke to post responses expressing his displeasure with the City’s approach to the pandemic. Freed deleted Lindke’s comments, and ultimately, blocked him from commenting altogether. Slip op. at 2-4.

Justice Barrett, writing for the Court, noted that the Court’s state action jurisprudence has largely focused upon “whether a nominally private person has engaged in state action,” not whether a state official had acted as a private citizen rather than a state actor.  Slip op. at 6.[7]  The Court noted that officials who “routinely interact[] with the public” may appear to be “on the clock” continuously, and thus “their every encounter with the citizenry” could be viewed as “part of the job.”  But, the Court explained, such “broad-brush assumptions” would denigrate public officials’ continuing status as private citizens imbued with their own constitutional rights.  By excluding from liability “acts of officers [with]in the ambit of their personal pursuits,” the state-action requirement serves to “protect[] a robust sphere of individual liberty” for such officials.  The Court noted public employees’ First Amendment right, in certain circumstances, to speak as . . . citizen[s] addressing matters of public concern,” and observed that “editorial control over speech and speakers on [the public employee’s] properties or platforms” is “part and parcel” of that right. Slip op. at 6-8.

The Court laid out its two-pronged test for deciding when a public official’s social-media activity qualified as state action, and then elaborated upon each prong, and its justification.  A public official’s social media activity constituted state action under §1983 only if the official both (1) possessed actual authority to speak on the state’s behalf, and (2) purported to exercise that authority when speaking on social media. Slip op. at 8.[8] Referencing the “appearance and function” test it was rejecting, the Court noted that a social media site’s appearance and function could assume relevance at the second step, but “cannot make up for a lack of state authority at the first.” Slip op. at 8-9.

The Court derived the test’s first prong from “the bedrock requirement that ‘the conduct allegedly causing the deprivation of a federal right be fairly attributable to the State,’” asserting that “[a]n act is not attributable to a State unless it is traceable to the State’s power or authority.” Slip op. at 9.

Freed’s conduct could be attributed to the State only if he was “possessed of state authority” to post city updates and register citizen concerns. If the State did not entrust Freed with responsibilities to communicate in such a manner, it could not “fairly be blamed” for Freed’s manner of doing so. The Court observed: “Freed can[not] conjure the power of the State through his own efforts.” Slip op. at 10.[9]

Thus Lindke had to show not just that Freed possessed “some authority to communicate with residents on behalf of Port Huron,” but that his alleged censorship of Linke’s posts was connected to Freed’s speech “on a matter within Freed’s bailiwick.” Slip op. at 10 (emphasis in original).[10]

In discerning the scope of a public official’s authority communicate on particular matters, the Court turned to section 1983, identifying “statute, ordinance, regulation, custom, or usage” as the “potential sources of such power.” The Court explained that “statutes, ordinances, and regulations” refer to “written law through which a State can authorize an official to speak on its behalf.” “Custom” and “usage,however,“ encompass[es] ‘persistent practices of state officials’ that are ‘so permanent and well settled’ that they carry ‘the force of law.’” Slip op. at 11.  Thus, the authority to speak need not always appear in written law.  However, whatever the source of the authority, “if an official has authority to speak for the State, he may have the authority to do so on social media.”  Id.

Moreover, the Court continued, sometimes the mere “grant of authority over particular subject matter may reasonably encompass authority to speak about it officially.”  But at other times it may not.  The key to navigating this distinction is not “whether making official announcements could fit within the job description; but whether making official announcements is actually part of the job that the State entrusted the official to do.”  Slip op. at 12.  The Court chose not to further elaborate on this apparently critical, yet subtle distinction.

With those observations, the Court turned to the second prong of its test, namely that the official must purport to be speaking in an official capacity.  The Court almost immediately turned to “a hypothetical from the offline world.” It posited the following series of events: “A school board president announces at a school board meeting that the board has lifted pandemic-era restrictions on public schools,” and, later, “at a backyard barbecue with friends whose children attend public schools,” shares the identical information. The Court explained that the first statement “is state action taken in his official capacity as school board president;” while “the latter is private action taken in his personal capacity as a friend and neighbor.” Though “the substance of the announcement is the same,” the Court explained, the context—an official meeting versus a private event—differs.” The School Board President had “invoked his official authority only when he acted as school board president.” Slip op. at 12-13.

Reasoning from that hypothetical, the Court observed — “[h]ad Freed’s account carried a label (e.g., ‘this is the personal page of James R. Freed’) or a disclaimer (e.g., ‘the views expressed are strictly my own’),” he would be entitled to a heavy, but rebuttable, “presumption that all posts on the site were personal.”  “Just as we can safely presume that speech at a backyard barbeque is personal, we can safely presume that speech on a ‘personal’ page is personal (absent significant evidence indicating that a post is official).”  Slip op. at 13.[11] Had Freed used “an account belonging to the political subdivision with which he was affiliated (e.g., a ‘City of Port Huron’ Facebook page) or [one] passed down to whomever occupies a particular office (e.g., an ‘@PHuronCityMgr’ Instagram account), the opposite presumption would be warranted.”  But Freed had failed to designate his page as either “personal” or “official,” and his posts encompassed the personal as well as the arguably “official,” suggesting that the site was devoted to “mixed use,” i.e., one including both personal and official posts. Slip op. at 13.

