Amicus Curiae Brief — Michelle O’Connor-Ratcliff, et al., Petitioners v. Christopher Garnier, et ux.

NO. 22-324

IN THE

Supreme Court of the United States

MICHELLE O’CONNOR-RATCLIFF AND T.J. ZANE,

Petitioners,

v.

CHRISTOPHER GARNIER AND KIMBERLY GARNIER,

Respondents.

On Writ of Certiorari to the United States Court of

Appeals for the Ninth Circuit

BRIEF OF MANHATTAN INSTITUTE

AS AMICUS CURIAE

IN SUPPORT OF RESPONDENTS

ILYA SHAPIRO

MANHATTAN INSTITUTE

52 Vanderbilt Ave.

New York, NY 20017

(212) 599-7000

CORTNEY C. THOMAS

BRIAN E. ROBISON

RUSSELL T. BROWN

BROWN FOX, PLLC

8111 Preston Rd.

Suite 300

Dallas, TX 75225

(214) 327-5000

JOHN J. BURSCH

Counsel of Record

RYAN L. BANGERT

JEREMY D. TEDESCO

CHRISTOPHER P. SCHANDEVEL

MICHAEL R. ROSS

MATHEW W. HOFFMANN

ALLIANCE DEFENDING

FREEDOM

44180 Riverside Pkwy

Lansdowne, VA 20176

(616) 450-4235

jbursch@ADFlegal.org

Counsel for Amicus Curiae

i

TABLE OF CONTENTS

TABLE OF AUTHORITIES ..................................... iii

INTEREST OF AMICUS CURIAE ........................... 1

SUMMARY OF THE ARGUMENT ........................... 2

ARGUMENT .............................................................. 4

I.

This Court distinguishes public officials’

public and private actions by looking to the

purpose and appearance of their conduct. .......... 4

II. The purpose and appearance of a public

official’s social-media account determines

state action. .......................................................... 8

A. Official social-media accounts are state

action. ............................................................ 9

B. Personal social-media accounts where

no official business is conducted are

just that: personal, non-state action........... 11

C. Election campaign profiles typically

remain personal but can become official

accounts. ...................................................... 14

D. Courts should assess mixed-use

accounts post by post................................... 16

ii

III. If a court finds state action, standard First

Amendment forum analysis applies to any

government restrictions on speech. .................. 20

A. Forum analysis preserves free and

uninhibited debate in the public

square. ......................................................... 21

B. Public officials’ activity on social media

can create either designated or limited

public forums. .............................................. 24

C. Forum analysis protects public debate

while preserving officials’ control over

their personal accounts and protecting

their own First Amendment freedoms. ...... 26

CONCLUSION ......................................................... 29

iii

TABLE OF AUTHORITIES

Cases

Arkansas Education Television Commission v.

Forbes,

523 U.S. 666 (1998)...................................... 23, 25

Biden v. Knight First Amendment Institute,

141 S. Ct. 1220 (2021).......................................... 9

Branti v. Finkel,

445 U.S. 507 (1980).............................................. 7

Brentwood Academy v. Tennessee Secondary

School Athletic Association,

531 U.S. 288 (2001).............................................. 6

Campbell v. Reisch,

986 F.3d 822 (8th Cir. 2021) ................... 9, 13–16

Charudattan v. Darnell,

834 F. App’x 477 (11th Cir. 2020) ............... 14–15

City of Lakewood v. Plain Dealer Publishing Co.,

486 U.S. 750 (1988)............................................ 24

Cohen v. California,

403 U.S. 15 (1971)................................................ 1

Cornelius v. NAACP Legal Defense & Education

Fund, Inc.,

473 U.S. 788 (1985)...................................... 23, 25

Davison v. Plowman,

247 F. Supp. 3d 767 (E.D. Va. 2017) ................. 25

Davison v. Plowman,

715 F. App’x 298 (4th Cir. 2018) ....................... 25

Davison v. Randall,

912 F.3d 666 (4th Cir. 2019) ............................. 10

iv

Felts v. Vollmer,

No. 4:20-cv-00821, 2022 WL 17546996 (E.D.

