LESSONS FROM TRUMP'S SUSPENSION: HOW TWITTER SHOULD CLARIFY AND STRENGTHEN ITS "PUBLIC INTEREST" APPROACH TO MODERATING LEADERS' VIOLENCE-INSPIRING SPEECH.

AuthorHolmberg, Erika Suh

TABLE OF CONTENTS I. INTRODUCTION 310 II. THE CURRENT LANDSCAPE 313 A. A Brief History and Overview of the "Public Interest Framework" 313 B. Trump and the Suspension 315 III. PROPOSED CHANGES 317 A. Continue and Clarify Treatment of Direct Threats as "Exceptions to the Exception" 318 B. Revise "Public Interest Framework" Analysis to Explicitly Consider Off-Platform Context 322 1. Memorialize and Exercise the Ability to Consider Off- Platform Circumstances and Leader Statements in Initial "Public Interest" Review of Framework- Eligible Leaders' Reported Tweets 325 2. Invest in Analysis of Interpretations of Leaders' Statements on and off Twitter, and Invest in Additional Staff Knowledgeable of Local Contexts Responsible for Monitoring Framework-Eligible Leaders 326 3. Articulate and Apply a Context-Aware Standard for Suspending Framework-Eligible Leaders for Repeated or Egregious Violations of Violence- Related Twitter Rules 328 C. The Case of T. Raja Singh: How More Context-Aware Moderation Would Operate 329 IV. CONCLUSION 332 I. INTRODUCTION

Twitter's decision to permanently suspend @realDonaldTrump after the January 6, 2021, insurrection marked a significant departure from both its stated policies and its recent practices. In 2019, the company had announced a "public interest framework," formalizing a practice of content moderation leniency toward major world leaders' Tweets. (1) Throughout Trump's presidency, Twitter had held world leaders to a more lenient standard than other users--declining to remove their Tweets or suspend their accounts, even if they violated platformwide rules, so long as the actual or potential harmfulness of their continued presence on Twitter did not outweigh the presumed inherent value of preserving public access to these leaders' online speech. (2) Trump's suspension marked the first time Twitter banned a head of state since adopting this framework. (3)

Twitter's decision to permanently suspend Trump and its stated justifications for doing so (4) raise important questions regarding how Twitter will moderate the actions of political leaders in the future. For example, why hadn't Twitter banned Trump earlier? Will Twitter revise its existing moderation policies and practices to enforce a similar context-aware approach against other global leaders who might use their accounts to incite violence? (5) Will Twitter apply a similar form of analysis as the one described in its statement on Trump's suspension if other leaders do in fact use Twitter to incite violence? (6) Until and unless Twitter meaningfully answers these open questions, world leaders and their followers will be left guessing as to which leaders Twitter might de-platform next, and the same failures that resulted from Twitter's current content moderation policies may enable another world leader to use Twitter to inspire political violence.

With these issues in mind, this Note argues that Trump's suspension and eventual ban from Twitter illustrate key defects and ambiguities in the policies and practices Twitter uses to moderate political leaders' "dangerous speech" (7) under the public interest framework. This Note proposes mechanisms through which Twitter can ameliorate these weaknesses without undervaluing the public interest benefits and free speech principles that the framework aims to protect. Given the particularly significant impact of world leaders' online speech on real-world safety, this Note offers reforms to the framework's approach to the specific harm of potential violence incitement as a first step in Twitter's broader process of rethinking its lenient approach to moderating world leaders. Though Twitter functions as a public square in some respects, as a private platform, under the current U.S. legal and regulatory landscape, it is free to create and enforce its own rules, suspend users, and remove or leave up Tweets at will. So long as this status quo continues, incorporating the free speech principles and trade-offs into the public interest framework in the manner this Note proposes can better ensure that Twitter will wield its tremendous speech-policing power more responsibly and transparently. (8)

Part II provides a brief overview of Twitter's publicized moderation policies on content posted by world leaders that potentially incites violence, the evolution of these policies during Trump's presidency, and tensions between the formal policy itself and Twitter's stated reasons for banning Trump.