In such circumstances, Justice Barrett cautioned courts to remain aware that “official[s] do[] not necessarily purport to exercise [their] authority simply by posting about a matter within the realm of the[ir] authority.” An official might post job-related information for any number of personal reasons, “from a desire to raise public awareness to promoting his prospects for reelection.”  Slip op. at 14 (emphasis added).  Moreover, “many public officials possess a broad portfolio of governmental authority that includes routine interaction with the public,” making it challenging to “discern a boundary between their public and private lives.”  Id.[12]

That said, “lest any officials lose” the right to speak about public affairs in their personal capacities, plaintiffs must shoulder the burden of establishing that the defendant official is purporting to exercise state authority in specific posts. Slip op. at 14.

The Court added that when doubt remains regarding the status of an official’s post, “additional factors might cast light,” providing one particularly clear guidepost — “an official who uses government staff to make a post will be hard pressed to deny that he was conducting government business.”[13]

The Court concluded by observing that “[t]he nature of the technology matters to the state-action analysis.”  Slip op, at 14.  When only deletion of particular posts is at issue, the only posts relevant to the state action inquiry are those from which the plaintiff’s comments were removed.  Slip op. at 15.  But if an official blocks a user from commenting altogether, as Freed had ultimately done, “a court would have to consider whether Freed had engaged in state action with respect to any post on which Lindke wished to comment.”  Id. (emphasis added.)[14]  The Court cautioned public officials that when page-wide blocking is the only option available, failing “to keep personal posts in a clearly designated personal account” would expose the public official “to greater potential liability.” Id.

The Court did not opine on the First Amendment standards applicable to an officeholder’s “official” social media sites and posts.[15]  Nor, of course, did it discuss the additional protection the qualified immunity doctrine affords officials unsure of whether their moderation efforts constitute “state action.”[16]

The Court essentially remanded the companion case, O’Connor-Radcliffe v. Garnier, to apply the revised “authority or duty” test it had laid out in Lindke v. Freed.[17]  

II. Observations

The Court’s opinion is surprisingly short for such a complex and multi-faceted topic as public officials’ control over their individual social media sites.  The Court resolved the case with a relatively minimalist opinion, deciding little more than necessary while providing some important guideposts to assist lower courts and public officials grappling with constitutional questions regarding their exercise of editorial and curatorial control on such websites.  Nevertheless, several preliminary observations are worthwhile.

A. Guideposts in the Opinion

In its opinion, the Court provides three helpful guideposts, two explicitly and one more subtly.  The first is the Court’s admonition that public officials keep their private and public social media accounts distinct.  For public officials, adherence to such a practice allows them to exercise virtually absolute curatorial control over discussions on private social media accounts while needing to observe constitutional constrains in exercising control only with respect to their public social media accounts.  It also allows officials to maintain a sphere of privacy in which they can share their private lives only with a select group of people.  Indeed, given the ease and wisdom of such a solution, the Court perhaps should have placed the onus on officials who fail to do so to demonstrate that work-related messages on bi-polar social media sites do not constitute state action subject to whatever constraints the First Amendment imposes upon exercising curatorial control over discussions related to such messages.

The second guidepost, that public officials should not expect social media sites to be treated as personal, rather than official, expression when they use public resources to post and exercise curatorial control on a social media site is unquestionably sound, and needs little elaboration.  However, one might wonder whether the same approach applies when the official uses staff only to monitor the official’s social media site and bring to his or her attention particularly important comments from followers.

The third guidepost is more subtly stated, but addresses an important concern. One troubling aspect of public officials’ use of social media sites on which they block access to some is that it may lead to distinctions among citizens in terms of access to information about government actions.  Thus, followers of the official’s social media site may receive information about how a program is administered or about potential dangers that other interested members of the general public lack.[18]  Moreover, even if government actions are ultimately announced through recognized official sources, followers may learn of the government action earlier than the general public if a public official first announces it on the official’s restricted social media site.  Indeed, this was a specific concern regarding access to Donald Trump’s Twitter account.  Knight First Amdt. Inst. v. Trump, supra, 928 F. 3d at 231-32, 235-36.