Mo. Dec. 9, 2022).................................................. 9

Filarsky v. Delia,

566 U.S. 377 (2012).............................................. 5

Garnier v. O’Connor-Ratcliff,

41 F.4th 1158 (9th Cir. 2022) .......9–15, 25–26, 28

Good News Club v. Milford Central School,

533 U.S. 98 (2001).............................................. 22

Griffin v. Maryland,

378 U.S. 130 (1964).......................................... 6–7

Hague v. Committee for Industrial Organization,

307 U.S. 496 (1939)............................................ 21

Iancu v. Brunetti,

139 S. Ct. 2294 (2019)........................................ 23

International Society for Krishna Consciousness,

Inc. v. Lee,

505 U.S. 672 (1992)............................................ 23

Knight First Amendment Institute v. Trump,

928 F.3d 226 (2d Cir. 2019) ..................... 9–10, 24

Lindke v. Freed,

143 S. Ct. 1780 (2023).......................................... 5

Lindke v. Freed,

37 F.4th 1199 (6th Cir. 2022) ............... 5–6, 9–10,

12–13, 16–18, 20, 26–27

Lindke v. Freed,

563 F. Supp. 3d 704 (E.D. Mich. 2021) ............. 27

Lugar v. Edmondson Oil Co.,

457 U.S. 922 (1982).......................................... 5–6

v

Manhattan Community Access Corp. v. Halleck,

139 S. Ct. 1921 (2019)........................................ 20

Packingham v. North Carolina,

582 U.S. 98 (2017).........................1, 12–13, 21, 24

Perry Education Association v. Perry Local

Educators’ Association,

460 U.S. 37 (1983).................................. 21–23, 25

Pleasant Grove City v. Summum,

555 U.S. 460 (2009)...................................... 23, 27

Polk County v. Dodson,

454 U.S. 312 (1981)...................6–7, 10, 16–17, 20

Rankin v. McPherson,

483 U.S. 378 (1987)........................................ 2, 13

Rosenberger v. Rector & Visitors of University of

Virginia,

515 U.S. 819 (1995)...................................... 22–24

Shuttlesworth v. City of Birmingham,

394 U.S. 147 (1969)............................................ 24

United States v. Classic,

313 U.S. 299 (1941).............................................. 6

Ward v. Rock Against Racism,

491 U.S. 781 (1989)............................................ 27

West v. Atkins,

487 U.S. 42 (1988).............................................. 20

Statutes

42 U.S.C. 1983 ........................................................ 2, 5

vi

Other Authorities

Abraham Lincoln, Gettysburg Address (Nov. 19,

1863) ..................................................................... 4

Facebook, Control who can see posts on your

Facebook timeline .............................................. 19

Justice Department,

https://twitter.com/TheJusticeDept .................... 9

Constitutional Provisions

U.S. Const. amend. I .................................................. 5

1

INTEREST OF AMICUS CURIAE 1

The Manhattan Institute (MI) is a nonprofit

public-policy research foundation whose mission is to

develop and disseminate new ideas that foster economic choice and individual responsibility. To that end,

it has historically sponsored scholarship supporting

the rule of law and opposing government overreach,

including in the marketplace of ideas.

MI understands that, in our digital age, so much

of the public discourse it seeks to enrich and influence

occurs on social media. “Social media allows users to

gain access to information and communicate with one

another about it on any subject that might come to

mind.” Packingham v. North Carolina, 582 U.S. 98,

107 (2017). Social media presents the “principal

sources for knowing current events . . . and otherwise

exploring the vast realms of human thought and

knowledge.” Id. Indeed, social-media platforms offer

the “most powerful mechanisms available to a private

citizen to make his or her voice heard.” Id. They are

“the modern public square.” Id.

And just like in the traditional public square,

government cannot pick winners and losers on social

media. The First Amendment “put[s] the decision as

to what views shall be voiced” where it should be—

“into the hands of each of us.” Cohen v. California, 403

U.S. 15, 24 (1971). Indeed, “no other approach would

comport with the premise of individual dignity and

choice upon which our political system rests.” Id. That

same logic applies with equal force when government

1 No counsel for a party authored this brief in whole or in part,

and no person other than amicus and its counsel made any monetary contribution intended to fund the preparation or submission of this brief.

2

officials censor speech while hiding behind the façade

of private action. Officials cannot imbue their socialmedia profiles with the trappings of their offices and

use them to communicate with their constituents, but

then disclaim liability when they censor views they

don’t like. When state officials miss the mark, 42

U.S.C. § 1983 allows injured citizens to seek redress.

An overly constrained notion of state action would

frustrate both the purposes of Section 1983 and MI’s

work to promote uninhibited debate on important

issues. Government officials certainly retain their

ability to speak on social media in their personal

capacities, but they must be held accountable for

preventing citizens from engaging them on issues of

public debate that they raise on those accounts.

SUMMARY OF THE ARGUMENT

Public discourse has moved online. Social media

now serves many of the same functions traditionally

filled by parks, squares, and sidewalks. Government

officials post about their work and interact with their

constituents on issues ranging from the local—like

preparing for a blizzard—to the national—like the

nomination of a new FBI director.

This new online forum does not exempt government officials from the traditional safeguards that

protect our public debate. The First Amendment

ensures that expression vital to our representative

democracy remains “uninhibited, robust, and wideopen.” Rankin v. McPherson, 483 U.S. 378, 387 (1987).

That means the government violates the Constitution

when it censors viewpoints online no less than when

it does so on the street.

3

But public officials are citizens with their own

First Amendment rights; they can only violate the

Constitution (and be held liable under Section 1983)

when they act under color of state law. This Court’s

precedents chart the line between a public official’s

personal and public conduct by examining its purpose

and appearance. Public-official conduct with a public

purpose—such as discussing city pandemic policies—

supports a state-action finding. And so does conduct

clothed with state authority—such as a social-media

profile that identifies its owner as a government

official and provides government contact information.

The purpose-and-appearance test properly holds

government officials responsible for actions attributable to the state while preserving personal freedoms.

And on either end of the spectrum, the inquiry is

straightforward. Official, state-action social-media

profiles will operate under an official title; focus on

government work; be open for public comment;

typically use state resources; and may transfer from

one officeholder to the next. Personal, non-stateaction social-media accounts will remain personal.

They will operate under an individual’s name and

remain under the individual’s control; are typically

not accessible by the general public; focus on personal

matters; and do not employ state resources.

Campaign and mixed-use accounts fall in between

these two ends. Campaign pages generally remain

personal by focusing on campaign business and not

including official titles or posts on behalf of government bodies or officeholders. But campaign pages and

other mixed-use accounts blending personal and official business trigger purpose-and-appearance scrutiny on a post-by-post basis to balance public accountability with the free-speech rights of all speakers.

4

A state-action finding triggers First Amendment

forum analysis. Just like in the traditional public

square, discourse happens in the interactive portions

of government officials’ social-media pages. These

interactive spaces are publicly accessible and invite

constituents to comment on government matters. By

opening these areas to public comment, the government makes them either designated or limited public

forums. That means—at the very least—that the

government cannot pick winners and losers in the

marketplace of ideas by discriminating based on

viewpoint. But the government also does not lose all

control over the forums it creates. It may impose

reasonable restrictions on public discourse consistent

with the purpose of the individual social-media forum.

This Court should apply its purpose-and-appearance test for state action and affirm the judgment of

the Ninth Circuit. That test hews closely to precedent

and shows which actions can be fairly attributed to

the state. It provides the twin benefits of holding

public officials accountable for online censorship that

is fairly attributable to the government while preserving their freedom to voice their own personal views.

And it ensures that dialogue in the modern public

square remains exactly as it should be: uninhibited,

robust, and equally open to all.

