Facebook v. Sullivan: Public Figures and Newsworthiness in Online Speech

St. John's University School of Law St. John's University School of Law

St. John's Law Scholarship Repository St. John's Law Scholarship Repository

Faculty Publications

2019

Facebook v. Sullivan: Public Figures and Newsworthiness in Facebook v. Sullivan: Public Figures and Newsworthiness in

Online Speech Online Speech

Thomas E. Kadri

Kate Klonick

St. John's University School of Law

Follow this and additional works at: https://scholarship.law.stjohns.edu/faculty_publications

Part of the First Amendment Commons, Privacy Law Commons, and the Torts Commons

This Article is brought to you for free and open access by St. John's Law Scholarship Repository. It has been

accepted for inclusion in Faculty Publications by an authorized administrator of St. John's Law Scholarship

Repository. For more information, please contact [email protected].

FACEBOOK V. SULLIVAN: PUBLIC

FIGURES AND NEWSWORTHINESS IN

ONLINE SPEECH

THOMAS E. KADRI

& KATE KLONICK

†

In the United States, there are now two systems to adjudicate disputes

about harmful speech. The first is older and more established: the legal

system in which judges apply constitutional law to limit tort claims alleging

injuries caused by speech. The second is newer and less familiar: the

content-moderation system in which platforms like Facebook implement the

rules that govern online speech. These platforms are not bound by the First

Amendment. But, as it turns out, they rely on many of the tools used by courts

to resolve tensions between regulating harmful speech and preserving free

expression—particularly the entangled concepts of “public figures” and

“newsworthiness.”

This Article offers the first empirical analysis of how judges and content

moderators have used these two concepts to shape the boundaries of free

speech. It first introduces the legal doctrines developed by the “Old

Governors,” exploring how courts have shaped the constitutional concepts

*. Resident Fellow, Yale Law School; Ph.D. Candidate, Yale University.

†. Assistant Professor of Law, St. John’s University School of Law; Affiliate Fellow, Yale Law

School Information Society Project. Authors listed alphabetically. Portions of this Article build on

research and ideas featured in Thomas Kadri’s and Kate Klonick’s doctoral dissertations and as well as

an essay published in the Emerging Threats series organized by the Knight First Amendment Institute at

Columbia University. See Kate Klonick, Facebook v. Sullivan, KNIGHT FIRST AMEND. INST. (Oct. 1,

2018), https://knightcolumbia.org/content/facebook-v-sullivan [https://perma.cc/W9R8-7K9B]. The

authors are grateful for the friends and colleagues whose insights improved this piece, especially Jack

Balkin, Molly Brady, Aaron Caplan, Danielle Citron, Jennifer Daskal, Evelyn Douek, Sarah Haan,

Margot Kaminski, Daphne Keller, Jennifer Rothman, Robert Post, Rory Van Loo, Morgan Weiland, and

colleagues at the Yale Information Society Project. Additional thanks for the opportunity to present this

work and receive excellent feedback at the Cornell Tech Speed Conference, Loyola Law School Los

Angeles Faculty Workshop, Washington School of Law Faculty Workshop, University of Colorado

Boulder Silicon Flatirons Conference, UCLA Social Media Conference, and Yale Freedom of Expression

Conference. A very special thank you to David Pozen, whose excellent, patient editing and thoughtful

feedback consolidated this Article’s central arguments and encouraged new ones.

   
38  SOUTHERN CALIFORNIA LAW REVIEW  [Vol. 93:37 
of  public  figures  and  newsworthiness  in  the  face  of  tort  claims  for 
defamation,  invasion  of  privacy,  and  intentional  infliction  of  emotional 
distress. The Article then turns to the “New Governors” and examines how 
Facebook’s  content-moderation  system channeled elements  of  the  courts’ 
reasoning for imposing First Amendment limits on tort liability.  
By  exposing  the  similarities  and  differences  between  how  the  two 
systems have understood these concepts, this Article offers lessons for both 
courts  and  platforms  as  they  confront  new  challenges  posed  by  online 
speech. It exposes the pitfalls of using algorithms to identify public figures; 
explores  the  diminished  utility  of  setting  rules  based  on  voluntary 
involvement  in  public  debate;  and  analyzes  the  dangers  of  ad  hoc  and 
unaccountable newsworthiness determinations. Both courts and platforms 
must adapt to the new speech ecosystem that companies like Facebook have 
helped create, particularly the way that viral content has shifted normative 
intuitions  about  who  deserves  harsher  rules  in  disputes  about  harmful 
speech, be it in law or content moderation. 
Finally,  the  Article  concludes  by  exploring  what  this  comparison 
reveals about the structural role platforms play in today’s speech ecosystem 
and how it illuminates  new solutions. These platforms act as  legislature, 
executive,  judiciary, and press—but without any separation  of powers to 
establish checks and balances. A change to this model is already occurring 
at  one  platform:  Facebook  is  creating  a  new  Oversight  Board  that  will 
hopefully provide due process to users on the platform’s speech decisions 
and transparency about how content-moderation policy is made, including 
how concepts related to newsworthiness and public figures are applied.  
TABLE OF CONTENTS 
INTRODUCTION .................................................................................. 39 
I. PUBLIC FIGURES AND NEWSWORTHINESS IN OLD 
GOVERNANCE: TORT LAW AND THE FIRST AMENDMENT  
IN THE COURTS ................................................................................ 42 
A. DEFAMATION .................................................................................. 43 
B. INVASION OF PRIVACY ..................................................................... 50 
C. INTENTIONAL INFLICTION OF EMOTIONAL DISTRESS........................ 55 
II. PUBLIC FIGURES AND NEWSWORTHINESS IN NEW 
GOVERNANCE: CONTENT MODERATION AND FREE  
SPEECH AT FACEBOOK .................................................................. 58 
A. CYBERBULLYING............................................................................. 60 
B. NEWSWORTHY CONTENT ................................................................. 64 
C. CONTEMPORARY APPROACHES ........................................................ 67 

2019] FACEBOOK V. SULLIVAN 39

III. FACEBOOK VERSUS SULLIVAN: LESSONS FOR COURTS

AND PLATFORMS IN THE DIGITAL AGE ...................................... 70

A. THE RATIONALES BEHIND THE RULES ............................................. 72

B. JUDGING OUR GOVERNORS, NEW AND OLD ..................................... 74

1. The Inaccuracies and Injustices of Algorithmic Authority ............ 75

2. Voluntariness and Sympathy in the Age of Virality ...................... 80

CONCLUSION ....................................................................................... 93

INTRODUCTION

In the summer of 2017, a group of American neo-Nazis convened for a

“Unite the Right” rally in Charlottesville, Virginia. Amid scenes of chaos,

James Alex Fields drove his car through a crowd and killed a young counter-

protestor, Heather Heyer. The next day, a blog post from the white

supremacist website, The Daily Stormer, was shared over 65,000 times on

Facebook: “Heather Heyer, Woman Killed in Road Rage Incident was a Fat,

Childless 32-Year-Old Slut.”

To some, this post looked like provocative but

permissible commentary about someone who was now newsworthy; to

others, it seemed like harmful speech that Facebook should remove as a

violation of its internal rules.

When people complained about the post,

Facebook’s rulemakers debated: Should it stay up or come down?

Ultimately, they hedged. The platform removed every link to the post unless

the user included a caption condemning The Daily Stormer.

This episode reveals something important about free speech in the

digital age: the judiciary is no longer the only actor that adjudicates claims

about harmful speech. In the United States, we now have two systems to

adjudicate these disputes. The first is older and more established: the legal

system in which judges apply constitutional law to limit tort claims alleging

injuries caused by harmful speech.

The second is newer and less familiar:

. Talia Lavin, The Neo-Nazis of The Daily Stormer Wander the Digital Wilderness, NEW

YORKER (Jan. 7, 2018), https://www.newyorker.com/tech/annals-of-technology/the-neo-nazis-of-the-

daily-stormer-wander-the-digital-wilderness [https://perma.cc/8XL7-UYKV]; see also Casey Newton,

Facebook Is Deleting Links to a Viral Attack on a Charlottesville Victim, THE VERGE (Aug. 14, 2017,

8:30 PM), https://www.theverge.com/2017/8/14/16147126/facebook-delete-viral-post-charlottesville-

daily-stormer [https://perma.cc/JLF2-AT2G].

. See Julia Angwin et al., Have You Experienced Hate Speech on Facebook? We Want to Hear

from You., PROPUBLICA (Aug. 29, 2017, 10:05 AM), https://www.propublica.org/article/have-you-

experienced-hate-speech-on-facebook-we-want-to-hear-from-you [https://perma.cc/A8PF-MC4U].

. Newton, supra note 1.

. See, e.g., Hustler Magazine, Inc. v. Falwell, 485 U.S. 46, 56 (1988) (intentional infliction of

emotional distress); Time, Inc. v. Hill, 385 U.S. 374, 381 (1967) (privacy); N.Y. Times Co. v. Sullivan,

376 U.S. 254, 271 (1964) (defamation). On the history and complications of constitutionalizing tort law,

40 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

the content-moderation system in which platforms like Facebook implement

the rules that govern online speech.

These platforms aren’t bound by the

First Amendment. But, as it turns out, they rely on many of the tools used by

courts to resolve tensions between regulating harmful speech and preserving

free expression—particularly the entangled concepts of “public figures” and

“newsworthiness.”

This Article analyzes how judges and content moderators have used

these two concepts to shape the boundaries of free speech. By exposing the

similarities and differences between how the two systems have understood

these concepts, this Article offers lessons for both courts and platforms as

they confront new challenges posed by online speech. Finally, the Article

briefly explores how this comparison reveals the structural changes that

platforms like Facebook should make to address these challenges and bring

oversight to their governance of online speech.

Part I introduces the legal doctrines developed by the “Old Governors,”

exploring how courts have shaped the concepts of public figures and

newsworthiness in the face of tort claims for defamation, invasion of privacy,

and intentional infliction of emotional distress. Part II turns to the “New

Governors” and examines Facebook’s content-moderation system. Drawing

on internal Facebook documents and several years of exclusive interviews

with current and former Facebook employees, this Part reveals for the first

time how and why the platform created its own rules related to public figures

and newsworthiness. In so doing, it shows that Facebook’s rulemaking—

consciously or unconsciously—channeled elements of the courts’ reasoning

when imposing First Amendment limits on tort liability.

see Daniel J. Solove & Neil M. Richards, Rethinking Free Speech and Civil Liability, 109 COLUM. L.

REV. 1650, 1672–84 (2009); Eugene Volokh, Tort Liability and the Original Meaning of the Freedom of

Speech, Press, and Petition, 96 IOWA L. REV. 249, 251–54 (2010).

. For foundational work on the development of the idea of private governance by speech

platforms, see REBECCA MACKINNON, CONSENT OF THE NETWORKED 149–65 (2012) (analyzing

platforms through the lens of “governance” at the “new digital sovereigns” of “Facebookistan” and

“Googledom”); see also Anupam Chander, Facebookistan, 90 N.C. L. REV. 1807, 1819–22 (2012);

Robert Gorwa, What is Platform Governance?, 22 INFO. COMM. & SOC’Y 854, 854 (2019); Kate Klonick,

The New Governors: The People, Rules, and Processes Governing Online Speech, 131 HARV. L. REV.

1598, 1616–18 (2018); Thomas Kadri, How Supreme a Court?, SLATE (Nov. 19, 2018, 1:59 PM),

https://slate.com/technology/2018/11/facebook-zuckerberg-independent-speech-content-appeals-

court.html [https://perma.cc/524G-QYLJ] [hereinafter Kadri, How Supreme a Court?]; Thomas Kadri,

Speech vs. Speakers, SLATE (Jan. 18, 2018, 12:56 PM), https://slate.com/technology/2018/01/twitters-

new-rules-blur-the-line-between-extremists-speakers-and-their-speech.html [https://perma.cc/JW9B-FF

Q6] [hereinafter Kadri, Speech vs. Speakers]; Kate Klonick & Thomas Kadri, Opinion, How to Make

Facebook’s ‘Supreme Court’ Work, N.Y. TIMES (Nov. 17, 2018), https://www.nytimes.com/2018/

11/17/opinion/facebook-supreme-court-speech.html [https://perma.cc/DRD4-N8AK].

2019] FACEBOOK V. SULLIVAN 41

Guided by this background, Part III exposes the similarities and

differences between how the private and public governance systems have

understood the concepts of public figures and newsworthiness. Through this

comparative analysis, this Part addresses how both courts and platforms

should confront new challenges posed by online speech. Judges and platform

policymakers must adapt to the new speech ecosystem that companies like

Facebook have helped create, particularly the way that virality has shifted

normative intuitions about who deserves harsher rules in disputes about

harmful speech, be it in constitutional law or content moderation.

This Part

first exposes the pitfalls of using algorithms to identify public figures,

critiquing Facebook’s use of online news aggregators and news sources to

determine public-figure status on the platform—a mechanism that results in

Facebook removing too much benign speech and preserving too much

harmful speech. It then explains how the internet has eroded traditional

reasons for specially protecting speech about public figures because these

reasons rest principally on assumptions that people become public figures by

choice and that, as public figures, they have greater access to channels of

rebuttal.

These assumptions are becoming increasingly outdated in the

digital age, given the dynamics of online speech

and the ubiquity of means

to engage in counter-speech.

As a result, this Part reassesses the normative

utility of distinguishing between “voluntary” and “involuntary” public

figures to judge who deserves harsher rules in courts and on platforms,

discussing the alternative use of “sympathy” as a normative barometer.

Lastly, this Part assesses the risks posed by platforms creating ad hoc

exceptions for newsworthy content. Although there are significant

advantages to increasing human review in content moderation, the current

. See generally THOMAS I. EMERSON, THE SYSTEM OF FREEDOM OF EXPRESSION 3−6 (1970)

(developing the idea of the First Amendment creating a “system” of free speech).

. See Gertz v. Robert Welch, Inc., 418 U.S. 323, 352 (1974); Curtis Publ’g Co. v. Butts, 388

U.S. 130, 154 (1967).

. See DANAH BOYD, IT’S COMPLICATED: THE SOCIAL LIVES OF NETWORKED TEENS 11–12

(2014) (explaining how social media creates new challenges because of the “persistence,” “visibility,”

“spreadability,” and “searchability” of content); Kate Klonick, Re-Shaming the Debate: Social Norms,

Shame, and Regulation in an Internet Age, 75 MD. L. REV. 1029, 1053–54 (2015) (describing how the

internet changes social norm enforcement by eliminating frictions of time, geography, personal reputation

and cost). See generally DANIELLE KEATS CITRON, HATE CRIMES IN CYBERSPACE (2014) (describing

cyber harassment and exploring ways to combat it); Jack M. Balkin, Old-School/New-School Speech

Regulation, 127 HARV. L. REV. 2296 (2014) (discussing the changing approach to speech regulation).

. See Eugene Volokh, Cheap Speech and What It Will Do, 104 YALE L.J. 1805, 1833 (1995)

(predicting, even before the advent of social media, that new technologies would “both democratize the

information marketplace—make it more accessible to comparatively poor speakers as well as rich ones—

and diversify it”); see also LAWRENCE LESSIG, CODE VERSION 2.0, at 19 (2006); Jack M. Balkin, Digital

Speech and Democratic Culture: A Theory of Freedom of Expression for the Information Society, 79

N.Y.U. L. REV. 1, 2 (2004).

42 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

structure of most platforms means that decisions in this area lack

transparency, expertise, and consistency.

This Article concludes by looking briefly at the Facebook Oversight

Board, an independent institution that Facebook is currently creating to

provide users with transparency and procedure around content moderation.

Ideally, such a Board will provide users with explanation and consistency

around how the concepts of newsworthiness and public figures are applied.

This institution may ultimately result in users gaining procedural and

structural protections in the new, private system regulating their expression.

The Article concludes that more platforms should adopt such oversight

mechanisms to protect users’ speech rights and provide accountability,

transparency, and due process.

I. PUBLIC FIGURES AND NEWSWORTHINESS IN OLD

GOVERNANCE: TORT LAW AND THE FIRST AMENDMENT IN

THE COURTS

Within the vast world of tort law, which aims to provide relief when

people are harmed, are what Jack Balkin calls “communications torts.”

Communications torts are “a category of legal causes of action in which

people are harmed by speech acts of others.”

The concept of

communications torts is particularly salient in the digital age because, as

Balkin foresaw, “all activity in virtual worlds must begin as a form of

speech” such that “[w]hen people injure each other in virtual worlds in ways

that the law will recognize, they are almost always committing some form of

communications tort.”

Unsurprisingly, then, legal claims involving communications torts will

often implicate the First Amendment because they are premised on harms

caused by speech. When adjudicating these claims, the Supreme Court has

looked at whether the claimant is a public figure or whether the underlying

speech is newsworthy. This Part focuses on how the concepts of public

figures and newsworthiness have curtailed three communications torts:

defamation, invasion of privacy, and intentional infliction of emotional

distress. By unfolding the history and underlying rationales of the Court’s

First Amendment doctrine in this area, this Part provides context for Part II’s

exploration of how Facebook developed its own rules surrounding these

. Jack M. Balkin, Law and Liberty in Virtual Worlds, 49 N.Y.L. SCH. L. REV. 63, 73 (2004).

. Id.

2019] FACEBOOK V. SULLIVAN 43

same concepts. As we will see in Part III, although these public and private

doctrines evolved in seemingly disparate contexts, they share many

similarities—and, in a new era of online speech, some critical differences.

A. DEFAMATION

The tort of defamation has played a central role in the development of

First Amendment law. Although liability for false speech that injures

someone’s reputation was long thought to raise no constitutional concern,

the Supreme Court eventually developed an intricate web of rules to restrain

the reach of defamation and protect free speech. The story begins in March

1960 when L.B. Sullivan, an elected commissioner from Alabama, sued the

New York Times for defamation after the newspaper published an

advertisement criticizing the Montgomery Police Department’s treatment of

civil-rights demonstrators.

Sullivan claimed that the advertisement

contained falsehoods that damaged his reputation. His case made it all the

way to the Supreme Court.

In addressing the threshold issue in New York Times Co. v. Sullivan,

Justice William Brennan explained that torts like defamation “can claim no

talismanic immunity from constitutional limitations” and “must be measured

by standards that satisfy the First Amendment.”

Although he

acknowledged that the advertisement contained falsehoods, he explained that

“erroneous statement is inevitable in free debate” and “must be protected if

the freedoms of expression are to have the ‘breathing space’ that they

‘need . . . to survive.’”

Defamation claims, he said, cannot create an

environment that “dampens the vigor and limits the variety of public

debate.”

To address this concern, the Court crafted a special constitutional

rule: public figures alleging defamation must prove that the offending

statements were made with “actual malice”—that is, with knowledge that the

statement “was false or with reckless disregard of whether it was false or

. Id. at 73–76.

. LEE LEVINE & STEPHEN WERMIEL, THE PROGENY: JUSTICE WILLIAM J. BRENNAN’S FIGHT TO

PRESERVE THE LEGACY OF NEW YORK TIMES V. SULLIVAN 3–4 (2014).

. See generally id. (providing an excellent account of this historic case); KERMIT L. HALL &

MELVIN I. UROFSKY, NEW YORK TIMES V. SULLIVAN: CIVIL RIGHTS, LIBEL LAW, AND THE FREE PRESS

(2011) (same); Mary-Rose Papandrea, The Story of New York Times Co. v. Sullivan, in FIRST

AMENDMENT STORIES 229 (Richard W. Garnett & Andrew Koppelman eds., 2012) (same).