Information regarding government actions should be available to all.  See, Bernard W. Bell, et al., Disclosure of Agency Legal Materials: Final Report 8-13 (May 30, 2023) (with Cary Coglianese, Michael Herz, Margaret Kwoka and Orly Lobel)(focusing on government legal materials).  Indeed, the Freedom of Information Act (“FOIA”), which is replicated in some form in almost every state, is premised on making government information equally available to all.  Nat’l Archives & Records v. Favish, 541 U.S. 157, 171-72 (2004).[19]  

The Court wisely suggests this guidepost, albeit obliquely, by the use of the school board president hypothetical.  Though not exactly the point of the vignette, the Court’s resolution of it suggests that an official who provides information regarding official actions unavailable anywhere else is likely to be viewed as engaging in state action.  That said, almost invariably, if the official is sufficiently detailed in response to individual’s concerns about the individual’s particular circumstances, the official will almost inevitably discuss some matters that may not have been addressed in similar detail in official sources.

Similar issues have arisen in the context of journalists’ access to government officials.  Particular journalists are sometimes “frozen out” by public officials, either because of their unfavorable coverage or for other reasons.  In an early notable case, Sherril v. Knight, 569 F.2d 124 (1977), a reporter asserted a procedural due process claim arising from his exclusion from the White House.  He argued that he had been denied a White House press pass without an appropriate opportunity to respond to allegations that he had previously assaulted a state official’s press secretary.  There, the Court distinguished access to the White House press facilities, publicly available to journalists as a source of information, from the President’s discretion to grant interviews and briefings to selected journalists.[20]  As one note-writer has suggested, any constitutional rule regarding discriminatory access “must walk a fine line between allowing socially useful, historically sanctioned exclusive access and giving government officials free rein to engage in viewpoint discrimination when determining who will get access to their administration.” Note, Viewpoint Discrimination and Media Access to Government Officials, 120 Harv. L. Rev. 1019, 1020 (2007).

B.  Policies Regarding Official’s Social Media Sites

The Court’s observation that officials cannot “conjure up” state action may loom large in the future.  The principle ensures that state and local governments possess the power to set policies that help define the distinction between public and private social media sites, and may well allow state and local governments to impose some limitations on officials with regard to their social media policies.[21] 

Thus, governments could quite probably require officials to put disclaimers on their individual social media sites advising followers that the sites do not represent the official views of the government and that official views can be found only on identified government websites.[22]  The disclaimer might also note that statements made by the official on the site have no official significance and that any statement made by a member of the public cannot be relied upon as a basis for notice to the government or the public official nor satisfy any obligation to submit claims or information to the government.

Presumably, the government could also preclude officials from announcing their actions or directives on social media sites before they become available on a government website or from some other source available to the public.  Alternatively, the government might also preclude public officials from banning individuals from the official’s social media site if they announce matters not generally available to the public on their sites.  Or the government might require officials to make any discussions on their social media page regarding the discharge of their responsibilities available on governmental websites as well, to ensure that all citizens have access to important information.  They could require officials to allow staff to monitor such sites to ensure that information that should be “of record” is maintained by the government or summarized on an official website or social media site. 

Some of the limitations suggested above might have “employee speech” implications that government entities will have to consider.  But such policies would surely provide important context with respect to whether exercising editorial control over a particular social media site does or does not involve state action.

C. The Scope of Officials’ Duties

The Court focuses on officials’ duties, suggesting that speech can sometimes be considered state action if the speech regards matters falling within the official’s responsibility.  The proper analysis of the role an official’s duties plays in the analysis will undoubtedly require further refinement and elaboration.  The Court neither adequately recognizes the inherent power of government officials to speak nor the role of government officials in facilitating citizen’s efforts to seek redress of their grievances and participate in government.

First, the Court’s apparently exclusive focus on statutes, regulations, ordinances, and even customs and usages regarding the authority to make official statements, is myopic.  Government, and presumably government officials, have an inherent power to speak to the public to facilitate the successful performance of their responsibilities as government officials.  See, MARK G. YUDOF, WHEN GOVERNMENT SPEAKS: POLITICS, LAW, AND GOVERNMENT EXPRESSION IN AMERICA 38-50 (1983).[23] For example, the power to prosecute surely includes the power not only to set enforcement policies, but also to announce those policies and discuss them with the public or members of the public.  The power to promulgate regulations surely encompasses the power to explain their meaning.  Indeed, in a seminal case undoubtedly familiar to many readers of this blog, the D.C. Circuit expounded upon the critical nature of government officials’ ability to communicate, indeed communicate ex parte, with regulated entities and those affected by the regulatory regime.  Sierra Club v. Costle, 657 F.2d 298, 401 (D.C. Cir. 1981).[24]  Thus, the absence of any explicit written authorization, or even an established “custom or usage,” should not preclude a finding that an official is engaging in state action in running a social media site.  Indeed, the power to speak officially on matters within one’s responsibilities should be presumed unless explicitly withheld.  Such a presumption would better reflect the inherent power of government officials to speak about their areas of responsibility.