ARGUMENT

I.

This Court distinguishes public officials’

public and private actions by looking to the

purpose and appearance of their conduct.

Our founders created a “government of the people,

by the people, [and] for the people.” Abraham Lincoln,

Gettysburg Address (Nov. 19, 1863). Government

5

officials—no less than private citizens—are part of

the “people.” And that doesn’t change when they

assume public office. Just like the rest of us, government officials still have personal pursuits. They

“visit[ ] the hardware store,” “chat[ ] with neighbors,”

and “attend[ ] church services.” Lindke v. Freed, 37

F.4th 1199, 1205 (6th Cir. 2022), cert. granted, 143 S.

Ct. 1780 (2023).

But by assuming public office, they also assume

public responsibilities. On a fundamental level, that

means they must follow the Constitution, including

the First Amendment’s prohibition against “abridging

the freedom of speech.” U.S. Const. amend. I. And

when they fall short, the people have the power to

hold them accountable for any actions taken “under

color of ” state law. 42 U.S.C. § 1983.

But because these officials do not shed their status

as members of the “people” when they take office,

courts must decide when their conduct “is fairly attributable to the State.” Filarsky v. Delia, 566 U.S. 377,

383 (2012) (cleaned up). Only then can courts decide

whether their conduct violated the Constitution.

In this way, the state-action doctrine balances the

need to hold government actors accountable for their

constitutional violations while still “preserv[ing] an

area of individual freedom by limiting the reach of

federal law.” Lugar v. Edmondson Oil Co., 457 U.S.

922, 936 (1982).

This Court’s decisions separating the public

actions of government officials from what those same

officials do as private citizens provides the proper

framework for deciding this case and its companion.

Cases examining when private actors can become

state actors may shed some light on the inquiry. E.g.,

6

Brentwood Acad. v. Tenn. Secondary Sch. Athletic

Ass’n, 531 U.S. 288, 295–96 (2001). But when state

officials are involved, the analysis changes. Contra

Lindke, 37 F.4th at 1203 (conflating the “state-official

test” with the “nexus test” traditionally applied to

private actors). Government officials are presumed to

be state actors: “state employment is generally sufficient to render the defendant a state actor under our

analysis.” Lugar, 457 U.S. at 935 n.18. Private actors

receive the opposite presumption: a “private party”

lacks the “apparent authority” of “the weight of the

State.” Id. at 937.

For that reason, a public-versus-private-property

approach—one that merely looks to who owns the

account, the government or a private citizen—

oversimplifies the analysis and misses the point of the

state-action doctrine. Contra Pets.’ Br. 23–24; SG Br.

20. When a public official “purports to act” under state

authority, he takes “state action.” Griffin v. Maryland, 378 U.S. 130, 135 (1964). “It is irrelevant that

he might have taken the same action had he acted in

a purely private capacity.” Id.

The proper inquiry asks whether the official

(1) exercised “power . . . by virtue of state law,” and

(2) was “clothed with the authority of state law.”

United States v. Classic, 313 U.S. 299, 326 (1941). In

other words, to distinguish between an official’s

public and private actions, the Court examines (1) the

conduct’s purpose (whether exercising power given by

state law to achieve a public end), and (2) its

appearance (whether acting while clothed with the

authority of state law). See Polk Cnty. v. Dodson, 454

U.S. 312, 318 (1981); Griffin, 378 U.S. at 135. And

that analysis requires a fact-specific inquiry. Griffin,

378 U.S. at 135.

7

As to purpose, a public official’s conduct will not

meet the state-action threshold when it serves “essentially a private function, traditionally filled by [a

private actor], for which state office and authority are

not needed.” Polk Cnty., 454 U.S. at 319. For example,

a public defender serves a private purpose while

representing her client because she “works under

canons of professional responsibility that mandate

[her] exercise of independent judgment on behalf of

the client” and remains “free of state control” in that

representation. Id. at 321–22. But when the same

public defender makes “hiring and firing decisions on

behalf of the State,” she serves a public purpose and

thus is a state actor. Id. at 325 (summarizing Branti

v. Finkel, 445 U.S. 507 (1980)).

As for appearance, when an official “possessed of

state authority . . . purports to act under that authority, his action is state action.” Griffin, 378 U.S. at 135

(emphasis added). Thus, this Court has recognized as

a state actor a deputy sheriff who “wore a sheriff ’s

badge and consistently identified himself as a deputy

sheriff rather than as an employee of the [private]

park” where he worked when he ordered civil-rights

protesters to leave the park and then arrested and

initiated prosecutions against them. Id. In that case,

statements in the amended warrant indicating that

the sheriff had acted as “an ‘agent’ of the park” had

“little, if any, bearing on the character of the

authority” the sheriff had “initially purported to

exercise.” Id. What mattered was how his authority

appeared to those he arrested.

8

II. The purpose and appearance of a public

official’s social-media account determines

state action.

On either end of the spectrum, the purpose-andappearance test yields determinate results. Official

social-media accounts bear the government body’s or

official’s title, focus exclusively on government work,

are open for public comment, typically use state

resources, and may transfer from one officeholder to

the next. They are therefore fairly attributable to the

state. Conversely, personal accounts are listed under

an individual’s name rather than his or her official

title, remain under the individual’s control even after

his or her term has expired, are often not publicly

accessible, focus almost exclusively on personal

matters, and do not use state resources. They are not

state action.

In the middle lie campaign and mixed-use

accounts. A campaign account will generally remain

personal because it exists to support personal election

efforts. But it may become a conduit of state action if,

while in office, the account adopts the characteristics

of an official account. Courts presented with a mixeduse account blending personal communications and

official business should employ a granular post-bypost approach. The granular approach comports with

precedent examining the specific function of the government official at issue, ensures that the state-action

doctrine does not swallow an official’s individual freedom, and prevents officials from circumventing the

First Amendment by occasionally using a personal

account to facilitate a public debate about government

issues while censoring disfavored views.