. N.Y. Times Co. v. Sullivan, 376 U.S. 254 (1964).

. Id. at 269.

. Id. at 271–72 (first alteration in original) (quoting NAACP v. Button, 371 U.S. 415, 433

(1963)).

. Id. at 279.

44 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

not.”

The justices in Sullivan identified two justifications for creating this

constitutional hurdle for public figures. The central reason—one that Justice

Brennan discussed at length in the majority opinion—was the democratic

imperative of preserving “debate on public issues.”

This rationale reflects

a theory of the First Amendment grounded in democratic self-governance.

Various scholars have shaped the contours of this theory and glossed it in

different ways,

but a central concern for many of them is the public’s need

to have the information necessary to engage in self-government.

Under this

rationale, the First Amendment protects the public’s entitlement “to all

information that is necessary for informed governance” because “the public,

in its role as the electorate, [is] ultimately responsible for political

decisions.”

The self-governance theory is often cached out in these

“educative” terms by justifying “speech protection not because of any

individual right of expression but instead because of the need to create an

informed public.”

. Id. at 279–80.

. Id. at 270.

. Compare, e.g., ALEXANDER MEIKLEJOHN, FREE SPEECH AND ITS RELATION TO SELF-

GOVERNMENT (1948) (developing a listener-focused account for the relationship between free speech

and self-government through the idea of the educative function of a town meeting), with ROBERT C. POST,

CONSTITUTIONAL DOMAINS: DEMOCRACY, COMMUNITY, MANAGEMENT (1995) (developing a speaker-

focused account that connects free speech and self-government by exploring the legitimating function

served by public discourse).

. See generally Thomas E. Kadri, Drawing Trump Naked: Curbing the Right of Publicity to

Protect Public Discourse, 78 MD. L. REV. 899, 905 (2019) (discussing theorists who hold this view and

dubbing them “educative” theorists).

. Robert C. Post, The Social Foundations of Privacy: Community and Self in the Common Law

Tort, 77 CALIF. L. REV. 957, 999 (1989).

. Kadri, supra note 23, at 913; see also OWEN M. FISS, LIBERALISM DIVIDED: FREEDOM OF

SPEECH AND THE MANY USES OF STATE POWER 13 (1996) (attesting that “[w]e allow people to speak so

others can vote” because “[s]peech allows people to vote intelligently and freely, aware of all the options

and in possession of all the relevant information”); ALEXANDER MEIKLEJOHN, POLITICAL FREEDOM: THE

CONSTITUTIONAL POWERS OF THE PEOPLE 55 (1960) (arguing that the First Amendment “has no concern

about the ‘needs of many men to express their opinions’” but rather is concerned with “the common needs

of all the members of the body politic”); id. at 56–57, 61 (criticizing Zechariah Chafee, Jr.’s “inclusion

of an individual interest within the scope of the First Amendment,” and Justice Oliver Wendell Holmes’s

“excessive individualism” on this front); Owen M. Fiss, Free Speech and Social Structure, 71 IOWA L.

REV. 1405, 1409–10 (1986) (arguing that “[t]he purpose of free speech is not individual self-actualization,

but rather the preservation of democracy, and the right of a people, as a people, to decide what kind of

life it wishes to live”); Owen M. Fiss, Why the State?, 100 HARV. L. REV. 781, 786 (1987) (framing the

individual speech right in instrumental terms, worthy of protection “only when it enriches public debate”);

Cass R. Sunstein, Television and the Public Interest, 88 CALIF. L. REV. 499, 501 (2000) (maintaining that

the primary purpose of free speech is to promote deliberative democracy— “a system in which citizens

are informed about public issues and able to make judgments on the basis of reasons”). For some skeptical

2019] FACEBOOK V. SULLIVAN 45

Today, Sullivan is often framed as a case that deals solely with the

importance of speech about “public figures,” but this undervalues its

significance to a much broader doctrine. Justice Brennan’s concern was not

limited to protecting speech about a plaintiff’s political position as a

government official, but rather what the justice called a “profound national

commitment to the principle that debate on public issues should be

uninhibited, robust, and wide-open.”

The focus was not simply on

Sullivan’s social status, but also—at least implicitly—on whether the speech

at issue related to issues that the public needed to know in order to self-

govern. The question whether a defamation plaintiff is a public figure is part

of a subtler inquiry about whether the speech in question touches on a matter

of public concern.

As a result, an important part of Sullivan’s legacy is that

it weaved self-governance theory into First Amendment doctrine.

Although the self-governance theory animated the Court’s opinion in

Sullivan, a second rationale for treating public figures differently appeared

in Justice Arthur Goldberg’s concurrence: public figures enjoy “equal if not

greater access than most private citizens to media of communication.”

Though not part of the majority’s reasoning, this additional justification

eventually animated the development of public-figure doctrine. When a

university athletic director and a prominent political activist sued two

newspapers for defamation in Curtis Publishing Co. v. Butts, the Court

adopted Justice Goldberg’s rationale by justifying its decision on the basis

that plaintiffs “had sufficient access to the means of counterargument” to

rebut the alleged falsehoods.

The Court also echoed the self-governance

rationale in explaining why Sullivan’s constitutional rule applied to plaintiffs

who were not government officials,

explaining that the “public interest” in

being informed about nonpolitical public figures was “not less” than being

informed about public officials.

Thus, although the Court’s holding hinged

on the plaintiffs’ social status, the justification for it was again rooted in the

treatment of these theorists, see J.M. Balkin, Populism and Progressivism as Constitutional Categories,

104 YALE L.J. 1935, 1935–90 (1995); Robert Post, Meiklejohn’s Mistake: Individual Autonomy and the

Reform of Public Discourse, 64 U. COLO. L. REV. 1109, 1109–23 (1993).

. Sullivan, 376 U.S. at 270

. See Catherine Hancock, Origins of the Public Figure Doctrine in First Amendment Defamation

Law, 50 N.Y.L. SCH. L. REV. 81, 85 (2005); Post, supra note 24, at 997.

. Sullivan, 376 U.S. at 304–05 (Goldberg, J., concurring).

. Curtis Publ’g Co. v. Butts, 388 U.S. 130, 135–41, 155 (1967).

. Id. at 155.

. Id. at 154–55; see also id. at 163 (Warren, C.J., concurring in result) (declaring that

distinguishing between public figures and officials would have “no basis in law, logic, or First

Amendment policy”).

46 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

importance of robust debate on “public issues,”

as well as the fact that the

plaintiffs could participate in that debate by virtue of their prominence.

Given the Court’s reliance on self-governance theory to protect speech

on “public issues,” one might wonder why the Court focused on the

plaintiff’s status instead of simply asking whether the speech at issue was

necessary for the public to know. The Court flirted with such a reformulation

of the doctrine in Rosenbloom v. Metromedia, Inc., in which a plurality

extended Sullivan’s rule to all defamation claims involving speech on

matters of public concern, regardless of whether the plaintiff was a public or

private figure.

Justice Brennan’s plurality opinion reasoned that a matter

“of public or general interest . . . cannot suddenly become less so merely

because a private individual is involved, or because in some sense the

individual did not ‘voluntarily’ choose to become involved.”

In his view,

“[t]he public’s primary interest is in the event,” not the individual’s social

status.

Any interest in the individual’s prominence was merely a corollary

to that primary interest, for “the public focus is on the conduct of the

participant and the content, effect, and significance of the conduct, not the

participant’s prior anonymity or notoriety.”

In order to “honor the

commitment to robust debate on public issues . . . embodied in the First

Amendment” that the Court had recognized in Sullivan, Justice Brennan

concluded that the constitutional rule must apply “to all discussion and

communication involving matters of public or general concern, without

regard to whether the persons involved are famous or anonymous,” though

quite what constituted a matter of “public or general concern” was left rather

opaque.

Rosenbloom’s doctrinal simplicity—if jettisoning the “public figure”

concept for the ambiguous notion of the “public or general concern” can be

considered simplicity—was fleeting. Just three years later, the Court held in

Gertz v. Robert Welch, Inc. that Sullivan’s rule should not apply to claims

brought by private figures.

In drawing lines between public and private

figures, the Court imagined two, and possibly three, types of public

figures: general public figures; limited-purpose public figures; and, perhaps,

. Id. at 147.

. Rosenbloom v. Metromedia, Inc., 403 U.S 29, 43–44 (1971) (plurality opinion).

. Id. at 43.

. Id. (emphasis added).

. Id.

. Id. at 43–44.

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 339–48 (1974).

2019] FACEBOOK V. SULLIVAN 47

involuntary public figures.

General public figures “occupy positions of

such persuasive power and influence that they are deemed public figures for

all purposes,”

while limited-purpose public figures “thrust themselves to

the forefront of particular public controversies in order to influence the

resolution of the issues involved.”

Notably, the Court left undefined the

third (possible) type of public figure, observing only that “truly involuntary

public figures must be exceedingly rare.”

This remark was prophetic: the

Court has never encountered its mythical character in the forty years that

have since passed.

The demarcation of different types of public figures has developed into

a central concern in the Court’s defamation jurisprudence. In contrast to the

atrophied idea of the involuntary public figure, the Court has developed the

“voluntariness” element at length. In Gertz, the Court explained that both

general and limited-purpose public figures “invite attention and comment”

and thus have “voluntarily exposed themselves to increased risk of injury

from defamatory falsehood concerning them.”

Building from this premise,

the Court in Time, Inc. v. Firestone held that a woman who divorced a

member of a wealthy socialite family was not a public figure.

The Court

cautioned against adopting a concept of public figures that might create too

large a class, and explained that a person needed more than local notoriety

to become a public figure.

The Court stressed that the divorcée did not

“freely choose to publicize issues as to the propriety of her married life” but

rather was “compelled” to go to court to end her marriage.

Her actions,

then, were “no more voluntary in a realistic sense” than those of a criminal

defendant “called upon to defend his interests in court.”

In addition, the

Court refused to equate “public controversy” with “all controversies of

interest to the public” because doing so would reinstate the Rosenbloom rule

. See W. Wat Hopkins, The Involuntary Public Figure: Not So Dead After All, 21 CARDOZO

ARTS & ENT. L.J. 1, 21 (2003) (“[T]here is disagreement as to whether the Supreme Court identified two

or three categories of public figure status.”).

. Gertz, 418 U.S. at 345.

. Id.

. See Jeffrey Omar Usman, Finding the Lost Involuntary Public Figure, 2014 UTAH L. REV.

951, 952 (claiming that “involuntary public figures” as category of individuals in First Amendment

defamation jurisprudence “has become lost”).

. Gertz, 418 U.S. at 345.

. Time, Inc. v. Firestone, 424.U.S. 448, 453 (1976).

. Id. at 450–53.

. Id. at 454.

. Id. (citation omitted).

48 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

that disavowed the significance of a plaintiff’s social status.

Since Firestone, the Court has repeatedly held that individuals who did

not voluntarily garner attention were not public figures. In Wolston v.

Reader’s Digest Association, for example, the Court declined to apply the

label to a witness who missed a grand-jury hearing investigating Cold War

espionage.

The witness had not “voluntarily thrust” or “injected” himself

into the public eye; rather, the Court declared that “[i]t would be more

accurate to say that [he] was dragged unwillingly into the controversy.”

Similarly, in Hutchinson v. Proxmire, in which a professor sued a U.S.

Senator who criticized him for wasting federal funds, the Court stressed that

the professor remained a private figure because he had not “thrust himself or

his views into public controversy to influence others.”

The Court rejected

the lower courts’ conclusion that the professor became a public figure in part

because he gained some prominence and access to the media after the

Senator had allegedly defamed him.

Although various media outlets

reported the professor’s response to the Senator’s criticism, the Court

stressed that the professor “did not have the regular and continuing access to

the media that is one of the accouterments of having become a public

figure.”

In other words, the controversy itself could not transform the

professor into a public figure simply because he appeared in the news as a

result.

As these cases show, although the Court glossed the public-figure

doctrine with a new taxonomy in Gertz and its progeny, the basic rationales

for affording heightened constitutional protection for speech about public

figures tracked the pair of rationales originally suggested in Sullivan and

Butts. First, speech about public figures requires protection for “debate on

public issues [to] be uninhibited, robust, and wide-open.”

The pervasive

power of some people makes their behaviors a matter of public interest, just

as the behaviors of someone who thrusts herself into a public controversy

become a matter of public interest, and the public needs to know about these

behaviors if it is to engage meaningfully in self-governance. Second, public

figures are “less vulnerable to injury from defamatory statements” because

. Id.

. Wolston v. Reader’s Digest Ass’n, 443 U.S. 157, 166–67 (1979).

. Id. at 166.

. Hutchinson v. Proxmire, 443 U.S. 111, 135 (1979).

. Id. at 134–36.

. Id. at 136.

. N.Y Times Co. v. Sullivan, 376 U.S. 254, 270 (1964).

2019] FACEBOOK V. SULLIVAN 49

they generally have greater ability to engage in “self-help” by “counter[ing]

criticism and expos[ing] the falsehood and fallacies of defamatory

statements” in the media.

These dual rationales sustain much of modern

public-figure doctrine to this day.

It’s crucial to note, however, that the Court’s doctrine also reflects two

caveats to these rationales. The first caveat relates to the significance of the

idea of “voluntariness.” The idea that a person chooses to place herself in the

public eye underlies the “normative consideration”

that public figures are

“less deserving”

of protection from defamation because, unlike private

figures, they have “voluntarily exposed themselves to increased risk of injury

from defamatory falsehood.”

In striking the balance between protecting

free speech and remedying harmful speech, the Court has continued to cling

to the importance of voluntariness, even when the idea of voluntariness

potentially conflicts with the public-figure doctrine’s embrace of self-

governance theory and its assertion that private figures lack the means of

rebuttal. After all, the fact that a plaintiff freely chose to enter the public

arena might not mean that he or she commands more public interest or that

he or she has greater access to the media—private figures might get caught

up in events that raise issues of great importance to the public, and the

attention that flows from these events might furnish these private figures

with ample opportunity to respond in the media. Nonetheless, the Court has

embraced this doctrinal tension to date.

The second caveat to the rationales developed in Sullivan and Butts

concerns the Rosenbloom retraction in Gertz. The Rosenbloom plurality had

sought to abandon distinctions based on social status,

and even based on

voluntariness,

but the Court in Gertz clawed back the significance of both

distinctions in the constitutional analysis. Had the Court’s sole concern been

to preserve “debate on public issues,” Rosenbloom’s rule would have carried

the day, for Justice Brennan was surely right when he observed in

. Wolston, 443 U.S. at 164.

. See generally Shlomit Yanisky-Ravid & Ben Zion Lahav, Public Interest vs. Private Lives—

Affording Public Figures Privacy in the Digital Era: The Three Principle Filtering Model, 19 J. CONST.

L. 975, 983–84 (2017) (outlining the various components of the Court’s public-figure doctrine as (1)

“access and control over the media”; (2) “enrollment in a special role in the public eye”; (3) “willingly

(voluntarily) choosing to engage in a public role, inviting invasion of privacy risks”; and (4) “public

controversy”).

. Wolston, 443 U.S. at 164.

. Id.

. Id. (quoting Gertz v. Robert Welch, Inc., 418 U.S. 323, 345 (1974)).

. Rosenbloom v. Metromedia, Inc., 403 U.S 29, 43–44 (1971).

. Id. at 43.

50 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

Rosenbloom that matters of public concern “cannot suddenly become less so

merely because a private individual is involved, or because in some sense

the individual did not ‘voluntarily’ choose to become involved.”

Yet

despite this fact, the Court chose not to embrace a more speech-protective

rule that would cover all speech on matters of public concern. Gertz and later

cases recognized the fundamental importance of free speech, but the Court

has nonetheless insisted upon a different constitutional regime for private

figures.

These two caveats reveal something important about the balance that

the Court has struck between preserving free speech and protecting people

from harmful speech. In the context of false and defamatory speech, the

Court has developed a constitutional commitment that the First Amendment

must limit tort liability in order to protect robust public debate. But the Court

has also recognized that the public’s eagerness to engage in such debate is

not always sufficient to override all attempts to tackle harmful speech

through defamation law. As the next section reveals, the Court has struck a

different balance when faced with other tort claims that raise free-speech

concerns.

B. INVASION OF PRIVACY

Given the distinctions drawn in defamation law between private and

public figures and matters of private and public concern, it is unsurprising

that privacy law has attracted similar constitutional concepts. Like the

defamation tort, privacy torts often involve allegations that a speech act has

caused harm to someone, and thus these claims regularly implicate the First

Amendment. After Sullivan constitutionalized defamation law through the

public-figure doctrine, defendants in privacy actions began raising

arguments that similar limitations should be placed on privacy torts in order

to preserve free speech. The resulting doctrine borrows heavily from the

Court’s defamation jurisprudence but differs in important respects.

Just a few years after Sullivan, the Court applied “the First Amendment

principles pronounced in [Sullivan]” within the privacy realm in Time, Inc.

v. Hill,

partially migrating Sullivan’s rule outside the defamation context

for the first time. The case developed after James Hill and his family were

. Id. at 43–44.

. Time, Inc. v. Hill, 385 U.S. 374, 390 (1967); see also Hancock, supra note 27, at 105–12

(discussing the relationship of Hill to Sullivan).

2019] FACEBOOK V. SULLIVAN 51

held hostage by escaped convicts.

Hill sued Life Magazine after it

published an article suggesting that a new play portrayed his family’s story.

Although he maintained that the article was “false and untrue,”

his claim

sounded not in defamation but in the privacy tort of unreasonably placing a

person in a “false light” before the public.

Hill argued that he was not a

public figure but a private citizen who had “involuntarily become

newsworthy” after he and his family were victims of a crime.

Justice Brennan again delivered the Court’s opinion, stressing—as he

had done in Sullivan—that “[f]reedom of discussion . . . must embrace all

issues about which information is needed.”

Foreshadowing an issue that

would become crucial in the Court’s defamation jurisprudence, Justice

Brennan declined to base a constitutional rule on “the distinction . . . between

the relative opportunities of the public official and the private individual to

rebut” harmful speech.

Indeed, he eschewed consideration of the plaintiff’s

social status entirely, as he would later attempt to do in Rosenbloom. He

focused instead on the need to preserve debate on “matters of public

interest”

and, in so doing, “declared an expansive view of the First

Amendment as protection for all newsworthy material,”

regardless of

whether the plaintiff is a private figure.

The decision reflected the self-

governance theory of the First Amendment, justified as it was by the fact that

the Hill’s ordeal concerned “issues about which information is needed” for

the public to govern itself effectively.

Hill gave constitutional weight to an idea that had deep foundations in

privacy law. Samuel Warren and Louis Brandeis’s seminal work advocating

for a vigorous right to privacy nonetheless stressed that such a right should

. Hill, 385 U.S. at 378.

. Id.

. See RESTATEMENT (SECOND) OF TORTS §§ 652A, 652E (AM. LAW INST. 1977).

. See Harry Kalven, Jr., The Reasonable Man and the First Amendment: Hill, Butts and Walker,

1967 SUP. CT. REV. 267, 279.

. Hill, 385 U.S. at 388 (citation omitted).

. Id. at 391.

. See id. at 387–88 (“We hold that the constitutional protections for speech and press preclude

the application of the New York statute to redress false reports of matters of public interest in the absence

of proof that the defendant published the report with knowledge of its falsity or in reckless disregard of

the truth.”).

. Samantha Barbas, When Privacy Almost Won: Time, Inc. v. Hill, 18 U. PA. J. CONST. L. 505,

508 (2015).

. Hill, 385 U.S. at 387–88.

. Id. at 388 (citation omitted).

52 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

not prohibit speech on matters “of public or general interest,”

and the first

decision acknowledging a right of privacy contained a similar

qualification.

Sometimes referred to as speech that’s “newsworthy,”

speech on matters of public concern is said to lie “at the core of the First

Amendment”

for reasons that self-governance theorists have detailed at

length.

The privacy tort of public disclosure of private facts has also been

cabined by constitutional concerns for free speech. As with the false-light

privacy claim in Hill, there can be no liability under the disclosure tort when

the speech in question is of “legitimate public concern.”

But while the Hill

Court seemed ambivalent to the plaintiff’s status, courts have superficially

entertained a distinction between private and public figures who bring

disclosure claims. As an initial matter, this is because people live much of

their lives out in the open; as a result, the disclosure tort will often be

inapplicable because many facts about people are not truly “private.”

complication arises, however, when people seek to shield certain aspects of

their lives from public view: whereas private figures might rely on the

disclosure tort, the “legitimate interest of the public” may extend “to some

reasonable extent” to facts about public figures “that would otherwise be

purely private.”

. Samuel D. Warren & Louis D. Brandeis, The Right to Privacy, 4 HARV. L. REV. 193, 214

(1890).

. Pavesich v. New England Life Ins., 50 S.E. 68, 74 (Ga. 1905) (“The truth may be spoken,

written, or printed about all matters of a public nature, as well as matters of a private nature in which the

public has a legitimate interest.”); see also Post, supra note 24, at 996.

. Time, Inc. v. Hill, 385 U.S. 374, 400 (1967) (Black, J., concurring); see also FLA. STAT.

§ 90.5015(1)(b) (2017) (“‘News’ means information of public concern relating to local, statewide,

national, or worldwide issues or events.”); Kadri, supra note 23, at 912–13; Mary-Rose Papandrea,

Citizen Journalism and the Reporter’s Privilege, 91 MINN. L. REV. 515, 578–81 (2007) (observing that

the newsworthiness standard “involves essentially the same inquiry as a ‘public concern’ test”).

. Frisby v. Schultz, 487 U.S. 474, 479 (1988); see also NAACP v. Claiborne Hardware Co., 458

U.S. 886, 913 (1982) (“[E]xpression on public issues ‘has always rested on the highest rung of the

hierarchy of First Amendment values.’” (citation omitted)); Schenck v. Pro-Choice Network of W. N.Y.,

519 U.S. 357, 377 (1997) (“Leafletting and commenting on matters of public concern are classic forms

of speech that lie at the heart of the First Amendment . . . .”).

. See, e.g., CASS R. SUNSTEIN, DEMOCRACY AND THE PROBLEM OF FREE SPEECH 121–65

(1993); Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 IND. L.J. 1, 20–35

(1971); Alexander Meiklejohn, The First Amendment Is an Absolute, 1961 SUP. CT. REV. 245, 255.

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. d (AM. LAW INST. 1977) (noting that this rule

applies as a matter of both common law and constitutional law).

. Id. § 652D cmt. e.

. Id. § 652D cmt. f; see also id. § 652D cmt. h (noting that interest about a public figure “may

legitimately extend, to some reasonable degree, to further information concerning the individual and to

facts about him, which are not public and which, in the case of one who had not become a public figure,

2019] FACEBOOK V. SULLIVAN 53

Although courts recognize the concept of the public figure under the

disclosure tort, they do not discriminate between the voluntary and

involuntary public figure as they do in defamation law. The Restatement

does draw a distinction between voluntary and involuntary public figures,

but it is a distinction without a doctrinal difference.

The involuntary public

figure is an enigmatic and elusive character in defamation law,

but in

privacy law she is pervasive: she is the person who commits a crime or is

accused of it,

she is the victim or witness of crime and catastrophe,

and

she is even the hapless soul who inadvertently gets caught up in “events that

attract public attention.”

Thus, even if she does nothing to “thrust” herself

voluntarily into the vortex of a public controversy, the involuntary public

figure is “properly subject to the public interest” and subject to the same

constitutional rules as those who freely enter the public arena.

Her desire

to avoid the limelight is irrelevant, as is her lack of access to the media as a

means of engaging in counter-speech.

This complex web of considerations creates a paradoxically simple rule:

there is no privacy invasion when the fact disclosed is of “legitimate public

concern.” This is true for disclosures about private figures and disclosures

about public figures, whether they be voluntary or involuntary. The only

wrinkle is that some facts about public figures would not be of “legitimate

public concern” were they instead about private figures. But all that means

is that certain facts about public figures—even facts that would otherwise be

“private”—are matters of legitimate public concern because of the person’s

social status. The court’s status determination, then, serves simply as a proxy

for deciding whether a particular fact is of legitimate public concern—that

would be regarded as an invasion of his purely private life”).

. A voluntary public figure “voluntarily places himself in the public eye, by engaging in public

activities, or by assuming a prominent role in institutions or activities having general economic, cultural,

social or similar public interest, or by submitting himself or his work for public judgment.” Id. § 652D

cmt. e. Involuntary public figures, by contrast, “have not sought publicity or consented to it, but through

their own conduct or otherwise have become a legitimate subject of public interest”—“[t]hey have, in

other words, become ‘news.’” Id. § 652D cmt. f.

. See supra text accompanying notes 21–25.

. See, e.g., Miller v. Nat’l Broad. Co., 157 F. Supp. 240, 241 (D. Del. 1957); RESTATEMENT

(SECOND) OF TORTS § 652D cmt. f. (noting that criminals are involuntary public figures because they

“may not only not seek publicity but may make every possible effort to avoid it”).

. See, e.g., Jones v. Herald Post Co., 18 S.W.2d 972, 972–73 (Ky. 1929); RESTATEMENT

(SECOND) OF TORTS § 652D cmt. f.

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. f; see, e.g., Jacova v. S. Radio & Television

Co., 83 So. 2d 34, 40 (Fla. 1955).

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. f.

. See id.

54 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

is, whether it implicates the public’s ability to engage in self-governance.

This means that defining “legitimate public concern” carries a lot of

analytical weight; and indeed, a privacy claim will often rise and fall on this

determination alone. One might think that the definition is both descriptive

and circular: a matter is of legitimate public concern when the public is

concerned about the matter. The Restatement gestures at this conception

when it observes that “[i]ncluded within the scope of legitimate public

concern are matters of the kind customarily regarded as ‘news’” or matters

that have “popular appeal.”

This descriptive definition means that, “[t]o a

considerable extent, . . . publishers and broadcasters have themselves

defined the term, as a glance at any morning paper will confirm.”

In short,

the reporting of a fact in the news is highly probative, if not conclusive, of

its newsworthiness.

Broad deference to the news media is an idea with some constitutional

pedigree. Given that many privacy claims are brought against the press, the

vaulted status of speech on matters of public concern has sometimes been

called the “privilege to report news” or the “privilege to publicize

newsworthy matters.”

In Cox Broadcasting Corp. v. Cohn, the Court

barred a disclosure claim brought against a media company that reported a

rape victim’s name obtained from public court records.

In holding that “the

press cannot be sanctioned for publishing” information found in public

documents, the Court stressed that “reliance must rest upon the judgment of

those who decide what to publish or broadcast.”

Despite the sweeping language in some judicial opinions, the reality is

that many courts have shunned a purely descriptive definition of

newsworthiness that would yield entirely to the press. The deference is broad

but not absolute.

Much hinges on whether the public’s interest in knowing

a particular fact is “legitimate.” The legitimacy determination, in turn,

considers “the customs and conventions of the community,” meaning that

. Id. § 652D cmt. g.

. Id.

. Post, supra note 24, at 995 (quoting, respectively, Harry Kalven, Jr., Privacy in Tort Law—

Were Warren and Brandeis Wrong?, 31 LAW & CONTEMP. PROBS. 326, 336 (1966) and Virgil v. Time,

Inc., 527 F.2d 1122, 1128 (9th Cir. 1975)).

. Cox Broad. Corp. v. Cohn, 420 U.S. 469, 470, 496–97 (1975).

. Id. at 496; see also Erin C. Carroll, Making News: Balancing Newsworthiness and Privacy in

the Age of Algorithms, 106 GEO. L.J. 69, 77–81 (2017).

. See, e.g., RESTATEMENT (SECOND) OF TORTS § 652D cmt. h (“The extent of the authority to

make public private facts [about public figures] is not, however, unlimited.”).

2019] FACEBOOK V. SULLIVAN 55

“what is proper becomes a matter of the community mores.”

In the

language of the Restatement:

The line is to be drawn when the publicity ceases to be the giving of

information to which the public is entitled, and becomes a morbid and

sensational prying into private lives for its own sake, with which a

reasonable member of the public, with decent standards, would say that he

had no concern. The limitations, in other words, are those of common

decency, having due regard to the freedom of the press and its reasonable

leeway to choose what it will tell the public, but also due regard to the

feelings of the individual and the harm that will be done to him by the

exposure.

These principles were at the heart of a recent blockbuster privacy

lawsuit between former wrestler Hulk Hogan and the now-defunct news

organization Gawker. Hogan sued for invasion of privacy after Gawker

published an excerpted sex tape that showed him ensconced with his best

friend’s wife. Although multiple courts refused to enjoin publication of the

tape on the grounds that it was newsworthy,

a jury disagreed and awarded

Hogan massive damages.

100

Evidently, the jury concluded that publishing

the tape amounted to “morbid and sensational prying” into Hogan’s private

life that violated “common decency,” despite Hogan’s clear status as a public

figure.

101

C. INTENTIONAL INFLICTION OF EMOTIONAL DISTRESS

The constitutional concepts of public figures and newsworthiness have

developed in relation to a third tort: intentional infliction of emotional

distress (“IIED”). Under this tort, a plaintiff must show that the defendant

“intentionally or recklessly engaged in extreme and outrageous conduct that

caused the plaintiff to suffer severe emotional distress.”

102

When the

. Id.; see also id. § 652D cmt. g (explaining that the media’s broad leeway to define

newsworthiness must still accord “with the mores of the community”); Amy Gajda, The Present of

Newsworthiness, 50 NEW ENG. L. REV. 145, 145–46 (2016).

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. h.

. For background on the case, see generally Bollea v. Gawker Media, LLC, 913 F. Supp. 2d

1325 (M.D. Fla. 2012); Gawker Media, LLC v. Bollea, 170 So. 3d 125 (Fla. Dist. Ct. App. 2015); Gawker

Media, LLC v. Bollea, 129 So. 3d 1196 (Fla. Dist. Ct. App. 2014).

100

. Nick Madigan & Ravi Somaiya, Hulk Hogan Awarded $115 Million in Privacy Suit Against

Gawker, N.Y. TIMES (Mar. 18, 2016), https://www.nytimes.com/2016/03/19/business/media/gawker-

hulk-hogan-verdict.html [https://perma.cc/EEZ3-NYMF].

101

. See RESTATEMENT (SECOND) OF TORTS § 652D cmt. h; see also Garner v. Triangle Publ’ns,

97 F. Supp. 546, 549–50 (S.D.N.Y. 1951).

102

. See Snyder v. Phelps, 562 U.S. 443, 451 (2011).

56 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

“extreme and outrageous conduct” consists of actions protected by the First

Amendment, courts have crafted rules to limit the tort’s incursion on free

speech.

The Court’s first tussle with the IIED tort came in Hustler Magazine,

Inc. v. Falwell, a legal battle fit—and indeed destined

103

—for Hollywood.

104

Jerry Falwell, a nationally renowned Christian minister, sued Hustler

Magazine and its antagonistic publisher, Larry Flynt, after the magazine

spoofed an interview with Falwell in a liquor advertisement entitled “Jerry

Falwell talks about his first time.”

105

A jury accepted Falwell’s claim that the

parody caused him grave emotional harm, awarding him compensatory and

punitive damages, but the Court resoundingly rejected Falwell’s claim and

extended Sullivan’s rule to IIED claims brought by public figures.

106

Justice

William Rehnquist wrote for a unanimous Court that public figures must

satisfy the rigors of actual malice. The decision was grounded in the Sullivan

and Hill self-governance rationales about the need for “robust political

debate” and “the free flow of ideas and opinions on matters of public interest

and concern.”

107

Portions of the Hustler decision stressed the significance of Falwell’s

social status as a public figure. The Court spoke, for instance, of the First

Amendment right to be “critical of those who hold public office or those

public figures who are ‘intimately involved in the resolution of important

public questions or, by reason of their fame, shape events in areas of concern

to society at large.’”

108

But this focus on social status faded away the next

time the Court addressed the constitutionality of an IIED claim. In Snyder v.

Phelps, the father of a soldier killed in Iraq sued parishioners from the

Westboro Baptist Church who protested near his son’s funeral, with an array

of signs, including “God Hates Fags,” “Thank God for Dead Soldiers,” and

“Priests Rape Boys.”

109

The father’s leading arguments in the Supreme Court

revolved around the fact that he was a private figure who “took no action to

inject himself into a public debate” and “did nothing to obtain the status of a

celebrity or a public figure.”

110

103

. THE PEOPLE VS. LARRY FLYNT (Phoenix Pictures 1996).

104

. Hustler Magazine, Inc. v. Falwell, 485 U.S. 46, 50 (1988).

105

. Id. at 48.

106

. Id. at 48, 56.

107

. Id. at 50–51; see also supra notes 16–27 & 64–74 and accompanying text.

108

. Hustler, 485 U.S. at 51 (quoting Curtis Publ’g Co. v Butts, 388 U.S. 130, 164 (1967) (Warren,

C.J., concurring in result)).

109

. Snyder v. Phelps, 562 U.S. 443, 448 (2011).

110

. Brief for Petitioner at 34, Snyder v. Phelps, 562 U.S. 443 (2011) (No. 09-751); see also Reply

2019] FACEBOOK V. SULLIVAN 57

The Court, however, was unpersuaded. Rather than focusing on the

father’s social status, the Court held that the First Amendment’s application

to the father’s claim turned on whether the parishioners’ speech was “of

public or private concern.”

111

Chief Justice John Roberts explained that the

First Amendment protections are “less rigorous” when speech regulations

target matters of private concern because “[t]here is no threat to the free and

robust debate of public issues; there is no potential interference with a

meaningful dialogue of ideas; and the threat of liability does not pose the risk

of a reaction of self-censorship on matters of public import.”

112

In short, the

Court cared only about whether the underlying speech concerned issues that

the public needed to know in order to govern itself. Given the analytical

importance of determining whether speech is on a matter of public concern,

one might think that the Court would clearly define this constitutional

concept. Yet the Court candidly admitted that its boundaries “are not well

defined”

113

before describing it in disjunctive—and potentially

contradictory—terms as speech that is (1) “fairly considered as relating to

any matter of political, social, or other concern to the community,” or (2) “a

subject of legitimate news interest; that is, a subject of general interest and

of value and concern to the public.”

114

The definition might be descriptive

(what the public does know or wants to know) or normative (what the public

ought to know or needs to know). As applied to the father’s claim, the Court

concluded that the parishioners’ signs highlighted issues that are “matters of

public import” and as such gained First Amendment protection that blocked

liability under the IIED tort.

115

The Court briefly raised the issue of

directionality, suggesting that at least two of the signs—“You’re Going to

Hell” and “God Hates You”—could be “viewed as containing messages

related to Matthew Snyder or the Snyders specifically.”

116

The fact that the

Court even flagged the issue of directionality might imply that speech

targeting particular people raises different constitutional considerations, but

the Court never reached this question because “the overall thrust and

dominant theme” of the signs “spoke to broader public issues.”

117

Brief at 10–14, Snyder v. Phelps, 562 U.S. 443 (2011) (No. 09-751) (arguing at length that Snyder was a

private figure, not a public one).

111

. Snyder, 562 U.S at 451.

112

. Id. at 452 (internal quotation marks omitted) (quoting Dun & Bradstreet, Inc. v. Greenmoss

Builders, Inc., 472 U.S. 749, 760 (1985)).

113

. Id. at 452 (quoting City of San Diego v. Roe, 543 U.S. 77, 83 (2004) (per curiam)).

114

. Id. at 453 (citations omitted).

115

. Id. at 454.

116

. Id.

117

. Id.

58 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

Snyder, the Court’s most recent case concerning communication torts

and the First Amendment, reveals several important points about the

constitutional dynamics at play when the Court sets rules for defamation,

privacy, and IIED claims. All three regimes reflect the Court’s concern for

protecting robust public discourse—a concern animated by a self-

governance theory of the First Amendment. This common thread runs

through the jurisprudence for all three torts, even as the rules differ slightly

between them. The Court does not disparage the state’s interests in protecting

people from harmful speech, nor does it dispute that the speech at issue in

these cases in fact inflicted harm. Rather, the Court frames the various

constitutional rules in prophylactic terms—as creating the conditions

necessary for an ecosystem in which free speech can flourish. As we’ll see,

Facebook has viewed its own rulemaking in strikingly similar terms.

II. PUBLIC FIGURES AND NEWSWORTHINESS IN NEW

GOVERNANCE: CONTENT MODERATION AND FREE SPEECH AT

FACEBOOK

Facebook is the preeminent social network of the digital age. With over

two billion users, the platform hosts vast amounts of content shared by

people scattered across the globe.

118

Though we might conceive of Facebook

as a “New Governor” because of its power over online discourse, it is, of

course, a private company that need not satisfy the First Amendment when

policing its users’ speech.

119

Nor is Facebook bound by any “Constitution”

of its own.

120

Instead, Facebook implements a system of semi-public rules

called “Community Standards,”

121

which are effectively Facebook’s “laws”

that govern what users may say on the platform.

122

In order to implement the Community Standards, Facebook has

developed an immense bureaucratic system to moderate user content and

118

. Company Info, FACEBOOK NEWSROOM, https://newsroom.fb.com/company-info

[https://perma.cc/G7SJ-V4E2].

119

. See Klonick, supra note 5, at 1658–62. The question whether the First Amendment restricts

the government’s use of social media presents discrete doctrinal challenges that others have expertly

analyzed. See Lyrissa Lidsky, Public Forum 2.0, 91 B.U. L. REV. 1975, 1979–2002 (2011) (discussing

how the public-forum doctrine might apply to when government actors use social media); Helen Norton

& Danielle Keats Citron, Government Speech 2.0, 87 DENV. U. L. REV. 899, 899 (2010) (analyzing how

the government-speech doctrine might adapt given that government’s increasing reliance on social

media).

120

. Though perhaps it should be. See Klonick & Kadri, supra note 5.

121

. See Community Standards, FACEBOOK, https://www.facebook.com/communitystandards

[https://perma.cc/J7BQ-96E2].

122

. DAVID KAYE, SPEECH POLICE: THE GLOBAL STRUGGLE TO GOVERN THE INTERNET 23 (2019).

2019] FACEBOOK V. SULLIVAN 59

adjudicate disputes arising from that content. Because an enormous volume

of content is posted every day, Facebook cannot and does not proactively

police all violations of its rules. Automated detection of violations is quite

sophisticated and successful for various types of visual content (such as child

pornography) but less so for written content that poses “nuanced linguistic

challenges” (such as harassment and hate speech).

123

As a result, the platform

still relies on users to reactively flag speech that might violate its rules.