Second, the Court fails to discuss an important part of the context of public officials’ speech, namely that the power to speak officially has certain constraints.  Think of customer service representatives at the Social Security Administration.  Such employees clearly have within their “bailiwick” the power to speak “officially” regarding eligibility for disability benefits and the requirements of the application process for such benefits.  However, they may do so only in a particular structural and temporal context.  Their speech is subject to supervision by their superiors and temporally limited to their duty hours.[25] 

Thus, the problem of social media sites is more acute when the official speaking has the authority to make final decisions on matters within their bailiwick.  The social media problem, in terms of state action, is correspondingly less acute for those who do not head of departments, units, or agencies.[26] 

Moreover, such a consideration of official’s more particularized context in discussing matters within their sphere of responsibility provides important insights that may prove useful in analyzing the liability of and constitutional constraints upon members of multi-members bodies who cannot take official actions by themselves.[27]  Thus, members of legislative bodies running their own social media sites might well be less likely to be viewed as engaging in “state action” than high level executive branch officials.  Though the former, like the latter, possess the power to speak on issues, the former can take action only collectively.[28]  Of course, local governments’ governing structures are more multifaceted than those of the federal government and can involve the melding of legislative and executive functions in more variegated ways.  Nevertheless, whether the official can act individually as opposed to acting only as a part of a larger body seems an important question.

Third, with respect to the authority to speak, some officials have the power to speak, but by tradition and usage, speak officially only via limited means.  Take trial judges, whose positions involving “speaking” on matters that come before them (and who are generally not members of a body that acts collectively).  However, by tradition, and, in part, by canons of judicial ethics, such official speech is limited to comments in a courtroom and written opinions.  See generally, Bernard W. Bell, Administrative Adjudicators Extrajudicial Statements,” 35 Notre Dame Journal of Law, Ethics & Policy 617, 627-34 (2021).  Thus, it would seem odd to view such judges as speaking in their capacity as a “state actors” in exercising editorial control over their social media sites.

Turning briefly to government officials’ use of social media sites for soliciting comments from constituents and making themselves available for constituents who wish to “redress their grievances,” certainly acting in such a role is an inherent part of most elected officials’ responsibilities (and quite possibly a responsibility of many appointed officials as well).  Such functions are essential to a legitimate and vital democracy.[29]  Indeed, while the “public function” prong of the “state action” doctrine is quite limited, see, Manhattan Community Access Corp. et al. v. Halleck, 587 U.S. 802, 809-11 (2019); Jackson v. Metropolitan Edison, 419 U.S. 345, 352-53 (1974), it arguably encompasses receiving and considering comments and grievances regarding government actions and contemplated government actions.[30]  It is difficult to conceive of a private entity executing such a function; listening and responding to constituents seems inextricably intertwined with the function of serving as an elected representative.  Indeed, Justice Barrett can be read to suggest as much in asserting that receiving comments on a proposed rule, albeit in the context of conducting notice-and-comment rulemaking, “belongs exclusively to the State [because] its exercise is necessarily governmental.”  Slip. Op. at 13 n.2. 

Citizens’ ability to interact with local governmental officials has been viewed as a particular strength of local government, and one of the attributes that makes such governments “closest to the people.”[31] The greater accessibility of local officials to their constituents, in additional to local officials’ greater knowledge of local conditions, has been viewed as a reason to protect local governments’ viability and devolve authority to the state and local levels of government.[32]  Thus, vehicles by which state and local officials ensure accessibility to their citizens would seems to serve a particularly important official function, suggesting care in characterizing a public official’s social media site, on which that official solicits comments and invites raising concerns, as a private one dissociated from the official’s role as a state actor.  Of course, these observations suggest that with respect to websites that offer a vehicle for public complaints, inquiries and comments, legislators’ social media sites should be particularly likely to be found official (in comparison to executive officials’ sites), given representatives’ arguably greater responsibilities with respect to hearing citizens’ concerns. 

D. The Officeholder as Candidate

Many elected officials seek reelection to their current office or election to another office.  One key component of such electoral efforts is the official’s record in their current (and past) offices and the political loyalty built through their actions in office.  And, of course, such concerns undoubtedly influence officeholders long before they declare their candidacies for reelection or election to a higher office.  Thus, communications with constituents, both soliciting their viewpoints and informing them of the official’s successes in office, serve not only an “official” function (tied to the officeholder actually holding public office), but also an electoral function.  Indeed, one criticism of the congressional “franking” privilege, i.e., the privilege of sending mass mailings free of charge to constituents, is that those communications largely provide a means of promoting re-election, giving incumbents a substantial advantage over challengers, who cannot mail out campaign literature for free.  See, Congressional Research Service, Franking Privilege: Historical Development and Options for Change, Report No. RL34274, 14 (Updated May 3, 2016).