9

A. Official social-media accounts are state

action.

When a government body or public official uses an

official social-media account, the account’s purpose

and appearance prove state action. These accounts

(1) use the government body’s name or official’s title;

(2) focus on government work; (3) are open for public

comment; (4) typically use state resources; and

(5) may transfer from one officeholder to the next.

First, official profiles operate under the name of

the government body, like the Department of Justice, 2

or the officeholder’s title, see Lindke, 37 F.4th at 1204

(“official Facebook account for the Governor of Kentucky titled @KentuckyGovernor” is state action). It

may even be identified as an “official” account.3

Second, these profiles focus on the body or official’s

work and thus become “an organ of official business.”

Campbell v. Reisch, 986 F.3d 822, 826 (8th Cir. 2021).

They “announce matters related to official government business”—like policy changes or judicial

nominees. Knight First Amend. Inst. v. Trump, 928

F.3d 226, 236 (2d Cir. 2019), vacated as moot sub nom

Biden v. Knight First Amend. Inst., 141 S. Ct. 1220

(2021). And they may serve “as a channel for communicating and interacting with the public about” the

government’s work. Id. at 235; accord Garnier v.

O’Connor-Ratcliff, 41 F.4th 1158, 1171 (9th Cir.

2022); Felts v. Vollmer, No. 4:20-cv-00821, 2022 WL

17546996, at *9 (E.D. Mo. Dec. 9, 2022) (finding state

action in account posting government links and other

“official activities”).

2 Justice Department, https://twitter.com/TheJusticeDept.

3 Id. (“Official DOJ Twitter account”).

10

Third, official accounts allow the general public to

view and comment on the content they post. The

ability of constituents to comment on posts factors

into the forum analysis, infra Part III, but it also

indicates an official profile. Government officials use

public social-media accounts to “provide[ ] information

to the public” and “solicit[ ] input from the public on

policy issues.” Davison v. Randall, 912 F.3d 666, 680

(4th Cir. 2019). Official accounts appear to be—and

are—“official channels of communication with the

public about the work” of the pertinent government

actor. Garnier, 41 F.4th at 1171.

Fourth, official accounts are often run by government employees at the government’s expense. A

“tech-savvy governor[’s]” use of the “state’s payroll” to

hire “a social-media team to manage her online

presence” supports the conclusion that such presence

is state action. Lindke, 37 F.4th at 1204.

Fifth, these accounts may transfer from one

officeholder to the next. For example, the “@POTUS”

and “@WhiteHouse” accounts “are official government

accounts” such that “the President and members of

the White House administration will not retain

control over those accounts upon leaving office.”

Knight, 928 F.3d at 235 n.6. Such accounts further the

public purpose of the office.

In sum, governments use official accounts for

official business—not “private functions.” Polk Cnty.,

454 U.S. at 319. The purpose and appearance of these

accounts, as revealed by the above factors, show that

they exercise power under state law and are clothed

in state authority, making them “fairly attributable”

to the state.

11

Applying these factors, the Ninth Circuit correctly

held the social-media profiles here to be state action.

The officials “identified themselves on their Facebook

pages as ‘government official[s],’ [and] listed their

official titles in prominent places on both their

Facebook and Twitter pages.” Garnier, 41 F.4th at

1171. One published her government email address,

and the other, a school-district board member,

identified his Facebook page as “the official page” for

him “to promote public and political information.” Id.

Both officials “regularly posted about school board

meetings, surveys related to school district policy

decisions, the superintendent hiring process, budget

planning, and public safety issues.” Id. And the

officials separated these publicly accessible accounts

from “their private Facebook pages.” Id. at 1163. As

the Ninth Circuit put it, “the pertinent factors all

indicate that [the officials] unequivocally cloaked

their social media accounts with the authority of the

state.” Id. at 1173 (cleaned up).

B. Personal social-media accounts where no

official business is conducted are just

that: personal, non-state action.

Personal social-media accounts reside on the

opposite end of the spectrum. These accounts

(1) operate under an individual’s name and remain

under the individual’s control even after his or her

term has expired; (2) are typically not accessible by

the general public; (3) focus almost exclusively on

personal matters; and (4) do not employ state

resources. Generally speaking, then, they would not

cross the line into the domain of state action.

12

First, a personal account is in an individual’s

name and stays with the individual. “[A] Facebook

page called @JohnDoe” evinces that it “belongs to

Doe-the-citizen—not Doe-the-governor.” Lindke, 37

F.4th at 1204. “That page will belong to Doe even after

he leaves office—it’s his, not the governorship’s.” Id.

These accounts do not prominently identify the holder

as a government official, nor do they provide government email addresses or website information. See

Garnier, 41 F.4th at 1171 (official social-media

profiles identified state actors as “‘government

official[s],’ listed their official titles in prominent

places on both their Facebook and Twitter pages,

and, . . . included [an] official . . . email address”).

Second, a personal social-media account often will

restrict public access. An account can be “private,”

meaning only those whom the accountholder has

“shared” it with will have access. See Garnier, 41

F.4th at 1163. A private account indicates the owner

has reserved it for “family and friends”—not government business. Id. What’s more, the existence of both

a private account and a publicly accessible account for

an official sharpens the divide between personal and

public function. There is little reason to have two

separate accounts—unless one exists for the government official as an official. See id. (government

officials had “private Facebook pages” apart from

their official social-media profiles).

Third, personal accounts focus on personal—not

government—pursuits. When they use the “modern

public square” to “gain access to information and

communicate with one another about it on any subject

that might come to mind,” government officials

engage in private activities. Packingham, 582 U.S. at

107. Just like everyone else, officials can use social

13

media to “debate religion and politics with their

friends and neighbors or share vacation photos.” Id.

at 104. And mere job talk about a government position

does not create state action. There is a dispositive

difference between “self-promotional” talk about a

recent raise or a discussion of a position’s long hours

and using social media “to communicate about

. . . official duties.” See Garnier, 41 F.4th at 1172. The

former reflects personal concerns while the latter

holds the “social media pages out to be official

channels of communication with the public about the

work of the [official].” Id. at 1171. An official account

focuses on the official business of the government

office or position and promotes the public office and

public interests more generally. Id. at 1172. Whereas

a personal account concerns and promotes the

individual’s “career[ ].” Id.