Content reported by users is placed into an online queue for review by human

content moderators—people working either directly for Facebook or as

contractors who are trained to apply Facebook’s rules and determine if

content violates the Community Standards.

124

The platform removes speech

found to be in violation; the rest remains.

125

Facebook’s first internal guidelines for content moderation were

created largely by Dave Willner in 2009, who then joined Jud Hoffman to

spearhead a small team that formalized and consolidated the ad hoc rules that

Facebook’s earliest content moderators had been using.

126

Ever since,

Facebook has devoted considerable attention to the rules and procedures it

uses to govern speech on the platform.

127

Somewhat like a common-law

legal system,

128

Facebook regularly adapts its Community Standards to

address changing circumstances, including new factual scenarios or

technologies; criticism or feedback from outside observers; changing norms

surrounding particular issues; and interventions from upper-level

123

. See Mark Zuckerberg, A Blueprint for Content Governance and Enforcement, FACEBOOK

(Nov. 15, 2018), https://www.facebook.com/notes/mark-zuckerberg/a-blueprint-for-content-governance-

and-enforcement/10156443129621634 [https://perma.cc/P4H6-6HVQ]; see also Community Standards

Enforcement Report, FACEBOOK (Nov. 2019), https://transparency.facebook.com/community-standards-

enforcement [https://perma.cc/S8FW-TXXV] (breaking down statistics for removal of nine different

types of content, including hate speech and child nudity and sexual exploitation).

124

. See Klonick, supra note 5, at 1630–48. For pathbreaking work on content moderation,

particularly on the experiences of the content moderators themselves, see generally SARAH T. ROBERTS,

BEHIND THE SCREEN: CONTENT MODERATION IN THE SHADOWS OF SOCIAL MEDIA (2019). And for

excellent exploration of platform policies from sociological and legal perspectives, see generally

TARLETON GILLESPIE, CUSTODIANS OF THE INTERNET: PLATFORMS, CONTENT MODERATION, AND THE

HIDDEN DECISIONS THAT SHAPE SOCIAL MEDIA (2018), and NICHOLAS SUZOR, LAWLESS: THE SECRET

RULES THAT GOVERN OUR DIGITAL LIVES (2019).

125

. See Klonick, supra note 5, at 1630–48.

126

. Telephone Interview with Jud Hoffman, Former Glob. Policy Manager, Facebook (Jan. 22,

2016); Telephone Interview with Dave Willner, Former Head of Content Policy, Facebook, & Charlotte

Willner, Former Safety Manager, User Operations, Facebook (Mar. 23, 2016). All interview notes are on

file with the authors.

127

. See Klonick, supra note 5, at 1630–48.

128

. See id. at 1645–47.

60 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

management.

129

Both Willner and Hoffman were heavily involved in

developing Facebook’s rules surrounding public figures and

newsworthiness—a history we turn to now.

130

A. CYBERBULLYING

Whereas the Supreme Court’s public-figure doctrine emerged from

claims of defamation,

131

Facebook’s rules surrounding public figures first

developed in response to claims about bullying.

132

In 2009, anti-

cyberbullying groups were pressuring Facebook to do more to protect

children from online abuse.

133

The problem, however, was that traditional

academic definitions of bullying seemed impossible to translate to online

content moderation. “How do we write a rule about bullying?” recounted

Willner.

134

What is bullying? What do you mean by that? It’s not just things that are

upsetting; it’s defined as a pattern of abusive or harassing unwanted

behavior over time that is occurring between a higher power [and] a lower

power. But that’s not an answer to the problem that resides in the

content—you can’t determine a power differential from looking at the

content. You often cannot even do it from looking at their profiles.

135

The apparent impossibility of employing a traditional definition of

bullying meant that Facebook had to make a choice. It could err on the side

of keeping up potentially harmful content, or it could err on the side of

removing all potential acts of bullying, even if some of the removed content

turned out to be benign. Faced with intense pressure from advocacy groups

and media coverage on cyberbullying, Facebook opted for the latter

approach, but with a caveat. The new presumption in favor of removing

speech reported to be “bullying” would apply only to speech directed at

129

. See, e.g., Simon Adler, Post No Evil, WNYC STUDIOS: RADIOLAB (Aug. 17, 2018),

https://www.wnycstudios.org/story/post-no-evil [https://perma.cc/J54X-5CWA]; Sheera Frenkel et al.,

Delay, Deny and Deflect: How Facebook’s Leaders Fought Through Crisis, N.Y. TIMES (Nov. 14, 2018),

https://nyti.ms/2DlsGPi [https://perma.cc/MD7T-LVDW].

130

. The following Sections contain excerpts from research and commentary discussed in Klonick,

supra note 5, at 1604–09.

131

. See supra Section I.A.

132

. Telephone Interview with Jud Hoffman, Former Glob. Policy Manager, Facebook (Mar. 6,

2018); Telephone Interview with Dave Willner, Former Head of Content Policy, Facebook (Mar. 7,

2018). On the distinct complications of criminalizing cyberbullying, see Lyrissa Lidsky & Andrea Pinzon

Garcia, How Not to Criminalize Cyberbullying, 77 MO. L. REV. 693 (2012).

133

. Telephone Interview with Dave Willner, supra note 132.

134

. Id.

135

. Id.

2019] FACEBOOK V. SULLIVAN 61

private figures. “What we said was, ‘Look, if you tell us this is about you,

and you don’t like it, and you’re a private individual—you’re not a public

figure—then we’ll take it down,’” said Hoffman.

136

“Because we can’t know

whether all of those other elements [of bullying] are met, we had to just make

the call to create a default rule for removal of bullying.”

137

Although Hoffman denies borrowing directly from the First

Amendment doctrine, his justification for creating this rule tracks some of

the reasoning in Sullivan and subsequent cases.

138

In order to preserve robust

public discourse on the platform, Hoffman’s team made the conscious

decision to treat certain targets of allegedly harmful speech differently on

account of their social status and the public interest in their doings.

139

According to Hoffman, this approach reflected Facebook’s mission

statement, which at that time was “Make the world more open and

connected.”

140

“Broadly, we interpreted ‘open’ to mean ‘more content.’ Yes,

that’s a bit of a ‘free speech’ perspective, but then we also had a concern

with things like bullying and revenge porn,” Hoffman recalled.

141

“But while

trying to take down that bad content, we didn’t want to make it impossible

for people to criticize the president or a person in the news. It’s important

there’s a public discussion around issues that affect people, and this is how

we drew the line.”

142

In trying to resolve these dilemmas, Hoffman and his colleagues sought

to “focus on the mission” of Facebook rather than adopt “wholesale . . . a

kind of U.S. jurisprudence free-expression approach.”

143

They quickly

realized, however, that the mission had to be balanced against competing

interests such as users’ safety and the company’s bottom line.

144

While

Hoffman and Willner were at Facebook, the balance was often struck in

favor of “leaving content up,” but they were always searching for new ways

to address concerns about harmful speech.

145

“We felt like Facebook was the

most important platform for this kind of communication, and we felt like it

136

. Telephone Interview with Jud Hoffman, supra note 126.

137

. Id.

138

. Telephone Interview with Jud Hoffman, supra note 132.

139

. Id.

140

. Id.

141

. Id.

142

. Id.

143

. Id.

144

. Id.

145

. Id.

62 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

was our responsibility to figure out an answer to this,” said Hoffman.

146

The policy required a way to determine if someone was a public figure.

When a piece of content was flagged for bullying, Facebook told its

moderators to use the news aggregator Google News.

147

If the person

allegedly being bullied appeared in a Google News search, moderators

would consider her a public figure—and the content would stay up.

148

Although some prominent people had blue “verification” checkmarks or

“public figure” titles on their Facebook pages, these symbols were not

actually part of the metric used to determine whether someone was a public

figure under Facebook’s Community Standards.

149

As Willner put it,

The blue checkmark was totally separate from the “public figure”

designation—an individual user getting a checkmark was much more

arbitrary and came from a totally different team. Ultimately, it had no

impact on how you were enforced against as a private or public figure—it

just meant that you’re one of the cool kids.

150

Facebook’s use of Google News to make public-figure determinations

provided the platform with a tool that its moderators could use quickly and

consistently. The ability to implement the underlying policy on such a mass

scale was, of course, its virtue, at least in Facebook’s eyes. But it was not

without its drawbacks, as even Facebook’s policymakers recognized. If

anyone who appeared in a Google News search became a public figure, there

was no way to know whether they had voluntarily entered the public eye.

This issue often arose when people were caught up in terrible circumstances

or publicly shamed in a way that went viral. As Willner put it, “you can think

of them as involuntary public figures, but another way of saying it might be

to think of them as sympathetic public figures.”

151

Whether it was fair to

apply the same bullying rules to these people was a question that Google’s

algorithm was simply unequipped to answer.

But not all involuntary public figures were plainly sympathetic.

Consider, for example, the case of Casey Anthony, who became a household

name after being accused of murdering her daughter in the “Social-Media

Trial of the Century.”

152

Or recall the furor surrounding Rolling Stone’s

146

. Id.

147

. Telephone Interview with Dave Willner, supra note 132.

148

. Id.

149

. Id.

150

. Id.

151

. Id.

152

. See John Cloud, How the Casey Anthony Murder Case Became the Social-Media Trial of the

2019] FACEBOOK V. SULLIVAN 63

article about an alleged rape at the University of Virginia, a story that was

later retracted and dubbed “a complete crock.”

153

In both episodes, it was

difficult for Facebook to identify the “sympathetic” parties caught up in the

online firestorm that engulfed the platform, especially as the winds changed.

With Casey Anthony, Willner recalled that Facebook felt torn between

accepting the court’s “not guilty” verdict and recognizing the hard reality

that “everyone in America thinks she killed her kid.”

154

Similarly, as Willner

recounted:

With the Rolling Stone story, when it starts, we believe the victims, and

that people shouldn’t say mean things about the victims, but then it turns

out all that’s not true—the victim there was actually the “bad” person. But

where along in the journey of learning about that entire story do people’s

minds shift and what do you decide to protect?

155

Despite the fact Google News could hardly provide the nuance to handle

these edge cases, the company stood by its use as the best way to strike the

balance between promoting free speech and remedying harmful speech.

In moderating content related to public figures, Facebook’s

policymakers began to blur the lines between social status and

newsworthiness, just as judges have done when applying the First

Amendment to privacy torts.

156

Willner reflected that “calling the exception

[an exception for] ‘public figures’ was probably a mistake—a more accurate

way of thinking about it is as a newsworthy person.”

157

All that a Google

News search could tell moderators was that someone’s name had appeared

in a news source—it could not reliably reveal the person’s true social status,

the reputability of the source, the veracity of the story, or the genesis of the

controversy. This framework meant that Facebook ran into many of the same

issues that have plagued courts in defining the boundaries of

newsworthiness. As Willner recounted, he saw “newsworthiness” as

representing a normative “post-hoc judgment that applies to the content as

it’s supposed to be—to be able to accurately assess it at the time literally

Century, TIME (June 16, 2011), http://content.time.com/time/nation/article/0,8599,2077969,00.html

[https://perma.cc/PB3Y-CLTU].

153

. Erik Wemple, Charlottesville Police Make Clear that Rolling Stone Story Is a Complete Crock,

WASH. POST (Mar. 23, 2015), https://www.washingtonpost.com/blogs/erik-wemple/wp/2015/03/23/

charlottesville-police-make-clear-that-rolling-stone-story-is-a-complete-crock [https://perma.cc/94PA-

532Q]; see also Ravi Somaiya, Rolling Stone Article on Rape at University of Virginia Failed All Basics,

Report Says, N.Y. TIMES (Apr. 5, 2015), https://nyti.ms/1NMyThP [https://perma.cc/C5ZM-ZFRM].

154

. Telephone Interview with Dave Willner, supra note 132.

155

. Id.

156

. See supra Section I.B.

157

. Telephone Interview with Dave Willner, supra note 132.

64 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

calls for time travel.”

158

Facebook’s engineers, while brilliant, had not realized that particular

Sci-Fi dream, so the platform had to settle for a descriptive concept that

included everything published in the “news.” But this approach was not

without its faults, and both Willner and Hoffman foresaw problems created

by the erosion of traditional media, the rise of self-publishing, and the

opportunities for mass amplification and virality offered by social media.

“Now that everyone is their own newsroom, it’s revealed that the emperor

has no clothes,” recounted Willner.

159

“We don’t like how democratized

reporting has gotten—everyone can be their own news station, and it’s very

upsetting to people.”

160

Nevertheless, given the volume of content on

Facebook and the subjectivity and unpredictability that would afflict case-

by-case newsworthiness determinations, the platform saw no viable

alternative to this broad deference to an increasingly unprofessional media

ecosystem. Moderators needed a tool to make quick and mechanical

decisions, and a normative newsworthiness standard was too relative to

measure consistently. As Hoffman remarked, “When we talk about a

newsworthiness standard, what do we mean? Newsworthy to who and how

many? If you don’t establish a minimum number of people, then random

gossip is newsworthy. How is somebody sitting in one of the [Facebook]

operations places . . . going to decide that?”

161

For similar reasons, both

Hoffman and Willner opposed building a general exception into the

Community Standards to prevent removal of all “newsworthy” content.

Facebook’s approach to this issue would develop on a slightly different track.

B. NEWSWORTHY CONTENT

For most of Facebook’s history, the platform made no exceptions for

content that violated Community Standards but was newsworthy.

162

Overtly

sexual, graphically violent, or “extremist” content would be taken down

regardless of whether it had cultural or political significance.

163

This was a

deliberate choice made by Hoffman and Willner, but the policy came under

increasing pressure in recent years.

164

158

. Id.

159

. Id.

160

. Id.

161

. Telephone Interview with Jud Hoffman, Former Glob. Policy Manager, Facebook (Mar. 8,

2018).

162

. Id.

163

. Telephone Interview with Dave Willner, supra note 132.

164

. Id.;

Telephone Interview with Jud Hoffman, supra note 132.

2019] FACEBOOK V. SULLIVAN 65

Members of the policy team recall an incident in 2013 as a turning point

toward the creation of an exception for newsworthy content.

165

In the wake

of the Boston Marathon bombing, a graphic image of a man in a wheelchair

began to circulate on Facebook.

166

The man was being wheeled away from

the carnage, one leg ripped open below the knee to reveal a long, bloody

bone.

167

What made this moderation question so fascinating was that there

were three versions of the photograph.

168

One was cropped so that the leg

was not visible.

169

A second was a wide-angle shot in which the leg was

visible but less obvious.

170

The third, and most controversial, clearly showed

the man’s “insides on the outside”—the content-moderation team’s

shorthand rule for when content was graphically violent.

171

Despite the fact

that multiple media outlets had published all three photographs, Facebook

removed any links to or images of the third version.

172

“Philosophically, if

we were going to take the position that [‘insides on the outside’] was our

definition of gore and we didn’t allow gore, then just because it happened in

Boston didn’t change that,” remembers one of the team members on call that

day.

173

Policy executives at Facebook disagreed, however, and reinstated all

such posts on the grounds of newsworthiness.

For some members of the policy team, who had spent years trying to

create administrable rules, the imposition of such an exception represented a

radical departure from the company’s commitment to procedural

consistency. Some of their complaints echo the Gertz Court’s rationale for

reining in the plurality’s rule in Rosenbloom.

174

In his opinion for the Court

in Gertz, Justice Lewis Powell worried openly about allowing “judges to

decide on an ad hoc basis which publications address issues of ‘general or

public interest’ and which do not.”

175

Many at Facebook worried similarly

that “newsworthiness as a standard is extremely problematic: the question is

really one of ‘newsworthy to whom?’ and the answer to that is based on ideas

165

. Adler, supra note 129.

166

. Id.

167

. Id.

168

. Id.

169

. Id.

170

. Id.

171

. Id.

172

. Id.

173

. Telephone Interview with Anonymous, Former Member of Policy Team, Facebook (Aug. 28,

2018).

174

. See supra notes 18–22 accompanying text.

175

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 346 (1974).

66 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

of culture and popularity.”

176

The result, some feared, would be a mercurial

exception that would, moreover, privilege American users’ views on

newsworthiness to the potential detriment of Facebook’s users in other

countries.

177

Although there were other one-off exceptions made for incidents like

the Boston Marathon photograph, Facebook’s internal content-moderation

policies continued to have no general exception for newsworthiness until

September 2016, when a famous Norwegian author, Tom Egeland, posted a

well-known historical picture to his Facebook page.

178

The photograph, “The

Terror of War,” depicts a nine-year-old Vietnamese girl naked in the street

after a napalm attack, and for this reason the photo is often called “Napalm

Girl.”

179

In part because of its graphic nature, the photo was a pivotal piece

of journalism during the Vietnam War.

180

Nonetheless, it violated

Facebook’s Community Standards.

181

Accordingly, Facebook removed the

photo and suspended Egeland’s account. Because of Egeland’s stature, the

takedown itself received news coverage. Espen Egil Hansen, the editor-in-

chief of the Norwegian newspaper Aftenposten, published a “letter” to

Facebook founder Mark Zuckerberg on Aftenposten’s front page calling for

Facebook to take a stand against censorship. Hours later, Facebook’s chief

operating officer Sheryl Sandberg admitted that the company had made a

mistake and promised that the rules would be rewritten to allow for posting

of the photo.

182

Shortly thereafter, Facebook issued a press release

underscoring the company’s commitment to “allowing more items that

people find newsworthy, significant, or important to the public interest—

even if they might otherwise violate [its] standards.”

183

176

. Telephone Interview with Anonymous, supra note 173.

177

. See Adler, supra note 129.

178

. Id.

179

. Id.

180

. See Kate Klonick, Facebook Under Pressure, SLATE (Sept. 12, 2016, 2:48 PM),

http://www.slate.com/articles/technology/future_tense/2016/09/facebook_erred_by_taking_down_the_n

apalm_girl_photo_what_happens_next.html [https://perma.cc/P2WX-YZ5U].

181

. The photo was likely removed because of the nudity, not because it was child pornography.

See Kjetil Malkenes Hovland & Deepa Seetharaman, Facebook Backs Down on Censoring ‘Napalm Girl’

Photo, WALL ST. J. (Sept. 9, 2016, 3:07 PM), http://www.wsj.com/articles/norway-accuses-facebook-of-

censorship-over-deleted-photo-of-napalm-girl-1473428032 [https://perma.cc/5GNF-DSKD].

182

. See Claire Zillman, Sheryl Sandberg Apologizes for Facebook’s ‘Napalm Girl’ Incident, TIME

(Sept. 13, 2016), http://time.com/4489370/sheryl-sandberg-napalm-girl-apology [https://perma.cc/L8ME

-Z637].

183

. Joel Kaplan & Justin Osofsky, Input From Community and Partners On Our Community

Standards, FACEBOOK NEWSROOM (Oct. 21, 2016), https://newsroom.fb.com/news/2016/10/input-from-

community-and-partners-on-our-community-standards [https://perma.cc/X48L-TYSD].

2019] FACEBOOK V. SULLIVAN 67

The “Terror of War” incident led Facebook to start looking more

broadly at how it evaluated newsworthiness outside the context of

cyberbullying. “After the ‘Terror of War’ controversy, we realized that we

had to create new rules for imagery that we’d normally want to disallow, but

for context reasons that policy doesn’t work,” said Peter Stern, head of

Product Policy Stakeholder Engagement at Facebook.

184

According to Stern,

this policy shift “led us to think about newsworthiness across the board.”

185

He acknowledged that Facebook has two considerations when making

newsworthiness determinations: “Safety of individuals on the one hand and

voice on the other.”