Justice Barrett, for the Court, suggests that communications discussing job-related matters aimed at “promoting [the office holder’s] prospects for reelection” qualify as personal.  Such an approach has the strength of making the rules applicable to candidates consistent — treating similarly candidates who currently hold governmental offices and candidates that do not.  Thus, both can host social media sites discussing issues relevant to the election without being burdened with the obligations that would arise if such communications constituted state action.  However, many communications will essentially have a dual purpose, soliciting views of constituents and informing them of developments as a part of serving in a particular office as well as increasing the office holder’s prospects for reelection or election to some other office. Such communications help accomplish the latter by cementing the loyalty of their constituents and ensuring that their “positions” on public matters are consistent with their constituents’ values.  Most officeholders’ twin goals, serving constitutes and furthering their own electoral prospects, are inextricably intertwined in ways that the Court’s cryptic remark about electoral motivations does not fully acknowledge.[33]  

Officials might avoid this conundrum by maintaining separate “electoral” and “governmental” social media sites.  If the “electoral” site is designated for posts related to the office holder’s re-election effort, and is not used to subvert the concerns about use of private social media sites for governmental business, those electoral sites could be viewed as not involving state action, thus freeing content-moderation decisions from First Amendment constraint.  The “governmental” site would continue to be limited by the First Amendment, as exercising curatorial control of the site would continue to be considered state action.

III. Conclusion

Government officials’ use of social media sites to interact with their constituents is generally a salutary development, giving the public an additional, more informal means of getting information about their government and interacting with elected and appointed officials.  But the use of such social media sites raises tricky questions about the line between public officials’ public and private lives. The Supreme Court, in its first foray into the area, could not be expected to resolve all the complexities surrounding public officials’ use of social media sites to discuss job-related matters.  As laid out above, lower courts will likely be called upon to address many issues in trying to draw the line between government official’s public roles and private selves in moderating social media sites.

[1] My thinking on this question was developed in supervising an independent study project undertaken by Michelle Nestor, J.D. candidate, Rutgers Law School, Class of 2024.  Though much of the content of this blogpost incorporates research leads and comments I provided Ms. Nestor, this post is also informed by sources Ms. Nestor found and insights she provided, which I hereby acknowledge.

[2] Leora Smith & Derek Kravitz, Governors and Federal Agencies Are Blocking Nearly 1,300 Accounts on Facebook and Twitter, PROPUBLICA (Dec. 8, 2017).

[3] Sarah Ludington, Lauren Smith, & Christian Bale, How Social Media Platforms Can Promote Compliance with the First Amendment, N.Y.U. L. REV. FORUM (Sept. 30, 2022)(“[o]ver the past several years, prospective clients have reached out . . . with what is becoming a common complaint: a public official . . . has censored their comments on the official’s social media page, or worse, has blocked them from participating on the page entirely”).

[4] This term the Court seems to have discovered it can no longer avoid constitutional issues raised by rise of social media, addressing cases involving state regulation of social media companies, Moody v. NetChoice, LLC, Dkt. No. 22-277, NetChoice, LLC v. Paxton, Dkt. No. 22-555, the Biden Administration’s cajoling of social media companies, Murthy v. Missouri Dkt. No. 23-411, and the constitutional implication of public officials blocking followers from their sites.

[5] As the Court observed the section 1983 “state action” requirement “tracks that of the Fourteenth Amendment, which obligates States to honor the constitutional rights that § 1983 protects.”  Slip op. at 5 (emphasis in original).

[6] One note writer asserts that the test originated in Davison v. Randall, 912 F.3d 666 (4th Cir. 2019).  See, John B. Tsimis, Looks Matter On Social Media: How Should Courts Determine Whether A Public Official Operates Their Social Media Account Under  Color Of State Law?, 91 Fordham L. Rev. 2061, 2074 (2023).

[7] Indeed, Polk County v. Dodson, 454 U. S. 312, 318–319 (1981), involving public defenders, was one of the rare cases in which the Supreme Court has found a public employee not to be a state actor.  Even so, the Court would later hold that in exercising peremptory challenges, defense attorneys, whether employed by the government or not, are engaged in state action, and thus the Batson v. Kentucky, 476 U.S. 79 (1986) proscription on racially-biased peremptory challenges applies to defense attorneys as well as prosecutors.  Georgia v. McCollum, 505 U.S. 42 (1992).  The distinction follows the Court’s pattern of being particularly expansive in finding “state action” in cases involving racial discrimination. See, Jackson v. Metropolitan Edison Co., 419 U.S. 345, 374 (1974)(Marshall, J. dissenting); Adickes v. S. H. Kress Co., 398 U.S. 144, 190-191 (1970)(Brennan, J., concurring and dissenting).

[8] In some ways this test resembles the Chevron Step Zero test coined in U.S. v. Mead Corp., 533 U.S. 218, 226-27, 233 (2001)(“[w]e hold that administrative implementation of a particular statutory provision qualifies for Chevron deference when it appears that Congress delegated authority to the agency generally to make rules carrying the force of law, and that the agency interpretation claiming deference was promulgated in the exercise of that authority.”)