Fourth, personal accounts will not include those

run by state-funded social-media teams, nor those

funded by government resources. Rather, they will

remain controlled and curated by the individual.

These factors separate personal conduct from

state action, making accounts “personal and free from

scrutiny under section 1983.” Lindke, 37 F.4th at

1204. They also fortify individual liberty. Most government workers do not interact with constituents or

the public at large, so they do not run their accounts

as “organ[s] of official business.” Campbell, 986 F.3d

at 826. And that is especially true for lower-level

employees. Cf. Rankin, 483 U.S. at 390–91 (When “an

employee serves no confidential, policymaking, or

public contact role, the danger to the agency’s

successful functioning from that employee’s private

speech is minimal.”). Meanwhile, public officials can

easily operate separate personal and public accounts.

14

C. Election campaign profiles typically

remain personal but can become official

accounts.

Private citizens campaigning for office have personal social-media accounts. “[I]t seems safe to say

that someone who isn’t a public official cannot create

an official governmental account.” Campbell, 986 F.3d

at 826. And being elected doesn’t “magically alter [an]

account’s character.” Id. An official can keep interacting with her campaign while in office “to promote

herself and position herself for more electoral success

down the road.” E.g., id. But as this case proves, a

campaign account can “evolve into something

different . . . if it becomes an organ of official

business,” Campbell, 986 F.3d at 826; accord Garnier,

41 F.4th at 1172 (state action when “[a]fter their

election,” officials “virtually never posted overtly

political or self-promotional material,” instead

focusing on “official District business or promot[ing]

the District generally”). Once again, purpose and

appearance will be dispositive.

An account focused on campaign-related topics—

even if it belongs to an elected official—is private

conduct. Campaign pages share many of the same

qualities as purely personal accounts. To remain

personal, they should not use official titles or post on

behalf of government bodies or officeholders. Charudattan v. Darnell, 834 F. App’x 477, 482 (11th Cir.

2020) (per curiam). Instead, they should discuss the

campaign, publish photos of campaign events, tout

endorsements, present the candidate’s background

and philosophy, and seek donations and other

support. See id. Any discussion of official work should

be for campaign purposes: “to create a favorable

impression of [her] in the minds of her constituents.”

15

Campbell, 986 F.3d at 827. In sum, an official should

use a campaign account “to convince her audience to

support her election bid.” Id. at 826.

Campaign accounts are likely to be publicly accessible and discuss an official’s work. See id. But those

factors do not by themselves transform a private

campaign page into an official social-media account.

To avoid confusion, elected officials should clearly

describe their purpose in using the page. One way is

to use a “disclaimer that the statements made on this

web site reflect the personal opinions of the author

and are not made in any official capacity.” Garnier, 41

F.4th at 1172 (cleaned up); see also Charudattan, 834

F. App’x at 479. A disclaimer cannot override other

purpose-and-appearance factors that make an

account state action, but it can show to the public the

official’s private campaign purpose.

Officials also can implicitly separate their campaign pages from their duties. A campaign page “does

not convert itself into an official page just because the

candidate chooses a handle that reflects the office she

is pursuing” or because she “posts a photo of herself

working at the job she was elected to perform and

hopes to be elected to perform again.” Campbell, 986

F.3d at 827. So a candidate can post about her work

“to create a favorable impression of [her] in the minds

of her constituents.” Id. But such posts should not

focus on official business, such as “announcing an

appointee” or “coordinating [the] county’s response to

a blizzard.” Id. The page should instead focus on

campaign topics like “provid[ing] information on [the]

local political party’s annual chili supper and Lincoln

Day banquet.” Id..

16

D. Courts should assess mixed-use accounts

post by post.

Mixed-use profiles present the most difficult

scenario for a state-action analysis. Officials may use

their accounts to “feature[ ] a medley of posts” ranging

from family photos to “administrative directives”

issued as a government official. Lindke, 37 F.4th at

1201. Such accounts may also identify the account

holder as “Daddy,” “Husband,” and “City Manager.”

E.g., id. When courts are presented with such profiles,

they should employ a more granular approach by

assessing each post individually within the context of

the page as a whole.

The dueling private and public nature of

individual posts can make evidence of personal or

official control more “equivocal.” Campbell, 986 F.3d

at 827. A public official may be engaging in state

action by publishing certain posts, by engaging with

constituents on those posts, and by limiting which of

his constituents can engage with those posts and how.

But the same official might not be engaging in state

action with respect to other, more overtly personal

posts, and thus might be entitled to greater control in

limiting access to them. A post-by-post approach

ensures proper respect for an official’s individual

liberty while not granting license to circumvent

constitutional guarantees.

Precedent supports the post-by-post analysis. The

state-employee status of a public defender does not

establish per se state action. Polk Cnty., 454 U.S. at

324–25. A public defender who “exercis[es] her independent professional judgment in a criminal proceeding” does not act under color of state law. Id. at 324.

But one who makes personnel decisions on the state’s

17

behalf does, as may one who “perform[s] certain

administrative and possibly investigative functions.”

Id. at 325. That is, the public-employee status of the

public defender does not determine the state-action

requirement in all circumstances. Instead, the Court

looks to the purpose and appearance of the particular

action taken by the public defender.

Take Lindke. The city-manager defendant—

Freed—mixed personal and official business posts.

Lindke, 37 F.4th at 1201. The plaintiff—Lindke—

criticized Freed’s posts about city pandemic policies,

which caused Freed to delete the critical comments

and block Lindke from his page. Id. at 1201–02.