186

But what exactly does “voice” mean in this context?

Here, again, Facebook has increasingly aligned itself with the courts’ view

of the relationship between free speech and self-governance. “When

someone enters the public eye,” Stern explained, “we want to allow a broader

scope of discussion.”

187

C. CONTEMPORARY APPROACHES

Over the last two years, Facebook’s content-moderation policies have

evolved to become somewhat less mechanical and more nuanced. For

example, Facebook modified its rules on bullying and harassment of public

figures in 2018. In place of its blanket rule that public figures could never

enjoy protection from bullying, the platform adopted more context-sensitive

standards to address some forms of harmful speech about prominent people.

“Our new policy does not allow certain high-intensity attacks, like calls for

death, directed at a certain public figure,” members of the Facebook policy

team reported on a recent call.

188

In the past, they explained, a statement such

as “Kim Kardashian is a whore” would never be removed for bullying or

harassment (whereas a statement calling a private individual a “whore”

would be). But now, Facebook will remove some speech directed at public

figures when it is posted on their own pages or accounts, depending on the

severity of the language.

189

Details of how moderators will make these

decisions are still vague, but it appears that the platform is beginning to draw

lines based on both substance (whether the speech is particularly vicious)

184

. Telephone Interview with Peter Stern, Head of Prod. Policy Stakeholder Engagement,

Facebook (Mar. 7, 2018).

185

. Id.

186

. Id.

187

. Id.

188

. Televideo Interview with Idalia Gabrielow & Peter Stern, Policy Risk Team, Facebook (Aug.

15, 2018).

189

. Id.

68 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

and directionality (whether the speech is targeted to reach particular public

figures).

190

Under this new regime, the platform has also tweaked its methods for

determining someone’s status as a public figure. Google News remains an

important instrument in the moderator’s toolkit—the platform will bestow

the label on people mentioned in multiple news stories within a certain

timeframe—but there are now other ways to qualify.

191

Regardless of what

Google’s algorithm regurgitates, Facebook now counts among its public

figures people elected or assigned through a political process to a

government position; people with hundreds of thousands of fans or followers

on a social-media account; and people employed by a news organization or

who speak publicly.

192

Facebook’s current policies on newsworthy content are somewhat

harder to pin down. Unlike the term “public figures,” which Facebook still

uses primarily for its bullying standards, “newsworthiness” is now a possible

exception to all of the company’s Community Standards.

193

And unlike the

public-figure designations made in the bullying context, newsworthiness

determinations do not rely on news aggregators and algorithms.

194

Instead,

Facebook employees review claims about possible newsworthy content on a

case-by-case basis.

195

In deciding whether to keep up otherwise-removable content because

of its newsworthiness, Facebook officials stress that they weigh the value of

“voice” against the risk of harm.

196

Assessments of harm are informed by the

nature as well as the substance of the objectionable content.

197

Hateful

speech on its own, for instance, might be seen as less harmful than a direct

call to violence.

198

Facebook officials maintain, however, that most of the

newsworthiness decisions relate to nudity. Difficult decisions include what

to do about nudity in public protests. “Just a few years ago, we took that

190

. In this sense, Facebook and the court might again be following a similar path. See supra notes

100−01 and accompanying text.

191

. Televideo Interview with Idalia Gabrielow & Peter Stern, supra note 188.

192

. Id.

193

. Id.

194

. Id.

195

. Id.

196

. Telephone Interview with Peter Stern, supra note 184.

197

. Televideo Interview with Ruchika Budhraja, David Caragliano, Idalia Gabrielow & Peter

Stern, Policy Risk Team, Facebook (Oct. 4, 2018).

198

. Id.

2019] FACEBOOK V. SULLIVAN 69

down,” stated David Caragliano, a policy manager at Facebook.

199

“But it’s

really important to leave this up consistent with our principles of voice.

That’s led to a policy change that’s now at scale for the platform.”

200

The

non-hateful, nonviolent expressive conduct of public protesters, it seems,

will today almost always be considered newsworthy and therefore will not

be taken down, even if it runs afoul of other Community Standards.

Compared to the thousands of day-to-day decisions made by hordes of

content moderators who compare content to rules, the “how” and “who”

behind Facebook’s newsworthiness determinations are more obscure. It is

unclear, for example, how a question of possible newsworthiness climbs the

Facebook policymaking ladder to become a new “policy change . . . at scale

for the platform,” let alone who makes that crucial call.

201

The lack of

transparency and accountability gives little comfort to those who worry

about the mercurial and subjective nature of newsworthiness determinations

at Facebook.

This might be about to change. In November 2018, in “A Blueprint for

Content Governance and Enforcement,” Mark Zuckerberg informed the

public that he “increasingly [has] come to believe that Facebook should not

make so many important decisions about free expression and safety on [its]

own.”

202

He announced that the platform would create an “Independent

Governance and Oversight” committee to make decisions about the kinds of

content users could post on the site.

203

Some have imagined this new body

as a “Supreme Court” of Facebook.

204

Indeed, Zuckerberg himself used this

analogy on an April 2018 podcast:

You can imagine some sort of structure, almost like a Supreme Court, that

is made up of independent folks who don’t work for Facebook, who

ultimately make the final judgment call on what should be acceptable

speech in a community that reflects the social norms and values of people

all around the world.

205

199

. Id.

200

. Id.

201

. See id.

202

. Zuckerberg, supra note 123.

203

. See Klonick & Kadri, supra note 5.

204

. Id.; evelyn douek, Facebook’s New ‘Supreme Court’ Could Revolutionize Online Speech,

LAWFARE (Nov. 19, 2018, 3:09 PM), https://www.lawfareblog.com/facebooks-new-supreme-court-

could-revolutionize-online-speech [https://perma.cc/PP5C-F668]; Kadri, How Supreme a Court?, supra

note 5.

205

. Ezra Klein, Mark Zuckerberg on Facebook’s Hardest Year, and What Comes Next, VOX (Apr.

2, 2018, 6:00 AM), https://www.vox.com/2018/4/2/17185052/mark-zuckerberg-facebook-interview-

fake-news-bots-cambridge [https://perma.cc/YTE5-JREW].

70 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

The platform is now in the process of soliciting feedback on this idea and

intends to establish the tribunal by the year’s end.

206

Whatever form Facebook’s Supreme Court takes, it will surely face

questions about the scope and application of Facebook’s “doctrine”

concerning public figures and newsworthiness. Actual courts, meanwhile,

will continue to grapple with these concepts and may face similar questions

as they confront challenges posed by tort claims arising from online speech.

With this in mind, Part III now considers the lessons to be learned by

comparing the public and private approaches to public figures and

newsworthiness. The constitutional law created by the Old Governors

greatly influenced the content moderation implemented by the New

Governors, and the New Governors’ experiences might now enlighten the

Old Governors in turn.

III. FACEBOOK VERSUS SULLIVAN: LESSONS FOR COURTS AND

PLATFORMS IN THE DIGITAL AGE

As we have seen, the two governance systems now used to adjudicate

complaints about harmful speech—tort lawsuits in courts and content

moderation on platforms—share some similarities. Both have developed

rules that seek to regulate harmful speech while protecting robust public

discourse that is essential to self-governance. To strike this balance, courts

and platforms have developed special rules that depend on the content of the

speech and the parties involved in the dispute. In the courts, defamation law

gives plaintiffs recourse for untruthful speech about them, but places a

substantially higher burden on plaintiffs who are public figures.

207

Facebook, an anti-bullying policy allows users to remove malicious speech

about them, but users who are public figures can rarely avail themselves of

this option.

208

In the courts, privacy and IIED law allow plaintiffs to hold

defendants liable for certain privacy invasions or outrageous conduct, except

when the underlying speech is deemed to be of legitimate public interest.

209

On Facebook, users can request that disturbing content like graphically

violent or hateful speech be taken down, except when the content is of

legitimate public interest.

210

Both judges and Facebook policymakers justify

these rules in similar terms, citing the importance of protecting free speech

206

. Zuckerberg, supra note 123.

207

. See supra Section I.A.

208

. See supra Section II.A.

209

. See supra Sections I.B−.C.

210

. See supra Section II.B.

2019] FACEBOOK V. SULLIVAN 71

and the legitimate public interest in discussion about people who are

powerful, famous, or at the forefront of a particular controversy.

211

The observation that First Amendment concepts like public figures and

newsworthiness have wended their way into Facebook’s content-moderation

policies is interesting as a descriptive matter. It is part of a broader story

about how American laws and norms have gained influence across the globe

as these potent American companies have expanded their reach abroad.

212

Even though Facebook need not adhere to the First Amendment, its content-

moderation policies were largely developed by American lawyers trained

and acculturated in American free-speech norms, and it seems that this

cultural background has affected their thinking.

213

By accurately describing

some of Facebook’s internal processes for moderating content in the way

that it does, we can better understand Facebook’s power inside and outside

of the United States—and perhaps some of the external resistance to it, as

non-Americans bristle at the exportation of foreign values.

214

But comparing these public and private governance systems provides

more than mere descriptive interest: it also teaches important lessons about

the challenges posed in the new speech ecosystem created by digital

discourse today. As an initial matter, the comparison tests intuitions about

why there are different standards concerning public figures—whether

through the courts or private platforms—and how those standards relate to

protections for newsworthy speech. With these rationales exposed, the

comparison also reveals problems with how courts and platforms have

implemented their standards for public figures and newsworthiness,

especially in an era when online speech often influences public discourse.

Finally, the comparison lays the groundwork to address some of those

problems—a task briefly undertaken in this Article’s conclusion.

211

. Compare, e.g., N.Y. Times Co. v. Sullivan, 376 U.S. 254, 270 (1964) (justifying heightened

protection for speech about public figures because “debate on public issues should be uninhibited, robust,

and wide-open”), with Telephone Interview with Peter Stern, supra note 184 (explaining that Facebook

“want[s] to allow a broader scope of discussion” once “someone enters the public eye”).

212

. See Danielle Keats Citron, What to Do About the Emerging Threat of Censorship Creep on the

Internet, CATO INST. (Nov. 28, 2017), https://www.cato.org/publications/policy-analysis/what-do-about-

emerging-threat-censorship-creep-internet [https://perma.cc/ATU4-DCH6].

213

. See Klonick, supra note 5, at 1621–22.

214

. It is worth noting, however, that similar concepts exist in jurisprudence outside of the United

States. For various international cases dealing with these examples, see generally Campbell v. MGN Ltd.

[2004] 2 AC 457 (HL) (Eng.); Reynolds v. Times Newspapers [1999] 4 All ER 609 (HL) (Eng.); Lingens

v. Austria, App. No. 9815/82, 8 Eur. H.R. Rep. 407 (1986); Tammer v. Estonia, App. No. 41205/98, 37

Eur. H.R. Rep. 43 (2001); Von Hannover v. Germany, 2004-VI Eur. Ct. HR 294.

72 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

A. THE RATIONALES BEHIND THE RULES

As this Article has revealed so far, courts and platforms have proffered

various rationales for creating rules that protect newsworthy speech or

hamstring public figures subjected to harmful speech. Unpacking these ideas

tests intuitions about whether those rationales are both descriptively sound

and normatively desirable. This scrutiny is particularly important today as

the challenges posed by the digital age force courts to rethink old doctrines

and lead the public to demand more from the private platforms that now

govern speech online.

There have historically been three reasons why public figures have

faced harsher standards than private figures when seeking recourse for

harmful speech. The first is that public figures are supposedly “less

vulnerable” because they have greater access to “channels of effective

communication” to rebut the harmful speech.

215

This rationale—which the

Court has also described as a public figure’s greater ability to perform “self-

help”—rests on an empirical postulate: that there is a meaningful difference

between the abilities of public and private figures to engage in counter-

speech that in some way redresses the harm.

216

This postulate explains, at

least in part, why the Court has adopted this rationale in the defamation

context only. When the harm caused by speech stems from untruth that

tarnishes someone’s reputation, counter-speech can be an effective tool to

rebut the lie and thereby address the injury.

217

But in the context of privacy

or emotional harms caused by speech, counter-speech does little to

ameliorate damage, which helps explain why a plaintiff’s status as a public

or private figure does little to change his or her legal rights when bringing

privacy or IIED claims.

218

It might also explain why Facebook’s

policymakers never raise this rationale in explaining their public-figure rules

related to cyberbullying, which again involves an array of speech harms that

are not easily redressed through counter-speech.

219

Indeed, both offline and

online, the harm caused by the types of speech that usually trigger privacy,

IIED, and cyberbullying concerns might actually increase with counter-

speech, either by amplifying the speech through more publicity or by forcing

215

. Wolston v. Reader’s Digest Ass’n, 443 U.S. 157, 164 (1979).

216

. Id.; see also Time, Inc. v. Hill, 385 U.S. 374, 391 (1967) (declining to base a constitutional

rule for privacy on “the distinction . . . between the relative opportunities of the public official and the

private individual to rebut” harmful speech because defamatory speech creates a different type of harm).

217

. Wolston, 443 U.S. at 164; see also Lyrissa Barnett Lidsky, Defamation, Reputation, and the

Myth of Community, 71 WASH. L. REV. 1, 7 (1996).

218

. See supra Sections I.B−.C.

219

. See supra Section II.A.

2019] FACEBOOK V. SULLIVAN 73

a victim to re-create or re-experience the harm in order to speak out against

it.

220

The second rationale for applying harsher rules to public figures is that

they are “less deserving” of protection because they assume the risk of

possible negative attention when they put themselves in the public eye.

221

The Court has explained that this is a “normative” rationale that depends on

the idea of voluntariness—the harsher rule is justified because public figures,

unlike private figures, have “voluntarily exposed themselves to increased

risk of injury.”

222

Once again, this reasoning appears crucial only in

defamation jurisprudence where the Court has repeatedly cabined the scope

of public-figure status by analyzing whether the plaintiff voluntarily “thrust”

herself into the vortex of a public controversy.

223

(And while the Court flirted

with the idea of an “involuntary” public figure, the justices have

conspicuously left this character undefined and undiscovered in the forty

years since he or she appeared hypothetically in Gertz.

224

) In privacy and

IIED law, meanwhile, this rationale has no purchase: for the former, courts

draw no doctrinal distinction between voluntary and involuntary public

figures who bring claims for public disclosure of private facts;

225

for the

latter, the Court in Snyder refused to give analytical weight to the plaintiff’s

status as a private figure involuntarily caught up in a newsworthy protest.

226

Facebook, too, has neglected to adopt this rationale, though it is unclear

whether this is a decision based on principle or practicality.

227

Early on, some

Facebook policymakers expressed concern about how the platform’s

algorithmic approach could not test for voluntariness, yet they still adopted

that approach.

228

The final reason to treat public figures differently rests on the idea that

their prominence makes them subjects of “legitimate public concern.” Under

this rationale, which hews most closely to the self-governance theory of the

220

. Julie E. Cohen, Law for the Platform Economy, 51 U.C. DAVIS L. REV. 133, 149–50 (2017)

(“Efforts to remove hurtful material typically backfire by drawing additional attention to it, intensifying

and prolonging the unwanted exposure.”).

221

. Wolston, 443 U.S. at 164.

222

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 345 (1974).

223

. See, e.g., Hutchinson v. Proxmire, 443 U.S. 111, 135 (1979).

224

. Supra note 41 and accompanying text.

225

. Compare RESTATEMENT (SECOND) OF TORTS § 652D cmt. e (AM. LAW INST. 1977)

(commenting on voluntary public figures), with id. § 652D cmt. f (commenting on involuntary public

figures).

226

. See Snyder v. Phelps, 562 U.S. 443, 474–75 (2011) (Alito, J., dissenting).

227

. See Telephone Interview with Dave Willner, supra note 132.

228

. Id.

74 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

First Amendment, the social status of public figures serves as a proxy for

their newsworthiness, and public figures thus face harsher standards for the

sake of cultivating robust public discourse.

229

This justification is most

visible in the standards surrounding privacy law, which immunize disclosure

of certain sensitive facts as matters of “legitimate public concern” only when

they relate to a public figure.

230

Facebook’s rules surrounding public figures

have a similar flavor. Using Google News to make status determinations

means that a person becomes a public figure on Facebook simply by

appearing in a news story. This mechanism tells the platform nothing

material about the person, aside from the fact that at least one news source

decided that the person was newsworthy. This is, of course, a purely

descriptive conception of newsworthiness—the person is newsworthy

because they appear in the news—but it nonetheless tracks the proxy

rationale to some extent.

This discussion reveals that these various rationales can take different

forms: they can be grounded in descriptive claims about public figures, such

as the notion that prominent people can more easily engage in effective self-

help; or they can stem from normative considerations, including ideas of

fairness, risk assumption, and legitimate scope of public curiosity.

Ultimately, courts and platforms have used these descriptive and normative

rationales to craft rules aimed at creating a speech ecosystem that preserves

robust debate and protects against harmful speech. Whether they have done

so effectively is another question—one that the following Section now

attempts to answer.

B. JUDGING THE GOVERNORS, NEW AND OLD

Drawing out the comparisons between the public and private

approaches to newsworthiness and public figures can teach valuable lessons

about free speech in terms of both constitutional law and content moderation.

This Section distills three insights that should enlighten both courts and

platforms.

231

Although Facebook policymakers may have channeled

doctrines created by courts, the platform’s algorithmic approach fails to

implement important normative protections baked into the courts’

229

. See, e.g., Gertz v. Robert Welch, Inc, 418 U.S. 323, 345 (1974) (explaining that, in defamation

law, public figures must satisfy a heightened burden because of their “roles of especial prominence in the

affairs of society”).

230

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. d.

231

. This Section incorporates and builds upon material in Kate Klonick, Facebook v. Sullivan,

KNIGHT FIRST AMEND. INST. (Oct. 1, 2018), https://knightcolumbia.org/content/facebook-v-sullivan

[https://perma.cc/W9R8-7K9B].

2019] FACEBOOK V. SULLIVAN 75

jurisprudence to justify harsher rules for public figures. Courts and platforms

should also rethink their approaches to defining public figures in light of new

phenomena created by social media, particularly given how virality alters

ideas about voluntariness. Finally, platforms could assuage concerns raised

by their ad hoc newsworthiness determinations by adopting structural

changes to become more like the court system.

1. The Inaccuracies and Injustices of Algorithmic Authority

The first lesson learned from a comparative analysis stems from

Facebook’s approach to unearthing public figures. Facebook’s use of Google

News to determine whether a person is a public figure provides a vivid

illustration of the problems raised when such definitions are outsourced

purely to the media marketplace. Although the platform has suggested that

this policy may be changing slightly, Facebook’s method for ascertaining

“public figure” status has traditionally turned on the presence or absence of

an individual’s name in news search results, which are effectively an

averaging algorithm of media outlets’ publication decisions. This runs

straight into the threat of what Clay Shirky has called “algorithmic

authority,” insofar as “an unmanaged process of extracting value from

diverse, untrustworthy sources” is treated as authoritative without any

human second-guessing or vouching for the validity of the outcome.

232

As commentators have pointed out for over fifty years in a closely

related context, if “newsworthiness” is defined solely in terms of news

outlets’ publication decisions, then granting a special legal privilege for

newsworthy content is liable to swallow torts such as invasion of privacy.

“The publisher has almost certainly published any given report because he

judged it to be of interest to his audience, and believed that it would

encourage them to purchase his publications in anticipation of more of the

same,” a student comment observed in 1963.