[9] This limitation resembles that governing invocation of “apparent authority” more broadly, where the representation on which a plaintiff claims to have relied must be that of the principal, not the agent.  See, RESTATEMENT (THIRD) OF AGENCY § 2.03 (“[a]pparent authority is the power held by an agent or other actor to affect a principal’s legal relations with third parties when a third party reasonably believes the actor has authority to act on behalf of the principal and that belief is traceable to the principal’s manifestations (emphasis added)); id,. comm. c (“[a] belief that results solely from the statements or other conduct of the agent, unsupported by any manifestations traceable to the principal, does not create apparent authority unless, . . . the agent’s conduct has been directed by the principal.”). 

However, the law of apparent authority is more nuanced that this straightforward principle suggests. The principal’s “manifestation” of the agent’s authority to act, as defined in Restatement § 1.03, i.e., written or spoken words or other conduct manifesting the principal’s “consent or intention,” may take many forms. These include explicit statements the principal makes directly to a third party, as well as statements made by others concerning an actor’s authority, traceable to the principal, that reach the third party.  Moreover, in some circumstances, the agent’s statements can provide a basis for a reasonable conclusion that such statements constitute the principal’s manifestation of intent to give the agent authority.  See also, RESTATEMENT (THIRD) OF AGENCY §§ 3.03, comm. b, 6.11, 7.08.

[10] The Court was careful to reaffirm the principle that the “[m]isuse of power, possessed by virtue of state law,” constitutes state action, noting that the raison d’être of every §1983 suit is an alleged misuse of power.  Slip op. at 10. In particular, such suits are premised on the principle that the Constitution precludes the valid exercise of state authority to deprive anyone of a federal right. Id. at 11.  But this did not blunt the Court’s main point, namely that to constitute state action an official’s conduct must have some source in state law. Succinctly put, “[t]o misuse power, . . . one must possess it in the first place.” Id.

[11] The Court cautioned that an official “cannot insulate government business from scrutiny by conducting such business on a personal page,” offering two examples. Slip. Op. at 13 n.2.  First, an official who designates space on a nominally personal page as the official channel for receiving comments on a proposed regulation would be engaged in state action when exercising control over comments on the regulation made on the social media site.  Id. Because the power to conduct notice-and-comment rulemaking belongs exclusively to the State, its exercise is necessarily governmental.  Id.  Second, a mayor would engage in state action when hosting a city council meeting online by streaming it only on his personal Facebook page. Id. (emphasis added).

[12] Presumably here the President of the United States comes to mind.  In an earlier blogpost, I suggested that the President might be treated distinctly among federal officials in terms of control over his or her social media sites. See A Little Blue Birdie, supra (“the Presidency melds together the individual office holder with a governmental institution, the Executive Branch”).

[13] For example, President Trump used at least one White House aide to craft and post tweets on his Twitter account. Knight First Amdt. Inst. v. Trump, supra, 928 F. 3d at 231, 235-36.

[14] Ludington and her colleagues also focus on technology in their discussion of public officials’ websites. However, they primarily focus on a different problem — ensuring that the public officials’ actions of deleting particular messages or blocking members of the public are more “transparent,” thus facilitating efforts to contest such actions in court.  Ludington, et al., supra note 3.

[15] The applicable First Amendment standard could, in part, address the court’s concerns about a public official retaining some curatorial control as a part of maintaining the official’s right to speak out on public issues as a citizen.  For example, some courts have held that public officials’ social media sites are designated “public fora,” severely limiting content-based constraints on participants’ speech, and seemingly requiring that followers be allowed to comment on posts.  See, e.g., Knight First Amdt. Inst. v. Trump, supra, 928 F.3d at 237.  But given the interest of the public official as speaker in maintaining control over a social media site, the Court could hold that such sites are “limited public fora” or perhaps even a non-public fora.  For a discussion of “forum” doctrine, see Minnesota Voters Alliance v. Mansky, 585 U.S. 1, 11-12 (2018)(finding a polling place to be a nonpublic forum).

Consider Arkansas Educational Television Commission v. Forbes, 523 U.S. 666 (1998).  There the Court held that while public broadcaster’s televised election debate was subject to “public forum” analysis, constraining the broadcaster’s discretion to exclude some candidates from the debate, “broad rights of access for outside speakers would be antithetical, as a general rule, to the editorial discretion that broadcasters must exercise to fulfill their journalistic purpose and statutory obligations.”  Id. at 673.  While different, the concerns regarding a public official’s ability to have a social media site in which they can get out their own message, keep the site from becoming unwieldy with irrelevant and off-topic messages, and ensure civility at least with respect to participants’ communications regarding each other, arguably justify treating a public official’s social media site somewhat more like at public meaning than a municipalities’ streets and parks.  And presumably officials could, like O’Connor-Ratcliff and Zane, limit followers’ reactions to emojis, a limitation that would not ordinarily be permitted in a public forum.