Freed’s page was public, he identified himself as a

“public figure” and by his official title, and he listed

the city’s website and contact information. Id. at 1201.

After the district court granted Freed’s motion for

summary judgment and Lindke appealed, the Sixth

Circuit tried to bring “the clarity of bright lines” to the

case, but in the process, the court inadvertently

blurred those lines. See id. at 1207. Essentially

summarizing “a version of the Supreme Court’s nexus

text,” the court stated that, when “analyzing socialmedia activity, [it] look[s] to a page or account as a

whole, not each individual post.” Id. at 1203.

Despite emphasizing the importance of “the

context of the entire page,” though, the court ultimately declined to consider either Freed’s account as a

whole or his individual posts. Id. Quite the opposite,

“[i]nstead of examining a page’s appearance or

purpose,” the court “focus[ed] on the actor’s official

duties and use of government resources or state

employees.” Id. at 1206 (emphasis added). And

because the court found that “Freed did not operate

18

his page to fulfill any actual or apparent duty of his

office,” or “use his government authority to maintain

it,” the court held that “he was acting in his personal

capacity—and there was no state action.” Id. at 1207.

Adopting such a narrow duty-or-authority test

would make it far too easy for government officials to

flout the First Amendment rights of their constituents, as the other circuits seem to have recognized.

See id. (“part[ing] ways with other circuits’ approach

to state action in this novel circumstance”). It would

allow government officials to maintain mixed-use

accounts with any appearance or purpose, to post any

content to them, and to censor any speech on them

however they saw fit—provided they stop short of

“fulfill[ing] any actual or apparent duty of [their]

office” or using their “governmental authority to

maintain” the accounts. Id. at 1207. Such a deferential standard would encourage public officials to

eschew official social-media accounts in favor of

mixed-use accounts if they can ensure—regardless of

how they appear—that such accounts are not

required by state law, do not use state resources, do

not arise from state authority, and do not use state

staff. See id. at 1203–04.

Assessing the general purpose and appearance of

the account as a whole without considering individual

posts likewise falls short. A public official could still

sprinkle in official content alongside private content

and censor speech in response to the official content

so long as the account stays mainly for personal use.

Having insulated their online presence from the Constitution’s oversight, public officials would be unaccountable for how they choose to limit or prohibit their

constituents from engaging with them online.

19

Conversely, a post-by-post approach solves those

problems by reasonably balancing an official’s

interest in regulating discussion of his or her personal

affairs with the public’s interest in speaking freely on

issues of public concern. If the Sixth Circuit in Lindke

had reviewed the district court’s summary-judgment

order under such a test, it could have reached a

clearer and better result. While the court should have

started its analysis by considering the overall purpose

and appearance of Freed’s once-private, now-public

Facebook account (which the court refused to do), it

also should have considered the purpose and appearance of each post that Freed prohibited Lindke from

engaging with after Freed blocked him from his page. 4

The purpose and appearance of Freed’s pandemicpolicy posts—to inform constituents about licit and

illicit conduct—considered within the context of the

page’s governmental identifications, made those posts

state action. Freed made those posts from his position

possessed by virtue of state law, clothed in the

authority of that position, and with the purpose of

communicating with his constituents. So those posts

were state action. And the same goes for his attempt

to limit access to them.

That doesn’t foreclose Freed’s ability to control his

personal postings on the same page. He can still limit

who can view his family photos and personal updates. 5 And he retains his “rights to exercise editorial

4 Under this approach, plaintiffs would have the burden of ident-

ifying posts they contend are state action. Courts would not be

required to sift through every post to carry that burden for them.

5 For example, Facebook lets users tailor the audience for specific

posts. See Facebook, Control who can see posts on your Facebook

timeline, https://www.facebook.com/help/246629975377810.

20

control over speech and speakers on” his personal

posts. Manhattan Cmty. Access Corp. v. Halleck, 139

S. Ct. 1921, 1932 (2019).

For mixed-use accounts, that approach works

better than “look[ing] to a page or account as a whole.”

Lindke, 37 F.4th at 1203. It’s also more consistent

with precedent. See, e.g., Polk Cnty., 454 U.S. at 324–

25; West v. Atkins, 487 U.S. 42, 56 (1988). If the Sixth

Circuit had followed this approach in Lindke, it

could—and should—have found that Freed’s efforts to

block access to his posts in which he was performing

part of his official role were state action.

Setting a clear purpose-and-appearance rule has

the benefit of reducing mixed-use profiles. It will

incentivize government officials to keep separate

personal and official accounts, which will protect both

them and the public. And it will encourage the use of

disclaimers and other clear indicators of private use

by public officials to separate campaigning from governing. A clear rule will allow officials to retain their

individual liberty to voice their personal views while

not depriving the public of the opportunity to participate in public debate. And that’s exactly what the

state-action doctrine should accomplish.

III. If a court finds state action, standard First

Amendment forum analysis applies to any

government restrictions on speech.

Once a court determines that a government

official’s social-media activity constitutes state action,

standard First Amendment public-forum principles

will apply to any resulting restrictions on speech. This

doctrine is well-suited to govern public officials’

social-media activity under this Court’s precedent. It

21

sets clear guardrails that would prevent official

government accounts from discriminatorily censoring

private citizens’ online speech. And it gives public

officials some flexibility to place reasonable restrictions on the way the public interacts with their

accounts, consistent with the account’s purpose,

provided they do not single out disfavored viewpoints

on issues of public concern.

A. Forum analysis preserves free and

uninhibited debate in the public square.

The public-forum doctrine “sharply circumscribe[s]” the “rights of the state to limit expressive

activity” and applies “[i]n places which by long

tradition or by government fiat have been devoted to

assembly and debate.” Perry Educ. Ass’n v. Perry

Local Educators’ Ass’n, 460 U.S. 37, 45 (1983). The

doctrine first developed in recognition of the special

role that streets and parks have held as places for

speech, assembly, and public debate. See Hague v.