233

“A plaintiff in a privacy

action would thus have lost almost before he started.”

234

Partly for this

reason, courts making these determinations have considered a range of

232

. A Speculative Post on the Idea of Algorithmic Authority, CLAY SHIRKY (Nov. 15, 2009),

http://www.shirky.com/weblog/2009/11/a-speculative-post-on-the-idea-of-algorithmic-authority

[https://perma.cc/64DR-P4TP]; see also James Grimmelmann, The Platform is the Message, 2 GEO. L.

TECH. REV. 217, 217 (2018) (observing “the disturbing, demand-driven dynamics of the Internet today,

where any desire, no matter how perverse or inarticulate, can be catered to by the invisible hand of an

algorithmic media ecosystem that has no conscious idea what it is doing”).

233

. Comment, The Right of Privacy: Normative-Descriptive Confusion in the Defense of

Newsworthiness, 30 U. CHI. L. REV. 722, 725 (1963).

234

. Id.

76 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

factors

235

and, especially in recent years, have been unwilling to defer

entirely to the media.

236

They have, in other words, refused to adopt a purely

descriptive conception of newsworthiness that abdicates definitional

responsibilities to the media.

The problems inherent in Facebook’s approach are exacerbated by two

features of the new speech ecosystem that platforms have themselves helped

create. The first is that the nature of these platforms can, in some sense, create

news, as when people become “Facebook Famous.” For example, when a

Facebook user killed a homeless man in Cleveland and then posted a video

of the murder on Facebook, it was difficult to determine where the genesis

of “publicity” begins.

237

As Hoffman described, “the problem with using

Google News to determine public figure is that sometimes . . . the source of

the Google News result would be Facebook.”

238

This feedback loop

undermines one potential virtue of delegating the public-figure

determination to an outside algorithm: the notion that doing so would

provide some legitimacy because it would defer to neutral sources external

to the platform. There was never an official solution during Hoffman’s tenure

on how to respond to the circular moment of a person becoming a public

figure because of their actions on the platform.

239

235

. See, e.g., Snyder v. Phelps, 562 U.S. 443, 453 (2011) (“Deciding whether speech is of public

or private concern requires us to examine the content, form, and context of that speech, as revealed by

the whole record.” (internal quotation marks omitted) (citation omitted)).

236

. See, e.g., Amy Gajda, Judging Journalism: The Turn Toward Privacy and Judicial Regulation

of the Press, 97 CALIF. L. REV. 1039, 1041–42 (2009) (explaining that some courts have become less

deferential to the media in determining newsworthiness, perhaps on account of “growing anxiety about

the loss of personal privacy in contemporary society” or “declining public respect for journalism”);

Sydney Ember, Gawker and Hulk Hogan Reach $31 Million Settlement, N.Y. TIMES (Nov. 2, 2016),

https://www.nytimes.com/2016/11/03/business/media/gawker-hulk-hogan-settlement.html [https://perm

a.cc/TG5X-2W62] (describing the groundbreaking jury verdict that awarded former professional wrestler

Hulk Hogan $140 million after the gossip news site Gawker.com published a sex tape featuring him). But

cf. Carroll, supra note 95, at 77–81 (discussing cases in which the courts have “largely left the role [of

determining what is newsworthy or of legitimate public interest] to the press”).

237

. Jonah Engel Bromwich, Cleveland Police Seek Suspect After a Killing Seen on Facebook, N.Y.

TIMES (Apr. 16, 2017), https://www.nytimes.com/2017/04/16/us/facebook-live-shooting.html [https://

perma.cc/Z7JC-FW3D].

238

. Telephone Interview with Jud Hoffman, supra note 132.

239

. Id. Professor Enrique Armijo has argued that First Amendment law would provide a “useful

heuristic for content moderation” in this area because a “defamation defendant cannot cause a plaintiff to

become a public figure by dint of the statements that gave rise to the claim.” Enrique Armijo, Meet the

New Governors, Same as the Old Governors, KNIGHT FIRST AMEND. INST., (Oct. 30, 2018),

https://knightcolumbia.org/content/meet-new-governors-same-old-governors [https://perma.cc/5QPY-

E45T]; see also Wells v. Liddy, 186 F.3d 505, 511, 541 (4th Cir. 1999) (explaining that the relevant

controversy “must have existed prior to the publication of the defamatory statement”). Armijo argues that

Facebook could channel “[g]ood old-fashioned First Amendment law” to set a new rule: “if the results of

2019] FACEBOOK V. SULLIVAN 77

The second—and related—problem is that Google News is a news

aggregator, not a newsroom or newspaper. Courts have justified harsher

standards for public figures because speech about them tends to be

newsworthy.

240

Facebook’s approach partially adopts this proxy rationale

because a person’s appearance in the news makes them a public figure.

241

But whereas judicial deference to the press has historically rested upon trust

in the press as an institution, it is difficult to ascribe the same wisdom to

Google’s algorithm. Facebook’s tactic of defining “newsworthy people”

using not the press but a news algorithm raises concerns because someone

can easily become “newsworthy” without “news judgment.” In the new

speech ecosystem brought about in part by platforms like Facebook, people

can be thrust into the public sphere and bypass the gatekeeping function of

the traditional press.

242

In short, there are no editorial desks at Google

News.

243

All of this creates a kind of inverse Goldilocks principle whereby

Facebook ends up removing too much benign speech and preserving too

much harmful speech. As to the former, recall that Facebook’s cyberbullying

policies traditionally meant that any user could have offensive speech about

them removed so long as they were not a public figure, and conversely all

speech about public figures would stay up.

244

A purely descriptive approach

to identifying public figures means that important or influential people slip

through the cracks and remain “private” figures. News aggregators may

struggle to capture localized power or notoriety in smaller communities—an

issue made more problematic by the very nature of social media, which has

enabled virtual communities to develop their own distinctive cultures and

social structures.

245

Provocative content flagged in these communities may

not seem to involve any “public figures” when judged against a global

[the Google News] search include only stories about the complained-of bullying itself, then the victim of

the bullying is a private person” and could still rely on the platform’s protections against cyberbullying.

Armijo, supra. Facebook might complain that this task will be hard for content moderators to handle at

scale, but Armijo is correct that it might provide a fairer system than one that defers entirely to Google’s

algorithm.

240

. See, e.g., Rand v. Hearst Corp., 298 N.Y.S.2d 405, 411 (App. Div. 1969) (explaining that

privacy invasions may be justified if the plaintiff “has achieved the position of a ‘public figure’ and thus

became newsworthy” (citation omitted)).

241

. Telephone Interview with Dave Willner, supra note 132.

242

. See infra Sections III.B.2−.3.

243

. See Amy Gajda, Newsworthiness and the Search for Norms, KNIGHT FIRST AMEND. INST. (Oct.

30, 2018), https://knightcolumbia.org/content/newsworthiness-and-search-norms [https://perma.cc/Y7

US-2NBR].

244

. See supra Section II.A.

245

. See BOYD, supra note 8, at 7–14.

78 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

Google News search and therefore may be removed even if it involves a

matter of intense interest within that community.

Facebook’s algorithmic approach also prioritizes scalability over

accuracy. Even by their own admission, the platform’s policymakers knew

that it could not ascertain whether someone was being cyberbullied by

looking only at the flagged content.

246

The use of Google News was thus a

haphazard way to ensure that prominent people could not use a powerful and

blunt tool to remove critical speech about them, while still allowing private

figures to remove potentially verboten content with a simple complaint.

247

According to the many different Facebook employees with whom we have

spoken over the years, the vast majority of content that gets flagged for

moderators is not speech that actually violates the platform’s rules but rather

speech that certain users simply do not like.

248

Making important distinctions

based on an unnuanced mechanism like Google News will inevitably lead to

false positives.

At the other end of the spectrum, Facebook’s public-figure

determinations can preserve too much harmful speech. As courts know all

too well,

249

deferring to the public’s curiosity can sometimes lead to unjust

results—and Facebook’s approach effectively implements such blind

deference. It does so in two ways, all of which can be understood as being

part of a larger problem with algorithmic authority: unlike in the tort system,

there is no “normative backstop” to prevent people facing unjustly harsh

rules. First, as we discussed already, courts handling privacy claims have

increasingly refrained from allowing the media to define newsworthiness by

insisting that a disclosed fact is a matter of “legitimate” public concern.

250

The idea of legitimacy in this context allows courts and juries to consider

community mores in defining newsworthiness as a way to prevent “morbid

and sensational prying into private lives for its own sake.”

251

Facebook’s

approach to defining public figures contains no such limitation; indeed, it is

tough to imagine how a news algorithm could ever engage in such a complex

and context-dependent inquiry. As a result, a Facebook user might be denied

protection against cyberbullying even if he or she appeared in a news story

246

. Telephone Interview with Dave Willner, supra note 132.

247

. Telephone Interview with Jud Hoffman, supra note 132.

248

. Telephone Interview with Dave and Charlotte Willner, supra note 126; Telephone Interview

with Jud Hoffman, supra note 126.

249

. See RESTATEMENT (SECOND) OF TORTS § 652D cmt. h (AM. LAW INST. 1977).

250

. See id.

251

. Id.

2019] FACEBOOK V. SULLIVAN 79

(and thus became a “public figure” on the platform) that violated notions of

“common decency.”

252

This might strike many as an unjust result.

Facebook’s approach lacks a second normative backstop that courts

have built into the doctrine: the voluntariness requirement from defamation

law. As we have seen, courts use the voluntariness rationale to limit harsher

rules to plaintiffs who are “less deserving” of protection against defamation

under the theory that they have assumed the risk of injury by voluntarily

entering the public eye.

253

A Google News search cannot provide any such

limitation. This failing is all the more concerning because involuntary public

figures—once considered “exceedingly rare” by the Court

254

—are

increasingly common in the online realm.

255

Countless stories exist of

relatively unknown individuals being filmed or photographed and then

finding themselves subject to widespread online shaming and related news

coverage.

256

Should such an individual report any particularly offensive

posts to Facebook for violating the company’s cyberbullying rules, the

Google News search results would indicate that the individual is a public

figure and—at least until recently—the posts would stay up. Google News

is unequipped to distinguish between situations in which people have

voluntarily “thrust themselves to the forefront of particular public

controversies”

257

and situations in which someone has been catapulted into

internet stardom through little or no action of their own.

258

Concern about

inequitable treatment for targets of harmful speech is precisely what led the

Court in cases like Firestone to limit the class of individuals who would face

the harsher defamation rules reserved for public figures, but Facebook has

yet to respond to similar fears on its platform.

259

252

. See id.

253

. Wolston v. Reader’s Digest Ass’n, 443 U.S. 157, 164 (1979) (quoting Gertz v. Robert Welch

Inc., 418 U.S. 323, 345 (1974)).

254

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 345 (1974).

255

. See infra Section III.B.2.

256

. See Klonick, supra note 8, at 1044–50.

257

. Gertz, 418 U.S. at 345.

258

. Not only does Facebook’s approach fail to test for voluntariness, it also features no way to

enforce the “public controversy” limitation that plays an important role in defamation law. See Silvester

v. Am. Broad. Cos., 839 F.2d 1491, 1494 (11th Cir. 1988) (explaining that “a public controversy must be

more than merely newsworthy”).

259

. Perhaps aware of the inequities built into its approach, Facebook has attempted one trick to

limit the scope of its public-figure designation: it limited the timeframe in which someone had to appear

in the news to qualify as a public figure. This approach has echoes of the courts’ treatment of “limited-

purpose” public figures, who gain the status because of their voluntary involvement in a particular public

controversy, but it is fundamentally different from the normative considerations that animate the doctrine.

Indeed, Facebook’s temporal approach creates censorship risks because important speech could be

80 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

2. Voluntariness and Sympathy in the Age of Virality

To be charitable to Facebook, the platform’s inattention to the

constitutional concept of voluntariness might reflect a skepticism about its

normative desirability in the digital age. The Court has insisted that those

who voluntarily enter the public arena are “less deserving” of protection

because they assume the risk of negative attention,

260

but this premise has

been undermined by certain features of the new speech ecosystem brought

about by platforms like Facebook. For this reason, both courts and platforms

should reform—and maybe even replace—their use of voluntariness as a

bellwether to decide which people deserve harsher rules in disputes about

harmful speech as a matter of constitutional law or content moderation.

For purposes of defamation claims, Gertz’s taxonomy splits the world

into four types of people: general public figures, who are in “positions of

such persuasive power and influence that they are deemed public figures for

all purposes”

261

; limited-purpose public figures, who “thrust themselves to

the forefront of particular public controversies in order to influence the

resolution of the issues involved;”

262

involuntary public figures, who become

famous “through no purposeful action of [their] own”

263

; and private figures,

who are “more vulnerable to injury” and thus “more deserving of

recovery.”

264

While the first two groups “invite attention and comment”

through their own purposeful actions, the last two do not—and, as a result,

the Court has not applied the harsher constitutional rules to involuntary

public figures or private figures in defamation actions.

265

The same cannot be said for privacy law, which differs from defamation

law in two important respects. First, privacy law appears to define public

figures more broadly. The Restatement squashes the general- and limited-

purpose public figure into one category—the “voluntary public figure”—and

defines a public figure as anyone who “voluntarily places [oneself] in the

public eye, by engaging in public activities, or by assuming a prominent role

removed simply because the subject of the speech has not recently been in the news. Cf. RESTATEMENT

(SECOND) OF TORTS § 652D cmt. k (AM. LAW INST. 1977) (“The fact that there has been a lapse of time,

even of considerable length, since the event that has made the plaintiff a public figure, does not of itself

defeat the authority to give him publicity or to renew publicity when it has formerly been given.”).

260

. Wolston v. Reader’s Digest Ass’n, 443 U.S. 157, 164 (1979) (quoting Gertz v. Robert Welch,

Inc. 418 U.S. 323, 345 (1974)).

261

. Gertz, 418 U.S. at 345.

262

. Id.

263

. Id.

264

. Id.

265

. See id.

2019] FACEBOOK V. SULLIVAN 81

in institutions or activities having general economic, cultural, social or

similar public interest, or by submitting [oneself] or [one’s] work for public

judgment.”

266

This lacks the limitation that the Court has crafted in

defamation law that limited-purpose public figures must embroil themselves

in a “public controversy,” which the Court tellingly refused to equate with

“all controversies of interest to the public” because doing so would reinstate

the Rosenbloom rule that disavowed the significance of a plaintiff’s social

status.

267

In other words, the “public controversy” requirement serves to limit

the class of public figures in a way that has no analog in privacy law, where

you can become a public figure by involving yourself in activities that have

“public interest.”

268

The second difference between defamation and privacy law is that

voluntary and involuntary public figures are treated identically in privacy

law, whereas the Court has yet to clarify the constitutional consequence of

being an involuntary public figure in defamation law. It seems, however, that

the logic of the Court’s post-Gertz jurisprudence could not support equal

treatment for involuntary public figures who bring defamation claims.

Decisions like Firestone, Wolston, and Proxmire all place great weight on

the idea of voluntariness,

269

which the Court has called the “normative

consideration” that justifies the harsher rules that public figures face under

defamation law.

270

In Wolston, for example, the Court refused to apply the

public-figure label to the grand-jury witness who “was dragged unwillingly

into the controversy.”

271

This means that Facebook’s rules for public figures

hew closer to privacy law because the platform also draws no distinction

based on voluntariness.

272

To understand why the normative salience of voluntariness might have

shifted with the rise of platforms like Facebook, it is helpful to run through

a few real-world examples to test intuitions. With each example, consider

whether applying a harsher rule (in the courts or on platforms) seems fair

based on a voluntary-involuntary distinction, and indeed whether that

distinction has any influence on popular or legal intuitions. This allows

266

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. e (AM. LAW INST. 1977).

267

. Time, Inc. v. Firestone, 424 U.S. 448, 454 (1976); see also Silvester v. Am. Broad. Cos., 839

F.2d 1491, 1494 (11th Cir. 1988).

268

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. e.

269

. See Hutchinson v. Proxmire, 443 U.S. 111, 135 (1979); Wolston v. Reader’s Digest Ass’n, 443

U.S. 157, 164 (1979); Firestone, 424 U.S. at 454.

270

. Wolston, 443 U.S. at 164.

271

. Id. at 166.

272

. Telephone Interview with Dave Willner, supra note 132.

82 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

courts and platforms to possibly reform the idea of voluntariness—and

indeed decide whether they should replace it with something else entirely.

In November 2014, a Twitter user posted a photo of a Target employee

bagging items behind a cashier.

273

The employee was wearing the nametag

“Alex.”

274

In one day, the tweet gained over one thousand retweets and two

thousand favorites.

275

In two days, the “#AlexFromTarget” hashtag had over

a million Twitter hits and the phrase “Alex From Target” racked up more

than 200,000 Google searches.

276

Before long, Twitter users managed to

identify “Alex” and unearth his Twitter account, which quickly amassed over

250,000 followers.

277

Alex appeared on the Ellen talk show two days later.

Death threats, denigrating posts, and “fabricate[d] stories” about him soon

followed.

278

It is hard to argue that Alex from Target, a “global celebrity”

with hundreds of thousands of social-media followers,

279

is merely a private

figure. Similarly, it is hard to argue that he is a voluntary public figure who

thrust himself into the vortex of a public controversy by bagging groceries

at his part-time job.

280

At most, then, he is the elusive involuntary public

figure that the Court believed was “exceedingly rare,”

281

though his fleeting

internet stardom seems different from the public’s interest in the two

involuntary public figures mentioned in the Restatement—perpetrators and

victims of crimes.

282

Unlike Alex from Target, some people play a more active role in

triggering the public’s attention by posting content online that then goes

viral. Consider the example of Justine Sacco, who tweeted before boarding

a flight to Cape Town: “Going to Africa. Hope I don’t get AIDS. Just

kidding. I’m white!”

283

The tweet went viral, and people all around the world

began following the “#HasJustineLandedYet” hashtag to track her progress

273

. Alex from Target / #AlexFromTarget, KNOW YOUR MEME, https://knowyourmeme.com/

memes/alex-from-target-alexfromtarget [https://perma.cc/ASB5-GMAK].

274

. Id.

275

. Id.

276

. Id.

277

. Id.

278

. Nick Bilton, Alex from Target: The Other Side of Fame, N.Y. TIMES (Nov. 12, 2014),

https://www.nytimes.com/2014/11/13/style/alex-from-target-the-other-side-of-fame.html [https://perma.

cc/J62U-QNXF].

279

. Id.

280

. See Hutchinson v. Proxmire, 443 U.S. 111, 135 (1979).

281

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 345 (1974).

282

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. f (AM. LAW INST. 1977).

283

. Klonick, supra note 8, at 1048–49 (citation omitted).

2019] FACEBOOK V. SULLIVAN 83

as they discussed what she had said.

284

When she landed in Cape Town

eleven hours later, she discovered that she was “the No. 1 worldwide trend

on Twitter” and had tens of thousands of responses to her tweet.

285

The story

was picked up by several major media outlets,

286

and Sacco soon lost her

job.

287

As with Alex from Target, it seems strange to think of Sacco as a

private figure given her sudden worldwide fame. But whether to dub her a

“voluntary” or “involuntary” public figure is far from clear. It seems like a

stretch to say that, under defamation law, she voluntarily “thrust [herself] to

the forefront of particular public controversies in order to influence the

resolution of the issues involved,”

288

but she might qualify as a voluntary

public figure under privacy law’s more permissive standard.