[16] The much criticized “qualified immunity” doctrine precludes imposition of liability upon public officials for “conduct [that] does not violate clearly established statutory or constitutional rights of which a reasonable person would have known.”  Harlow v. Fitzgerald, 457 U.S. 800, 806 (1982).  Importantly, given the “highly fact-specific” nature of the “state action” inquiry in this context, slip op. at 8, the contours of the law must be sufficiently clear in the particularized context “that a reasonable official would understand that what he is doing violates” a constitutional right.  Anderson v. Creighton, 483 U.S. 635, 640 (1987).  See generally, Kit Kinports, The Supreme Court’s Quiet Expansion of Qualified Immunity, 100 Minn. L. Rev. Headnotes 62 (2016).  For a recent controversial case involving the scope of qualified immunity, see Villarreal v. City of Laredo, 94 F.4th 374 (5th Cir. 2024)(en banc).

[17] The defendants, Michelle O’Connor-Ratcliff and T. J. Zane, had created public Facebook pages to promote their campaigns for election to a local school district’s Board of Trustees, and continued to use the site after taking office to discuss school district matters.  O’Connor-Ratcliff v. Garnier, supra, slip op. at 1.  Their posts included “board-meeting recaps, application solicitations for board positions, local budget plans and surveys, and public safety updates.” Id. “They also used their pages to solicit feedback and communicate with constituents.”  Id. At some point, the Garniers, critics of the board, began posting “lengthy and repetitive comments” on O’Connor-Radcliff’s and Zane’s posts.  Unremarked by the Supreme Court, sometime after they blocked the Garniers, the defendants began using Facebook’s “word filter” to preclude all verbal comments on their public pages.  O’Connor-Ratcliff v. Garnier, 41 F.4th 1158, 1166, 1168 (9th Cir. 2022).  This allowed unblocked followers to register non-verbal reactions, using one of Facebook’s reaction buttons (such as the thumbs up symbol). Because they were blocked, the Garniers were unable to leave these nonverbal reactions on the Trustees’ Facebook pages.  Id. at 2.

[18] The Court seems to view “rais[ing] public awareness” as a non-official motivation, but many public officials might quite properly view such efforts at public education as essential aspect of their official responsibilities.

[19] Accord, Maricopa Audubon Society v. U.S. Forest Service, 108 F. 3d 1082, 1088 (9th Cir. 1997); Yonemoto v. Department of Veteran Affairs, 686 F. 3d 681, 689-90 (9th Cir. 2011).  Even in the FOIA context, very limited “selective disclosure” of classified information has been found permissible.  See, Bernard W. Bell, Plausible Deniability: Selective Disclosure of Information and FOIA, YALE J. ON REG.: NOTICE & COMMENT (May 17, 2018).

[20] Two other cases that addressed exclusions of journalist due to their unfavorable coverage are Baltimore Sun v. Ehrlich, 437 F.3d 410 (4th Cir. 2006)(taking an approach similar to Sherril v. Knight), and Borecca v. Fasi, 369 F. Supp. 906 (D. Haw. 1974).

[21] See Municipal Research and Service Center, Social Media Policies (Last Modified: February 23, 2024) (Bothell Personnel Policies and Procedures (2014) – See section 10.12.4.D).

[22] Perhaps the policy could require that an active link to the relevant government websites be provided in the disclaimer.

[23] Elsewhere, I, and others, have suggested that governments have an obligation to explain their actions. I have defended that proposition at length in Bernard W. Bell, Legislative History Without Legislative Intent: The Public Justification Approach to Statutory Interpretation, 60 OHIO STATE L.J. 1, 9-20 (1999).  There I use such an obligation to argue for the legitimacy of the use of some elements “legislative history” without relying on the concept of “legislative intent” that is so heavily criticized by textualists.

[24] In Costle, Judge Wald, writing for a three-judge panel, observed:

the importance to effective regulation of continuing contact with a regulated industry, other affected groups, and the public cannot be underestimated. Informal contacts may enable the agency to win needed support for its program, reduce future enforcement requirements by helping those regulated to anticipate and shape their plans for the future, and spur the provision of information which the agency needs.

[25] Granted, the Court arguably recognizes the temporal limitations on official speech in its school board president example, and quite properly notes the limits of a temporal analysis for officials who are continuously “on the clock.”

[26] In Lindke v. Freed, Justice Barrett observes that there are “approximately 20 million state and local government employees across the Nation, with an extraordinarily wide range of job descriptions—from Governors, mayors, and police chiefs to teachers, healthcare professionals, and transportation workers.”  Slip op. at 8.  But many neither head agencies, department, or units nor hold elective office. The contextual constraint discussed above would ordinarily suggest that many such officials’ content moderation decisions with respect to their individual social media sites should not be attributable to the government. That would make Free Speech Clause constraints inapplicable, no matter how closely related to the officials’ jobs the discussion on their sites are.

[27] Though Lindke, as an executive official, could presumably take some action by himself, the defendants in O’Connor-Radcliffe v. Garnier, members of a local school board, might only have been able to act collectively.