Comm. for Indus. Org., 307 U.S. 496, 515 (1939).

Today, the internet and social media have taken

on a similar role as streets and parks, serving as significant hubs for public expression and debate. “While

in the past there may have been difficulty in identifying the most important places (in a spatial sense) for

the exchange of views, today the answer is clear.”

Packingham, 582 U.S. at 104. “It is cyberspace—the

vast democratic forums of the Internet in general, and

social media in particular.” Id. (cleaned up).

Forum analysis links the character of the place

where expression occurs to the permissible level of

government regulation.

22

1. For traditional public forums, such as streets

and parks, “the government may not prohibit all

communicative activity,” meaning that any contentbased restrictions must survive strict scrutiny. Perry,

460 U.S. at 45. But the government may enforce

“content-neutral [and] narrowly tailored” regulations

of the “time, place, and manner of expression,” id.,

prohibiting, for example, loud events in residential

areas provided that the regulations apply equally no

matter the event’s content or its viewpoint.

2. The public-forum doctrine does not end with the

“traditional” public square—the government may also

establish designated public forums, areas that “the

state has opened for use by the public as a place for

expressive activity.” Id. The key difference from

traditional public forums is that the government need

not create and need not keep open designated public

forums. Otherwise, the rules governing speech

restrictions for designated public forums are the

same: “Reasonable time, place and manner regulations are permissible, and a content-based prohibition

must be narrowly drawn to effectuate a compelling

state interest.” Id. at 46.

3. The government may also create limited public

forums by “reserving [a forum] for certain groups or

for the discussion of certain topics.” Rosenberger v.

Rector & Visitors of Univ. of Va., 515 U.S. 819, 829

(1995); accord, e.g., Perry, 460 U.S. at 46 n.7. When it

does, restrictions on the forum must be (1) reasonable,

and (2) viewpoint neutral. Good News Club v. Milford

Cent. Sch., 533 U.S. 98, 106–07 (2001).

4. Finally, nonpublic forums are “[p]ublic property

which . . . by tradition or designation” have not been

“forum[s] for public communication.” Perry, 460 U.S.

23

at 46. In a nonpublic forum, the government can have

a “selective access” policy in which “individual nonministerial judgments” govern forum participation,

again subject to the same two limitations: any policy

must be (1) reasonable, and (2) viewpoint neutral.

Ark. Educ. Television Comm’n v. Forbes, 523 U.S. 666,

680 (1998); accord, e.g., Perry, 460 U.S. at 46.

The government may not discriminate based on

viewpoint—in any forum. Int’l Soc’y for Krishna Consciousness, Inc. v. Lee, 505 U.S. 672, 679 (1992);

accord Pleasant Grove City v. Summum, 555 U.S. 460,

469–70 (2009) (requiring viewpoint neutrality for

traditional, designated, and limited public forums);

Cornelius v. NAACP Legal Def. & Educ. Fund, Inc.,

473 U.S. 788, 806 (1985) (same for nonpublic forums).

Viewpoint discrimination occurs “[w]hen the government targets not subject matter, but particular views

taken by speakers on a subject.” Rosenberger, 515

U.S. at 829. This Court has repeatedly condemned

viewpoint discrimination as “an egregious form of

content discrimination,” id., and “poison to a free

society,” Iancu v. Brunetti, 139 S. Ct. 2294, 2302

(2019) (Alito, J., concurring). Thus, “[t]he government

must abstain from regulating speech when the

specific motivating ideology or the opinion or

perspective of the speaker is the rationale for the

restriction.” Rosenberger, 515 U.S. at 829.

The prohibition on viewpoint discrimination also

means public officials cannot be granted unbridled

discretion to censor speech in any forum. This Court

“consistently condemn[s]” speech regulations that

“vest in an administrative official discretion to grant

or withhold a permit based upon broad criteria

unrelated to proper regulation of public places.”

Shuttlesworth v. City of Birmingham, 394 U.S. 147,

24

153 (1969). With vague or non-existent criteria on

which to make their decisions, government officials

“may decide who may speak and who may not based

upon the content of the speech or viewpoint of the

speaker.” City of Lakewood v. Plain Dealer Publ’g Co.,

486 U.S. 750, 763–64 (1988). Speech restrictions

therefore must contain “narrow, objective, and definite standards to guide” officials. Shuttlesworth, 394

U.S. at 150–51. In the social-media context, as in any

context, government officials must employ policies

with “narrow, objective, and definite standards” when

those policies result in restrictions on speech.

B. Public officials’ activity on social media

can create either designated or limited

public forums.

After a court determines that a government

official’s social-media activity constitutes state action,

the court still must determine whether the official has

created a “forum” for speech or whether the communication is truly one-way and limited to government

speech. When the public is given the ability to interact

with a government-controlled account, the account

becomes a forum for speech by its very nature. See,

e.g., Knight, 928 F.3d at 236 (describing the socialmedia account at issue as having “interactive features

open to the public, making public interaction a

prominent feature of the account,” and stating that

“[t]hese factors mean that the account is not private”);

accord, e.g., Packingham, 582 U.S. at 104 (describing

social media as particularly significant spaces “for the

exchange of views” today). Moreover, a forum does not

need to be “spatial or geographic.” Rosenberger, 515

U.S. at 830. The “same principles” apply to a

“metaphysical” forum. Id.

25

Assuming the public can access and interact with

them, government officials’ social-media accounts can

create either designated or limited public forums. If

state-action social-media accounts and posts allow the

public to openly comment on and otherwise interact

with them, they become designated public forums. So

the typical, run-of-the-mill government-controlled

social-media account that is open to the public for

“indiscriminate use” (without any governmental

speech-regulating policy in place) would become a

designated public forum because “the state has

opened [it] for use by the public as a place for

expressive activity.” Perry, 460 U.S. at 45, 47. As

such, an official could place content-neutral time,

place, and manner restrictions on the public’s

interactions with the account, but any contentdiscriminatory regulations would have to survive

strict scrutiny. Id. at 46.