289

Any

designation is further complicated by the fact that the public furor on social

media surrounding Sacco’s tweet arguably created a newsworthy event in its

own right, even if the underlying events did not.

Sacco, at least, was an adult who presumably had some sense of the

possible ramifications of her actions, even if she could not have predicted

their scale or intensity. Matters become more complicated, however, when

children become internet sensations. This can be so when they lack the

maturity to grasp the consequences of their actions, as one might think

occurred recently with students from Covington Catholic whose interaction

with a Native American activist became a huge story after videos of the

incident went viral on social media.

290

As one journalist quipped in the

aftermath, “The thing to remember about the dumbass teens of Covington

Catholic is that while they are dumbasses, they are also teens.”

291

Even

though the students voluntarily took part in a public march in the middle of

the nation’s capital while they knew people were filming events on their

phones, to say that they assumed the risk of such colossal negative

attention—in the way courts speak of public figures in defamation law—

284

. Jon Ronson, How One Stupid Tweet Blew Up Justine Sacco’s Life, N.Y. TIMES MAG. (Feb. 12,

2015), https://www.nytimes.com/2015/02/15/magazine/how-one-stupid-tweet-ruined-justine-saccos-

life.html [https://perma.cc/3CD3-MY7E].

285

. Id.

286

. Klonick, supra note 8, 1048–49.

287

. Ronson, supra note 284.

288

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 345 (1974).

289

. RESTATEMENT (SECOND) OF TORTS § 652D cmt. e (AM. LAW INST. 1977).

290

. Mike Pesca, Covington Boys: The Difference Between Jerks and Monsters, SLATE (Jan. 24,

2019, 7:16 PM), https://slate.com/news-and-politics/2019/01/covington-catholic-the-scandal-that-isnt-a-

scandal.html [https://perma.cc/RSM2-UR6P].

291

. Id.

84 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

seems farfetched.

292

Perhaps more troubling are those children who are “thrust” into the

limelight by their parents. Six-year-old Adalia Rose Williams, for example,

gathered nearly six million Facebook fans after her mother set up a page

about her rare and fatal condition that made Williams age much faster than

normal.

293

The Facebook page was a source of widespread support and

empathy, but it also drew vile abuse and even a hoax story that Williams had

died.

294

Williams’s mother admitted that she was trying to raise awareness

about her daughter’s condition by placing her in the public eye.

295

Williams

was, in a sense, “thrust” into a public forum created intentionally to provoke

widespread engagement and discussion, but Williams herself did nothing to

encourage her fame. Nonetheless, whether she voluntarily or involuntarily

garnered public attention, it is hard to think of someone tracked by millions

of people as a private figure.

Finally, there are people who have chosen to be in the spotlight but who

might nonetheless seem deserving of protection against certain types of

online reactions. Take even the most obvious public figures, like actress and

comedienne Leslie Jones, who understandably spend much of their time

trying to gain public attention. Jones was inundated with racist and sexist

comments on Twitter after she starred in the all-female Ghostbusters

remake.

296

In the wake of the abuse, Jones tweeted “I feel like I’m in a

personal hell. I didn’t do anything to deserve this.”

297

There is no doubt that

Jones’s fame makes her a public figure under defamation and privacy law.

292

. See Wolston v. Reader’s Digest Ass’n, 443 U.S. 157, 164 (1979). It is true, of course, that

Covington students might not qualify as defamation public figures in the courts, despite their online fame.

See Eugene Volokh, Libel Law and the Covington Boys, VOLOKH CONSPIRACY (Jan. 24, 2019, 1:01 PM),

https://reason.com/volokh/2019/01/24/libel-law-and-the-covington-boys [https://perma.cc/TH9S-

ZGWJ] (arguing that the students are still private figures because “they weren’t famous or influential

before this event” and that “just showing up at a rally would [not] qualify” as “voluntarily entering some

particular debate” to make them limited-purpose public figures).

293

. Simon Tomlinson, Six-Year-Old Girl with Body of an Old Woman… Who Sings Vanilla Ice,

Dances Gangnam Style and Has Own Fan Club: Adalia Rose Suffers Rare Premature Aging Condition,

DAILY MAIL (Feb. 25, 2013, 7:06 AM), https://www.dailymail.co.uk/news/article-2284110/Adalia-Rose-

The-year-old-girl-body-old-woman-progeria.html [https://perma.cc/C3FH-ZCGY].

294

. Id.

295

. Jonathan Weiss, Girl with Rare Genetic Disorder Bullied Online, MED. DAILY (Feb. 26, 2013,

3:24 PM), https://www.medicaldaily.com/girl-rare-genetic-disorder-bullied-online-244525 [https://per

ma.cc/D9CS-UNZL].

296

. Anna Silman, A Timeline of Leslie Jones’s Horrific Online Abuse, THE CUT (Aug. 24, 2016),

https://www.thecut.com/2016/08/a-timeline-of-leslie-joness-horrific-online-abuse.html [https://perma.cc

/2ACG-6STU].

297

. Id.

2019] FACEBOOK V. SULLIVAN 85

But whether—as a normative matter—she deserves the harsher

cyberbullying rules that accompany public-figure status on Facebook is a far

harder question.

What can these anecdotes teach courts and platforms as they handle

disputes in the digital age? The first lesson is that the voluntary-involuntary

distinction needs reform if it is to retain normative appeal as a barometer for

how “deserving” a person is of harsher rules. A person’s viral internet fame

might make it difficult to conceive of her as a private figure, but it might also

strike us as unfair if that person’s decision to post on social media suffices

to strip that person of protection under tort law or content-moderation rules.

This intuition might stem from at least three points about life in the digital

age. First, even if a person intentionally posts online, it is not safe to presume

that he or she was intending to provoke an online firestorm and suddenly

become famous. Second, we might question whether the person truly

assumed the risk of a viral reaction and all of the baggage that can

accompany it.

298

And third, it is no longer so “rare” to become involuntarily

famous with little or no “purposeful action of [his or her] own.”

299

These

three points bring to mind a person walking through a forest when the heel

of her shoe sets off a spark that creates a massive forest fire. He or she may

have voluntarily taken actions in the world that led to a catastrophic incident,

but it feels wrong to say that he or she assumed that risk or deserves to be

punished for arson.

300

This might mean that, in defamation law, courts should narrowly define

public figures to prevent the exception from swallowing the rule. If, as one

court has remarked, a public figure is “anyone who is famous or infamous

because of who he is or what he has done,”

301

there will be far too many

public figures in this world. Courts should vigilantly apply the limitations

placed on public-figure status in defamation law, insisting that people have

voluntarily embroiled themselves in a “particular public controversy” and

298

. Cf. Gertz v. Robert Welch, Inc., 418 U.S. 323, 345 (1974) (explaining that public figures

“invite attention and comment” and thus have “voluntarily exposed themselves to increased risk of injury

from defamatory falsehood concerning them”).

299

. See supra notes 257−81 and accompanying text.

300

. To any graduate of an American law school, such an example might call to mind the classic

torts case of Palsgraf v. Long Island Railroad Co., 162 N.E. 99, 99 (N.Y. 1928), in which the plaintiff,

Helen Palsgraf, was waiting to board a train when another passenger stumbled while running to catch the

train, dropping a package of fireworks, which subsequently exploded and caused a large scale on the

platform to hit her. The case is typically taught to introduce the idea of foreseeability and causation as

potential limiting factors on tort liability. Foreseeability might provide insights for why we might not

want to increase the burden on the plaintiff in certain instances.

301

. Cepeda v. Cowles Magazines & Broad., Inc., 392 F.2d 417, 419 (9th Cir. 1968).

86 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

not simply been swept up in events that suddenly become “of interest to the

public.”

302

At the very least, the Supreme Court should bring clarity to the

constitutional significance of being an involuntary public figure. The Court’s

post-Gertz jurisprudence suggests that such people do not deserve harsher

defamation rules, but the time has come for certainty now that they are no

longer “exceedingly rare.”

To reform privacy law, courts would need to both narrow the definition

of a voluntary public figure and distinguish between voluntary and

involuntary public figures. Under current privacy doctrine, a vast range of

people could be voluntary public figures: posting a Tweet, photo, blog post,

or status update would all seem to qualify as having “submitted [oneself] or

[one’s] work for public judgment” so as to make every human who has ever

had a social-media account a voluntary public figure.

303

Even those who

refrain from social media but get mentioned on it by “engaging in public

activities” might qualify.

304

If not a voluntary public figure, such hapless

individuals might be involuntary public figures, who are subject to identical

treatment under privacy law. This lack of distinction is particularly

troublesome because, as we have seen, platforms themselves have created an

environment in which virality is a feature of their products and involuntary

public figures are everywhere.

305

Given the challenges posed by the digital age, perhaps the better

approach is to replace voluntariness altogether as a normative measurement.

Now that people voluntarily engage in all sorts of activities that unexpectedly

spawn virality and fame, one might question whether voluntariness is doing

the work that it once did to assess who is “less deserving” of protection

against harmful speech. Put differently, has the ease with which everyone

can now publish and amplify their speech changed the social (and perhaps

legal) significance of voluntary engagement in public debate? Moreover,

does the newfound prevalence of involuntary public figures undermine

distinctions based on voluntariness now that it is no longer “exceedingly

rare” to be thrust into the limelight against your will?

Platforms like Facebook have created an environment that reveals

problems with penalizing people for “thrusting” themselves into a “public

302

. Time, Inc. v. Firestone, 424.U.S. 448, 453–54 (1976).

303

. See RESTATEMENT (SECOND) OF TORTS § 652D cmt. e (AM. LAW INST. 1977).

304

. See id.

305

. Cf. James Grimmelmann, Saving Facebook, 94 IOWA L. REV. 1137, 1137 (2009) (observing

that “Facebook offers a socially compelling platform that also facilitates peer-to-peer privacy violations”

in part because “people use Facebook with the goal of sharing information about themselves”).

2019] FACEBOOK V. SULLIVAN 87

controversy.” To apply harsher rules to people who speak out risks chilling

valuable expression. This is particularly concerning when targets of harmful

speech might surrender certain protections as the price of publicly

responding to online abuse—a dynamic that risks “blaming the victim,”

heightening their vulnerability, and worsening the harm they might suffer.

306

These concerns might explain why platforms like Facebook eschew

considerations of voluntariness and instead reach for other normative

concepts to judge what is “fair” in their systems of private governance. To

borrow the language of one Facebook policymaker, what might matter is

whether a person is “sympathetic.” The sympathetic public figure is often

involuntary, but not always. As Willner put it, “you can think of them as

involuntary public figures, but another way of saying it might be to think of

them as sympathetic public figures”

307

because they most frequently came

up with people who were caught up in terrible circumstances or had been

publicly shamed and that shaming had gone viral.

It is unclear whether Facebook ever formalized this concept in any

actual rules, but it is nonetheless useful to consider how it might have

influenced the platform’s line-drawing in this context. Rather than using an

empirical fact as a basis for a normative conclusion—such as determining

whether somebody voluntarily invited public attention as a way to conclude

that they are “less deserving” of protection—the description of someone as

“sympathetic” skips straight to an opaque normative analysis about whether

applying the harsh rule is fair under the circumstances. This makes it a tough

standard to administer, particularly on a mass scale where such subjectivity

can breed inconsistency. But the same can be said for legal standards that

rest on community mores—such as the inquiry into whether a matter is of

“legitimate” public concern in determining its newsworthiness. Perhaps,

then, the “sympathetic” public figure is not so different from the person who,

under privacy or IIED law, must face harsher rules for the sake of ensuring

robust public discourse.

We might begin to give some shape to the concept of the sympathetic

306

. For discussion of this general problem online and in other areas of law, see DANIELLE KEATS

CITRON, HATE CRIMES IN CYBERSPACE 77–78 (2014); Heidi M. Hurd, Blaming the Victim: A Response

to the Proposal That Criminal Law Recognize a General Defense of Contributory Responsibility, 8 BUFF.

CRIM. L. REV. 503, 510 (2005); Josephine Ross, Blaming the Victim: ‘Consent’ Within the Fourth

Amendment and Rape Law, 26 HARV. J. ON RACIAL & ETHNIC JUST. 1, 3 (2010); JoAnne Sweeny,

Gendered Violence and Victim-Blaming: The Law’s Troubling Response to Cyber-Harassment and

Revenge Pornography, 8 INT’L J. TECHNOETHICS 18, 23 (2017).

307

. Telephone Interview with Dave Willner, supra note 132.

88 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

public figure by analyzing how it might fare better than the notion of

voluntariness in the digital age. As we have seen, voluntariness can still

influence the normative calculus in certain circumstances: Alex from Target

might eventually become a public figure following his online stardom, but it

feels odd to treat him identically to someone like Donald Trump given that

Alex did nothing (aside from being handsome while bagging groceries!) to

encourage his fame.

308

But we might also worry about how fair it is for courts

and platforms to apply harsher rules to the likes of Justine Sacco, the

Covington Catholic students, and Adalia Rose even though all of them, in

some sense, became famous through voluntary actions taken by themselves

or their legal guardians.

309

What ties these three anecdotes together is that it

seems unlikely that the people involved foresaw that their actions would spur

such a significant reaction and transform them into public figures, nor would

a reasonable person have foreseen it either. It therefore seems unfair to say—

as we might in defamation law—that they are “less deserving” of protection

because they assumed the risk of becoming targets of harmful speech due to

their own voluntary actions.

310

If Facebook and other platforms are serious about replacing the concept

of voluntariness with something like “sympathy,” they cannot rely on

algorithms and automation to do the job. These judgments are too nuanced

and depend on understanding context and complicated social facts, so

mechanical application of the standard will fail. Google News, then, must be

replaced or supplemented by other tools that, at least at some point in the

process, incorporate human review. Without it, platforms will run headlong

into the same problems discussed in the previous Section, whereby

algorithmic authority removes any normative backstop and implements a

purely descriptive approach. Still, this case-by-case approach is not without

its drawbacks, as we will now see.

3. The Perils of Ad Hoc Exceptions

Merely including humans in the mix is not necessarily a solution to the

problems posed by algorithmic authority and the new speech ecosystem that

platforms have helped create. Facebook’s experiences implementing its ad

hoc newsworthiness exception are proof of the perils of case-by-case

adjudication in this area. Given courts’ parallel experiences, this should

come as no surprise. Indeed, similar concerns animated the Supreme Court’s

decision to overrule Rosenbloom in Gertz, in which Justice Lewis Powell

308

. Bilton, supra note 278.

309

. See supra notes 267−79 and accompanying text.

310

. Wolston v. Reader’s Digest Ass’n, 443 U.S. 157, 164 (1979).

2019] FACEBOOK V. SULLIVAN 89

warned of the dangers of “forcing state and federal judges to decide on an ad

hoc basis which publications address issues of ‘general or public interest’

and which do not—to determine . . . ‘what information is relevant to self-

government.’”

311

If the Court “doubt[ed] the wisdom of committing this task

to the conscience of judges” in the court system,

312

we might also worry

about Facebook executives acting similarly behind closed doors.

As things stand, Facebook and other platforms seem to deviate from

their rules if a high-ranking policymaker at the company intervenes and

determines that a piece of content is newsworthy. This overarching exception

applies to all types of content on Facebook. In the context of Napalm Girl,

for instance, Sheryl Sandberg intervened and Facebook’s rules against

nudity ultimately gave way to a determination within the platform that it was

“an iconic image of historical importance.”

313

And when then-candidate

Donald Trump’s posts about his Muslim ban appeared to run afoul of

Facebook’s rules against hate speech, they stayed up after Mark Zuckerberg

and senior members of Facebook’s policy team concluded that they were

“newsworthy, significant, or important to the public interest.”

314

No doubt

some newsworthiness judgments fall to policymakers who are lower on the

Facebook totem pole, but the important point is that these decisions are made

by humans—not algorithms—and often stem from company higher-ups.

Facebook’s newsworthiness exception in some ways mirrors the courts’

treatment of the same concept in privacy and IIED law. The interests served

by both torts give way to a constitutional commitment to protecting robust

discourse on matters of public concern, a determination that is often

influenced by the judgment of the press.

315

But newsworthiness

determinations at Facebook are made by an anonymous and somewhat

arbitrary group of policy executives.

316

This infects the adjudication process

with various problems and exposes a more systemic issue: the dearth of

accountability that a private company like Facebook has to its users, despite

impacting its users’ speech rights on the platform.

311

. Gertz v. Robert Welch, Inc., 418 U.S. 323, 346 (1974) (quoting Rosenbloom v. Metromedia,

Inc., 403 U.S 29, 79 (1971) (Marshall, J., dissenting)).

312

. Id.

313

. Zillman, supra note 182.

314

. Deepa Seetharaman, Facebook Employees Pushed to Remove Trump’s Posts as Hate Speech,

WALL ST. J. (Oct. 21, 2016, 7:43 PM), http://www.wsj.com/articles/facebook-employees-pushed-to-

remove-trump-posts-as-hate-speech-1477075392 [https://perma.cc/63C2-LY4U].

315

. See, e.g., Gajda, supra note 236, at 1041.

316

. Adler, supra note 129.

90 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

The first problem created by ad hoc exceptions at platforms is that they

might suffer from a lack of expertise or professional norms. At least in the

courts, newsworthiness determinations are often shaped by deference to the

traditional news media, where society has come to expect some level of skill,

experience, and professionalism.

317

Even when tort claims do not involve a

media defendant, judges themselves are usually proficient at reading and

interpreting past decisions and deciding how they should apply in analogous

contexts. It is at best unclear if we can expect the same from Facebook

policymakers. Some may have expertise in content moderation, but others

are there to represent the platform’s business interests. Because users do not

really know who is making these decisions, they are unable to judge what

their skewed incentives might be.

318

But one thing is for sure: the decisions

are currently made by Facebook insiders, and not by any neutral overseer

less likely to prioritize the company’s profits or prestige.

What little we do know about Facebook’s decisionmaking process in

this area is concerning. At least anecdotally, it seems that powerful and

connected people are more likely to get favorable treatment. This

materializes in two ways. For one, when prominent people complain about

content takedowns, the platforms seem more likely to listen. When a Danish

politician posted a photo of Copenhagen’s iconic Little Mermaid statue on

Facebook, the platform removed it for violating its nudity rules because the

image featured “too much bare skin or sexual undertones.”

319

After the

politician complained publicly, Facebook backtracked and reinstated the

post.

320

Similarly, after Facebook removed a post by a group of journalists

in the Philippines criticizing President Rodrigo Duterte, the group was able

to mobilize public pressure and persuade the platform to reinstate it.

321

Facebook may have provided everyone with a platform, but that does not

317

. Even courts might have to become more skeptical of news deference in the digital age, in which

everyone has a platform and professional norms of journalism are arguably deteriorating. See Gajda,

supra note 236, at 1041–42; cf. Sarah C. Haan, The Post-Truth First Amendment, 94 IND. L.J.

(forthcoming 2019).

318

. See Sarah C. Haan, Profits v. Principles, KNIGHT FIRST AMEND. INST. (Oct. 30, 2018),

https://knightcolumbia.org/content/profits-v-principles [https://perma.cc/6J89-U5ND] (observing that

“[t]ensions exist between Facebook’s business interests and its aspiration to create a prosocial expressive

environment”).

319

. Denmark: Facebook Blocks Little Mermaid Over ‘Bare Skin,’ BBC (Jan. 4, 2016),

https://www.bbc.com/news/blogs-news-from-elsewhere-35221329 [https://perma.cc/A69H-5XUH].

320

. Id.