[28] Of course, this statement itself may be too simplistic.  Aside from assisting constituents in redressing their grievances with respect to administrative agencies’ decision (which does not require collective action), members can use their official position to influence private entities, sometimes with the threat of a legislative investigation in the background, without producing any legislation.

For example, in Kennedy v. Warren, 66 F.4th 1199 (9th Cir. 2023), U.S. Senator Elizabeth Warren sent a letter to Amazon accusing it of “peddling misinformation about COVID-19 vaccines and treatments” by using algorithms bringing such misleading materials to the top of its search results.  (The letter identified plaintiff Robert F. Kennedy Jr’s The Truth About COVID-19 as one such item.)  Senator Warren requested that Amazon “perform an immediate review of [its] algorithms and, within 14 days, provide both a public report on the extent to which Amazon’s algorithms are directing consumers to books and other products containing COVID-19 misinformation and a plan to modify these algorithms so that they no longer do so.”  

In another example, a series of letters from a House committee chair and subcommittee chair to pharmaceutical companies indicating the Committee’s plan to investigate insulin prices, the chairs sought information regarding pharmaceutical companies’ pricing decisions.  Letter from the Honorable Frank Pallone, Jr. and the Honorable Diane DiGetter to David A. Ricks, Chair and Chief Executive Officer Eli Lilly dated January 30, 2019.  In some circumstances, such a letter might well lead a recipient to change its practices.

Such efforts, and any discussion of such efforts by a legislator on his or her social media site and subsequent moderation of followers’ replies, might well be viewed as engaging in state action even if no legislative action takes place.

[29] See Gregory A. Mark, The Vestigial Constitution: The History and Significance of the Right to Petition, 66 Fordham L. Rev. 2153, 2161-62, 2176-79, 2199-2200, 2208-09 (1989).

[30] And indeed, one can be concerned about significantly unequal access to elected official to redress one’s grievances.  See, Sarah Ludington, et al., How Social Media Platforms Can Promote Compliance, supra note 3 (public officials’ use of moderation tools to block certain members of the public “effectively creates two classes of citizens—those who can communicate easily with public officials through social media and those who cannot”).  Granted, there is no requirement that elected officials be equally attentive to their constituents.  In fact, public officials will often speak to specific groups composed of people of a certain ideological viewpoint or certain interests, in settings where the more general public is not invited. 

[31] 1 MATTHEWS MUNICIPAL ORDINANCES §1.2 (“[e]lected local government officials live in the community and ordinarily are accessible to the public for requests, questions, and comments”)(available in westlaw database); Executive Order 13132—Federalism, §2(a)(Aug. 4. 1999), 64 Fed. Reg. 43255 (1999)(“[f]ederalism is rooted in the belief that issues that are not national in scope or significance are most appropriately addressed by the level of government closest to the people” (emphasis added)); Richard M. Nixon, Statement About the General Revenue Sharing Bill (Oct. 20, 1972)(announcing “revenue sharing” program to reverse the trend of power “flowing away from those levels of government which are closest to the people” and ensuring that on a range of matters “it will be local officials responding to local conditions and local constituencies who will decide what should happen” (emphasis added)).  Granted scholars have cast doubt on this oft-stated postulate. E.g., Donald R. Songer, Government Closest to the People: Constituent Knowledge in State & National Politics, 17 Polity 387 (1984).

[32] Gregory v. Ashcroft, 501 U.S. 452, 458 (1991)(“decentralized government … increases opportunity for citizen involvement in democratic processes”); Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528, 575 n.18 (1985)(Powell, J., dissenting) (“[t]he Framers recognized that the most effective democracy occurs at local levels of government, where people with firsthand knowledge of local problems have more ready access to public officials responsible for dealing with them[;] [t]his is as true today as it was when the Constitution was adopted”); Board of Education v. Pico, 457 U.S. 853, 894 (1982)(Powell, J., dissenting)(“[i]t is fair to say that no single agency of government at any level is closer to the people whom it serves than the typical school board”); Executive Order 13132—Federalism, supra, note 31, at §2(a); Statement About the General Revenue Sharing Bill, supra, note 31.

[33] Indeed O’Connor-Ratcliff and Zane argued that their social media sites were “personal campaign pages designed only to advance their own political careers.”  O’Connor-Ratcliff v. Garnier, 41 F.4th 1158, 1172 (9th Cir. 2022).  The panel acknowledged that many of the posts could serve dual purposes, acknowledging that “[m]any of the Trustees’ posts . . . concern[ed] workaday visits to schools and the achievements of PUSD’s students and teachers,” and that such material “could promote the Trustees’ personal campaign prospects.” Id. (emphasis added).  It rejected the argument because the Trustees’ posts about school activities “generally do not read as advertising ‘campaign promises’ kept or touting their own political achievements.”  Id.  Relying on judges to make such subtle distinctions post hoc based on the language used in such posts is problematic, and such an approach encourages government officials to feature their desire to be reelected more prominently in their communications (which should ot be viewed as a welcome development).

Print Friendly, PDF & Email