State officials also can place certain “reasonable”

limitations on their official social-media accounts,

thereby transforming them into limited public

forums. See Forbes, 523 U.S. at 680. For example,

government social-media accounts can establish

narrow, objective, and definite “polic[ies] and

practice[s]” restricting discussion to certain topics

relevant to the officials’ work or setting rules against

unlawful harassment. Cornelius, 473 U.S. at 802;

Garnier, 41 F.4th at 1179 (discussing how the use of

keyword filters that automatically block comments

created a limited public forum); Davison v. Plowman,

247 F. Supp. 3d 767, 777 (E.D. Va. 2017), aff’d, 715 F.

App’x 298 (4th Cir. 2018) (approving “clearly off topic”

restriction). But “unwritten rule[s] of decorum” do not

qualify. Garnier, 41 F.4th at 1167.

26

Based on these standard First Amendment principles, the Ninth Circuit here correctly held that the

challenged government social-media accounts, absent

a “policy or practice of regulating the content” of

speech that the public posted on them, created

designated public forums, and that the later addition

of word-filter limitations transformed them into

limited public forums. Id. at 1179.

C. Forum analysis protects public debate

while preserving officials’ control over

their personal accounts and protecting

their own First Amendment freedoms.

Using this Court’s standard public-forum analysis,

the Ninth Circuit correctly held that the Trustees

violated the Garniers’ First Amendment rights. The

same cannot be said for the Sixth Circuit in Lindke. If

the Sixth Circuit had used a post-by-post test, it

should have recognized that at least some of the

defendant’s social-media activity was state action.

It then could have held that the defendant violated

the plaintiff ’s First Amendment rights—a vital

question that the Sixth Circuit’s state-action analysis

prevented it from even entertaining.

Because Freed, the government official in Lindke,

converted his Facebook page from a private to a public

account, and because he employed no restrictions on

who could follow his account or comment on his posts,

any of his posts in which he carried out his official

functions created designated public forums. Lindke,

37 F.4th at 1201; cf. Garnier, 41 F.4th at 1179

(“Where, as here, the government has made a forum

available for use by the public and has no policy or

practice of regulating the content posted to that

forum, it has created a designated public forum.”)

27

(cleaned up). In a designated public forum, officials

can employ content-neutral and “reasonable restrictions on the time, place, or manner of protected speech,

provided the restrictions” are “narrowly tailored to

serve a significant governmental interest” and “leave

open ample alternative channels.” Ward v. Rock

Against Racism, 491 U.S. 781, 791 (1989) (cleaned

up). But as in any public forum, “restrictions based on

viewpoint are prohibited.” Pleasant Grove City, 555

U.S. at 469.

Under this Court’s standard forum analysis, the

Sixth Circuit could have held that Freed’s retaliatory

actions of deleting Lindke’s comments and then

blocking Lindke from the page constituted viewpoint

discrimination. After all, Freed used his public

account to share “information about City programs,

policies, and actions,” post “about the COVID-19

pandemic and the City’s response to it,” and comment

on news articles that reported on city actions. Lindke

v. Freed, 563 F. Supp. 3d 704, 706–07 (E.D. Mich.

2021); Lindke, 37 F.4th at 1201. But Freed then

deleted Lindke’s comments that “questioned and

criticized the response of Port Huron governmental

officials, including Freed, to the COVID-19

pandemic.” Lindke, 563 F. Supp. 3d at 707; see

Lindke, 37 F.4th at 1201–02. And Lindke was not

alone: “four other individuals testified that Freed

deleted their comments on Freed’s posts that were

critical of Freed or the City’s actions on different

issues.” Lindke, 563 F. Supp. 3d at 707.

Freed also likely could not have met his burden to

show that his decision to completely block Lindke was

narrowly tailored. Freed’s decision to block Lindke

likely “burden[ed] substantially more speech than is

necessary.” Ward, 491 U.S. at 799. Like the Garniers,

28

Lindke could not “leav[e] any comments at all, no

matter how short, relevant, or non-duplicative they

might be.” Garnier, 41 F.4th at 1182. If the Sixth

Circuit had held that Freed’s posts were state action

and that his actions against Lindke and other constituents violated the First Amendment, then Lindke

and other concerned citizens could have held their city

manager accountable for his online censorship.

That drastically different outcome highlights the

importance of the questions presented in these cases,

especially for organizations whose mission is to

influence the climate of public opinion and shape

public policy. Much public engagement now occurs

online, where many public officials—at all levels of

government—are present and active, personally and

professionally. In the ever-evolving world of social

media, where public officials blend their personal and

professional lives, it is possible to balance a government official’s interest in regulating discussion of his

or her personal affairs with the public’s interest in

speaking freely on matters of public concern. But it is

only possible if courts employ the correct test to

properly identify state action in all its forms—

including discrete posts on otherwise blended socialmedia accounts—and hold public officials accountable

for free-speech violations while allowing them proper

breathing room to convey their messages and exercise

their own First Amendment rights.

29

CONCLUSION

The Court should apply the purpose-andappearance test for social-media state action and

affirm.

Respectfully submitted,

ILYA SHAPIRO

MANHATTAN INSTITUTE

52 Vanderbilt Ave.

New York, NY 20017

(212) 599-7000

CORTNEY C. THOMAS

BRIAN E. ROBISON

RUSSELL T. BROWN

BROWN FOX, PLLC

8111 Preston Rd.

Suite 300

Dallas, TX 75225

(214) 327-5000

AUGUST 2023

JOHN J. BURSCH

Counsel of Record

RYAN L. BANGERT

JEREMY D. TEDESCO

CHRISTOPHER P. SCHANDEVEL

MICHAEL R. ROSS

MATHEW W. HOFFMANN

ALLIANCE DEFENDING

FREEDOM

44180 Riverside Pkwy

Lansdowne, VA 20176

(616) 450-4235

jbursch@ADFlegal.org