321

. Aries Joseph Hegina, Facebook Restores Journalists’ Page with Anti-Duterte Post,

INQUIRER.NET (June 7, 2016, 11:15 AM), https://technology.inquirer.net/48270/facebook-restores-

journalists-page-with-anti-duterte-post [https://perma.cc/92S4-FCCT].

2019] FACEBOOK V. SULLIVAN 91

mean that everyone is heard equally.

322

The second way that platforms entrench power with the powerful is by

allowing influential people to speak in ways that the general public cannot.

Again, at least anecdotally, Facebook seems more likely to find the speech

of public figures to be newsworthy, meaning that powerful users will often

get a pass even when their content violates the platform’s rules. Zuckerberg

conceded as much when he justified the decision to treat Trump differently

because his status as a “public figure” made the posts about the Muslim ban

“newsworthy.”

323

In the aftermath of that decision, Facebook announced that

it would “begin allowing more items that people find newsworthy,

significant, or important to the public interest—even if they might otherwise

violate [its] standards.”

324

This preferential treatment is worsened by what

Jillian York has called the “tornadoes of celebrity,” whereby different

platforms reinforce each other in a way that creates new celebrities and then

helps them remain powerful.

325

Even though Facebook has now openly said that it makes exceptions

for newsworthy content, it has given little insight into how it defines

newsworthiness. Senior Facebook executive Peter Stern has said that the

platform considers “safety of individuals on the one hand and voice on the

other,” but “voice” is a nebulous concept to say the least.

326

If policymakers

are making these calls on a case-by-case basis, without the benefit of a body

of reasoned decisions from past cases, there is a risk that mere human whim

can lead to erratic and arbitrary results. Of course, a nuanced and ambiguous

standard like newsworthiness is always susceptible to inconsistent

application, but Facebook has given us little confidence that it can reliably

and fairly make these calls.

One cause for skepticism is the apparent American-centric conclusions

that Facebook has drawn in this area. As we saw, the platform was quick to

abandon its anti-gore rules to allow images of a Boston Marathon victim to

322

. See Press Release, Elec. Frontier Found., EFF, Human Rights Watch, and Over 70 Civil

Society Groups Ask Mark Zuckerberg to Provide All Users with Mechanism to Appeal Content

Censorship on Facebook (Nov. 13, 2018), https://www.eff.org/press/releases/eff-human-rights-watch-

and-over-70-civil-society-groups-ask-mark-zuckerberg-provide [https://perma.cc/TR6L-TERT].

323

. Seetharaman, supra note 314.

324

. Kaplan & Osofsky, supra note 183.

325

. Jillian York, Director of Int’l Freedom of Expression, Elec. Frontier Found., Speech at the

re:publica 2018 Conference: The New Kingmakers: How Silicon Valley Created a New Culture of

Celebrity (May 6, 2018), https://www.youtube.com/watch?v=Rl9JX5jmY0M [https://perma.cc/FGP6-

CDJ7].

326

. Telephone Interview with Peter Stern, supra note 184.

92 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

remain online.

327

But when users made similar requests concerning videos

depicting brutal violence committed by Mexican drug cartels, the result was

quite different.

328

Lower-level Facebook employees deemed the Mexican

videos newsworthy, but a high-ranking executive overruled them following

intense backlash in the media.

329

The platform’s rule seemed to shift

depending on what side of the border the event occurred. Likewise, while

Trump’s Islamophobic posts about banning all Muslim immigration were

saved due to their newsworthiness, Facebook removed posts by the son of

Israeli prime minister Benjamin Netanyahu wishing that “all Muslims leave

the land of Israel.”

330

This form of American exceptionalism is, in some sense, unsurprising.

Facebook is an American company run largely by Americans, so a normative

concept like newsworthiness has predictably been influenced by American

norms. Yet what makes this trend at Facebook so interesting is that it seems

contrary to how the leaders at the platform conceive of their mission. When

Zuckerberg first raised the idea of creating a “Supreme Court” for content

moderation, he extolled its virtue as a body that could “make the final

judgment call on what should be acceptable speech in a community that

reflects the social norms and values of people all around the world.”

331

This

is an ambitious—and perhaps impossible—goal: the nature of norms and

values is that they often develop differently within dissimilar communities.

Facebook may wish that it could create one set of Community Standards that

would satisfy everyone within its polity, but the reality is that these global

platforms simply cannot craft policies that represent worldwide “norms and

values” that do not exist—particularly surrounding such complex and

contestable notions as the boundaries of free speech. Facebook will need to

choose the values it wants to reflect in its content moderation, but the

questions of how it will choose them and what those values might be remains

unanswered.

In sum, several lessons emerge from a comparative analysis of courts’

and platforms’ treatment of public figures and newsworthiness. For starters,

327

. Adler, supra note 129.

328

. Id.

329

. Id.

330

. Facebook Temporarily Bans Israeli PM’s Son Over Posts, BBC (Dec. 17, 2018),

https://www.bbc.com/news/world-middle-east-46591270 [https://perma.cc/5FUE-KG2V]; cf. Jake

Evans, Fraser Anning’s Public Facebook Page Removed for Reported Hate Speech, ABC (Sept. 28, 2018,

12:53 AM), https://www.abc.net.au/news/2018-09-28/fraser-annings-facebook-page-taken-down/1031

7638 [https://perma.cc/MU6M-MUHJ] (discussing Facebook’s banning of Australian politician from the

platform after he violated rules prohibiting hate speech).

331

. Klein, supra note 205 (emphasis added).

2019] FACEBOOK V. SULLIVAN 93

Facebook’s use of Google News for its public-figure determinations

underscores the dangers of reducing such judgments to mechanical

calculations: when followed strictly, they can result in either taking down or

keeping up too much speech. This approach, which necessarily jettisons the

“voluntariness” caveat recognized by the courts, can lead to inaccurate and

unjust results. Although Facebook and other platforms have proliferated the

once-rare “involuntary public figure” imagined in Gertz, they—and the

courts—have yet to reckon with this new reality. These issues reveal that the

new speech ecosystem created by platforms like Facebook has indeed eroded

the empirical postulate at the core of the public-figure doctrine—that is, if

you are a public figure, you have greater access to channels of

communication—now that everyone has a platform. Yet despite this

apparent equalization of speech access, the result is often less than

egalitarian. The powerful and connected often get better rules and treatment,

not all speech receives equal attention or amplification, and the mechanisms

used by Facebook to regulate speech—though increasingly transparent—

remain largely opaque and unaccountable to users. Facebook’s struggle to

create principled exceptions for newsworthy content underscores how the

company straddles the roles of the legislature, executive, judiciary, and press

in controlling access to speech for both speakers and listeners.

CONCLUSION

As this Article reveals, platforms like Facebook have fundamentally

altered the nature of the global speech ecosystem.

332

They have given

everyone a platform to speak, dramatically reducing the barriers to entry in

public debate. They have facilitated the amplification of speech, allowing

users to reach broad audiences in places near and far. They have created both

opportunities and dangers by enabling virality, accelerating and extending

332

. This Article has focused largely on Facebook for the purposes of comparing the private and

public modes of adjudicating claims about harmful speech. But although Facebook is currently the most

powerful and ubiquitous speech platform, it is of course just one of several players in this space. Content

moderation is a key part of many internet platforms, from online marketplaces (for example, Etsy and

Amazon) to gaming streaming (for example, Twitch and Steam), and from smaller niche communities

(for example, Ravelry and Patreon) to global speech platforms (for example, Twitter and YouTube). The

multitude of features offered by Facebook gives it a unique role in this new speech ecosystem and allows

us to analogize to the various roles played by all of these actors, making its use as a case study particularly

valuable. The lessons we can learn from Facebook’s experience will be helpful to addressing parallel

concerns raised by other online platforms. See, e.g., Twitter Will Hide Rule-Breaking Politicians’ Tweets,

BBC (June 27, 2019), https://www.bbc.com/news/technology-48791094 [https://perma.cc/NPJ5-6UY2]

(describing Twitter’s new policy to hide politicians’ “newsworthy” tweets that break the platform’s rules

but are left online “in the public interest”).

94 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

the influence of speech by previously obscure people. And they have

developed a new system of governance, adjudicating the boundaries between

protecting free speech and preventing harmful speech.

In the age dominated by the Old Governors, speech governance was

essentially split between the legislature, executive, judiciary, and press.

333

At least in the United States, this governance system was divided between

three branches of government, created and backed by values enshrined in the

Constitution. Congress and state legislatures would make laws, those laws

would be enforced by executive branches, and the legality of those laws

would be measured against the Constitution by courts.

334

As a check against

all of this official governance was the so-called “Fourth Estate” of the press,

which enjoyed its own constitutional protection in its role as watchdog of the

government and caretaker of an informed electorate.

335

The press’s decisions

of what to publish—what was “newsworthy”—were made by editorial

boards and given some deference in courts.

336

Collectively, as Balkin has

argued, this was the system of “old-school speech regulation.”

337

Today, in the age of the New Governors, we can see shadows of these

various roles, but in a quite different construct. Much of the governance of

online speech is done by private platforms that fulfill all of these roles—

legislature, executive, judiciary, and press—at once. At Facebook, policy

teams make rules on content, moderators enforce those rules in response to

flagged content, and escalations teams review whether those enforcements

were correct or if larger changes should be made in policy.

338

And while the

media still plays an important role in raising awareness about problematic

content-moderation rules or decisions,

339

platforms are also publishers of

speech and act as editorial boards determining what types of content see the

light of day. Platforms are both the governors, setting speech policies and

adjudicating speech disputes, and the publishers, controlling access to speech

on behalf of speakers and listeners. They are the Sullivan Court, and they are

333

. See Balkin, supra note 8, at 2301, 2306−08.

334

. See Marbury v. Madison, 5 U.S. (1 Cranch) 137 (1803).

335

. As Edmund Burke once said, “there were Three Estates in Parliament; but, in the Reporters

Gallery yonder, there sat a Fourth Estate more important far than they all.” THOMAS CARLYLE, ON

HEROES, HERO-WORSHIP, AND THE HEROIC IN HISTORY 202 (John Chester Adams ed., Houghton Mifflin

& Co. 1907) (1842); see also Rachel Luberda, The Fourth Branch of the Government: Evaluating the

Media’s Role in Overseeing the Independent Judiciary, 22 NOTRE DAME J.L. ETHICS & PUB. POL’Y 507,

507−10 (2008).

336

. Gajda, supra note 236, at 1041–42.

337

. Balkin, supra note 8, at 2298.

338

. Klonick, supra note 5, at 1630–58.

339

. See, e.g., Hegina, supra note 321.

2019] FACEBOOK V. SULLIVAN 95

the New York Times.

Facebook seems to realize these parallels as well. Though it is just one

of many speech platforms, it is leading the way in attempting to build

accountability and oversight into its speech policies. By creating an

independent “Oversight Board” to make policy and appeals determinations

concerning the content users may post, Zuckerberg spoke directly to

concerns raised by the consolidation of power at the platform.

340

Indeed, in

his blog post making the announcement, he stated that creating the tribunal

was an attempt to “prevent the concentration of too much decision-making

within [its] teams” and to “provide assurance that these decisions are made

in the best interests of [the] community and not for commercial reasons.”

341

Concerns about Facebook’s incentives have led some to worry that the body

will be “more soundbite than substance,”

342

but there are promising signs

that Zuckerberg will put his money where his mouth is—both literally and

figuratively. In January 2019, the platform released a few more specifics

about how the body might operate. In order to “render independent

judgment” on the platform’s speech policies, the body may “[r]everse

Facebook’s decisions when necessary.”

343

According to the announcement,

members of the board “will be obligated to the people who use Facebook—

not Facebook the company.”

344

Facebook even published a “Draft Charter”

340

. Zuckerberg, supra note 123. The idea of such a tribunal did not spring from nowhere. Its

genesis built on years of pressure from outside groups and Facebook’s own attempts to create some form

of appellate review for content-moderation decisions. In the 2015 Manila Principles, civil-society groups

demanded that “[l]aws and content restriction policies and practices must respect due process” and

provide “[t]ransparency and accountability.” ELEC. FRONTIER FOUND., MANILA PRINCIPLES ON

INTERMEDIARY LIABILITY 4−5 (2015), https://www.eff.org/files/2015/10/31/manila_principles_1.0.pdf

[https://perma.cc/HNB9-F9EY]. David Kaye, the United Nations Special Rapporteur on Freedom of

Expression, echoed these recommendations in his report to the Human Rights Council on free speech and

the private sector in the digital age. David Kaye, Rep. of the Special Rapporteur on the Promotion and

Protection of the Right to Freedom of Opinion and Expression, ¶ 89, U.N. Doc. A/HRC/32/38 (May 11,

2016) (“It is also critical that private entities ensure the greatest possible transparency in their policies,

standards and actions that implicate the freedom of expression and other fundamental rights.”). In a

similar vein, the 2018 Santa Clara Principles declared that platforms “should provide a meaningful

opportunity for timely appeal of any content removal or account suspension” and outlined several

minimum standards that platforms should meet. THE SANTA CLARA PRINCIPLES ON TRANSPARENCY AND

ACCOUNTABILITY IN CONTENT MODERATION (2018), https://santaclaraprinciples.org [https://perma.cc/

K8ZN-7SNF].

341

. Zuckerberg, supra note 123.

342

. douek, supra note 204.

343

. Nick Clegg, Charting a Course for an Oversight Board for Content Decisions, FACEBOOK

NEWSROOM (Jan. 28, 2019), https://newsroom.fb.com/news/2019/01/oversight-board [https://perma.

cc/5ZA5-QG8L].

344

. Id.

96 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

for the board, which included commitments that the board would “share its

decisions transparently and give reasons for them” and that “Facebook will

accept and implement the board’s decisions.”

345

Most significantly, the

Charter also revealed that the board will base its decisions not only on

Facebook’s pre-existing Community Standards, but also on “a set of values,

which will include concepts like voice, safety, equity, dignity, equality and

privacy.”

346

These values, which sound akin to constitutional provisions, will

be included in a final charter “that will serve as the basis for board

governance.”

347

Between January and June 2019, Facebook sought external advice on

what the Oversight Board should look like.

348

Over those six months, it

“heard from more than 650 people from 88 countries represented at 22

smaller global roundtables,” gathered “feedback from more than 250 experts

in one-on-one meetings,” and established “an online ‘public consultation’

process, which encouraged users to both answer polls and submit essays on

what they thought the board should look like.”

349

As one might expect from

such a global listening tour, the lengthy report released in late June

summarizing the platform’s findings raises more questions than answers.

350

But it does demonstrate Facebook’s commitment to seeking out the opinions

and thoughts of users and stakeholders worldwide and providing

transparency in reporting those findings, which will hopefully influence the

Oversight Board’s final charter.

351

Facebook is in the midst of its own kind of Constitutional Convention

that could fundamentally alter its nature and the way it governs online

345

. FACEBOOK, DRAFT CHARTER: AN OVERSIGHT BOARD FOR CONTENT DECISIONS 1 (2019),

https://fbnewsroomus.files.wordpress.com/2019/01/draft-charter-oversight-board-for-content-decisions-

1.pdf [https://perma.cc/EH7R-YZ7P].

346

. Id. at 3.

347

. Id. at 5.

348

. Evelyn Douek & Kate Klonick, Facebook Releases an Update on Its Oversight Board: Many

Questions, Few Answers, LAWFARE (June 27, 2019, 3:41 PM), https://www.lawfareblog.com/facebook-

releases-update-its-oversight-board-many-questions-few-answers [https://perma.cc/7V2D-HFTV].

349

. Kate Klonick & Evelyn Douek, Facebook’s Federalist Papers, SLATE (June 27, 2019, 9:44

AM), https://slate.com/technology/2019/06/facebook-oversight-board-community-standards-federalist-

papers.html [https://perma.cc/BBL9-5N34].

350

. Douek & Klonick, supra note 348; see also Brent Harris, Global Feedback and Input on the

Facebook Oversight Board for Content Decisions, FACEBOOK NEWSROOM (June 27, 2019),

https://newsroom.fb.com/news/2019/06/global-feedback-on-oversight-board [https://perma.cc/C5PX-

6866].

351

. For an insightful analysis of the potential benefits and limitations of Facebook’s Oversight

Board, see generally Evelyn Douek, Facebook’s “Oversight Board:” Move Fast with Stable

Infrastructure and Humility, 21 N.C. J.L. & TECH. 1 (2019).

2019] FACEBOOK V. SULLIVAN 97

speech.

352

Zuckerberg seems to have finally come to terms with his

tremendous power, acknowledging that decisions about how to “balance

safety and free expression . . . are too consequential for Facebook to make

alone.”

353

With the advent of an oversight body that aspires to bring

“independent judgment to hard cases,”

354

the platform is on the cusp of

creating a meaningful check on its own power that could have ripple effects

throughout the industry and reshape public discourse on the internet. This

may seem hyperbolic, but Facebook’s 2.3 billion users leave a giant footprint

on the character of online speech—and the nature of the platform’s

governance over them is among the most pressing issues concerning freedom

of expression in the digital age.

Facebook’s influence over online speech makes critical oversight all the

more important. As the comparative analysis in this Article has shown,

Facebook’s approach to issues surrounding public figures and

newsworthiness raises a host of problems. The platform’s use of Google

News to determine public-figure status is purely descriptive and lacks a

normative backstop to consider concepts like voluntariness or community

mores.

355

Yet if Facebook deviates from its algorithmic tools to make

exceptions based on human judgment, it risks creating arbitrary and

inconsistent results through an opaque process that is largely hidden from its

users.

356

Facebook’s exceptions for newsworthy content raise similar

concerns.

357

Although the platform strives vaguely to balance the “voice” of

its users against the “safety” of it users,

358

when it comes to difficult issues

surrounding “sympathetic” public figures and highly contextual

newsworthiness determinations, the lack of transparent and granular

articulations of the platform’s decisions can lead to a host of problems.

359

This Article has revealed the inner workings of Facebook’s content

moderation surrounding the crucial concepts of public figures and

newsworthiness. The rules and processes that the platform has adopted have

352

. See generally David Pozen, Authoritarian Constitutionalism in Facebookland,

BALKINIZATION (Oct. 30, 2018), https://balkin.blogspot.com/2018/10/authoritarian-constitutionalism-

in.html [https://perma.cc/AV6K-UYSC] (observing that Facebook’s existing model is closer to

“authoritarian constitutionalism” than a “common law system,” in part because Facebook lacks “(i)

formally independent dispute resolution bodies, paradigmatically courts, that issue (ii) precedential, (iii)

written decisions”).

353

. Clegg, supra note 343.

354

. Id.

355

. See supra Sections III.B.1−.2.

356

. See supra Section III.B.3.

357

. See supra Section III.B.3.

358

. Telephone Interview with Peter Stern, supra note 184.

359

. See supra Section III.B.3.

98 SOUTHERN CALIFORNIA LAW REVIEW [Vol. 93:37

deep roots in First Amendment law, but they differ in critical respects. By

comparing the old and new systems of speech governance, this Article has

exposed flaws in both. But these flaws are not fatal—both judges and

platform policymakers can change their doctrines to adapt to challenges

posed by the new speech ecosystem brought about by companies like

Facebook. The battle over how to protect free speech while regulating

harmful speech must now be fought on two fronts: through tort law in courts

and content moderation on platforms. While the Old Governors have long-

established structures to adjudicate the public’s claims, the New Governors

are still building theirs.

2019] FACEBOOK V. SULLIVAN 99