Part III, Section A argues that whenever framework-eligible leaders Tweet "clear and direct threats of violence against an individual," (9) Twitter must hold them to the same enforcement standard as the general public without consideration of the possible public interest value of the Tweet(s) due to the particularly high risk of inciting grievous harm and the relative weakness of public interest justifications in this context. In order to resolve existing ambiguity in Twitter's definition of prohibited "clear and direct threats of violence against an individual" that will "result in enforcement action" regardless of the author's status, (10) Twitter should clarify that the only type of "context" that may inform these decisions is the context of on-platform surrounding circumstances. Twitter must also clarify whether similar threats against groups are also subject to immediate removal.

Part III, Section B addresses Tweets by framework-eligible leaders that may carry a substantial risk of causing violence even when they do not constitute "clear and direct threats against an individual," as illustrated by Trump's Tweets that emboldened his supporters to violently storm the Capitol but fell short of explicitly threatening or encouraging those specific violent acts. In these myriad "harder cases," accurately assessing the risk of violence requires a more context-aware approach that considers how users interpret and react to both on- and off-platform content. Part III, Section B proposes three ways in which Twitter can and should change its policies and practices on moderating ambiguous but potentially violence-inspiring Tweets by framework-eligible leaders to more effectively gauge the risk of violence to be balanced against the speech's public interest value.

Part III, Section C explores an example of Tweets posted by a particular politician to illustrate how these proposed changes would function in practice. Finally, this Note offers concluding thoughts on solutions to mitigate concerns that this proposed approach grants undue discretion and power to Twitter in policing online speech or heightens the risk of selective enforcement.

  1. THE CURRENT LANDSCAPE

    1. A Brief History and Overview of the "Public Interest Framework"

      Prior to 2019, Twitter responded to criticism of its non-interference with controversial Tweets by pointing to the inherent "newsworthiness" and public interest value of world leaders' speech, emphasizing the "critical role" political leaders play in public conversation. (11) In June 2019, Twitter created a new formalized "notice" system, under which world leaders' Tweets that violate the Twitter Rules remain on the platform, but are shielded behind an interstitial user notice explaining that the shielded Tweet violates a rule, (12) but "it may be in the public's interest for the Tweet to remain available." (13) This system exclusively applies to Tweets from accounts that (1) represent a "government/elected official," a candidate for public office, or a government appointee, (2) have over 100,000 followers, and (3) are verified. (14)

      In October 2019, Twitter expanded the notice system by announcing a broader "public interest framework" in an official policy statement (the "World Leaders Policy Statement"). (15) The platform also added a provision to its general guidelines on content moderation that codified the new "public interest exception" system. (16) To qualify for framework analysis, an account must meet the same criteria defined under the notice system. (17) Under the framework, Twitter's Enforcement Team evaluates reported Tweets from framework-eligible leaders against Twitter's rules, focusing on the plain language of the Tweet without "attempt[ing] to determine all potential interpretations of the content or its intent." (18) If the Enforcement Team determines the Tweet violates a rule, the Trust & Safety Team then shares a recommendation on "whether or not continued access to the Tweet is in the public interest" with a "cross-functional set of leaders across different internal teams with diverse and multidisciplinary backgrounds... as well as inmarket teams with an understanding of the cultural context in which the Tweet was posted." (19) After receiving the stakeholders' feedback, Trust & Safety Team senior leaders decide whether to leave the Tweet up behind a notice--thereby implementing the "public interest exception" to standard enforcement of the Twitter Rules--or remove the Tweet, balancing "potential risk and severity of harm" against public interest value. (20)

      According to its general guidelines provision on the public interest exception, Twitter is "less likely to make [public interest] exceptions" when a Tweet "threatens or glorifies violence," (21) and will "especially err on the side of removal... where there is evidence the content may be leading to actual or likely offline harm," though in "rare instances," the exception may apply "if there is a more attenuated connection to actual violence." (22) Twitter is also "more likely" to remove a Tweet after public interest review if it includes a "declarative call to action that could harm a specific individual or group." (23)

      The World Leaders Policy Statement further clarifies that Tweets containing "clear and direct threats of violence against an individual" will result in "enforcement action... without consideration of the potential public interest value in allowing the Tweet to remain visible behind a notice."...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT