Microsoft Word - Turley.Testimony.CISA. FINAL.docx

Written Statement

Jonathan Turley

Shapiro Professor of Public Interest Law

The George Washington University Law School

“Censorship Laundering:

How the U.S. Department of Homeland Security Enables the Silencing of Dissent”

Subcommittee on Oversight, Investigation, and Accountability

Committee on Homeland Security

United States House of Representatives

May 11, 2023

I. INTRODUCTION

Chairman Bishop, ranking member Ivey, members of the Subcommittee, my

name is Jonathan Turley, and I am a law professor at George Washington University,

where I hold the J.B. and Maurice C. Shapiro Chair of Public Interest Law.

It is an honor

to appear before you today to discuss free speech and government censorship.

For the purposes of background, I come to this subject as someone who has

written,

litigated,

and testified

in the areas of congressional oversight and the First

I appear today on my own behalf, and my views do not reflect those of my law school or the

media organizations that feature my legal analysis.

In addition to a blog with a focus on First Amendment issues (www.jonathanturley.org), I have

written on First Amendment issues as an academic for decades. See, e.g., Jonathan Turley, THE

INDISPENSABLE RIGHT: FREE SPEECH IN THE AGE OF RAGE (forthcoming 2024); Jonathan

Turley, The Unfinished Masterpiece: Speech Compulsion and the Evolving Jurisprudence of Religious

Speech 82 MD L. REV. (forthcoming 2023); Jonathan Turley, Rage Rhetoric and the Revival of American

Sedition, 65 William & Mary Law Review (forthcoming 2023), Jonathan Turley, The Right to Rage in

American Political Discourse, GEO. J.L. & PUB. POL’Y (forthcoming 2023); Jonathan Turley, Harm and

Hegemony: The Decline of Free Speech in the United States, 45 HARV. J.L. & PUB. POL’Y 571 (2022);

Jonathan Turley, The Loadstone Rock: The Role of Harm in The Criminalization of Plural Unions, 64

EMORY L.J. 1905 (2015); Jonathan Turley, Registering Publicus: The Supreme Court and Right to

Anonymity, 2002 SUP. CT. REV. 57-83.

See, e.g., Eugene Volokh, The Sisters Wives Case and the Criminal Prosecution of Polygamy,

WASH. POST, Aug. 28, 2015 (discussing challenge on religious, speech, and associational rights); Jonathan

Turley, Thanks to the Sisters Wives Litigation, We have One Less Morality Law, WASH. POST, Dec. 12,

2013.

See, e.g., “Hearing on the Weaponization of the Federal Government,” United States House of

Representatives, House Judiciary Committee, Select Subcommittee on the Weaponization of the Federal

Government, February 9, 2023 (statement of Jonathan Turley); Examining the ‘Metastasizing’ Domestic

Terrorism Threat After the Buffalo Attack: Hearing Before the S. Comm. on the Judiciary, 117th Cong.

(2022) (statement of Jonathan Turley); Secrecy Orders and Prosecuting Leaks: Potential Legislative

Responses to Deter Prosecutorial Abuse of Power: Hearing Before H. Comm. on the Judiciary, 117th

Cong. (2021) (statement of Jonathan Turley); Fanning the Flames: Disinformation and Extremism in the

Media: Hearing Before the Subcomm. on Commc’n & Tech. of the H. Comm. on Energy & Com., 117th

Cong. (2021) (statement of Jonathan Turley); The Right of The People Peacefully to Assemble: Protecting

Amendment for decades. I have also represented the United States House of

Representatives in litigation.

My testimony today obviously reflects that past work and I

hope to offer a fair understanding of the governing constitutional provisions, case law,

and standards that bear on this question.

As I recently testified before the House Judiciary Committee, the growing

evidence of censorship and blacklisting efforts by the government raises serious and

troubling questions over our protection of free speech.

There are legitimate

disagreements on how Congress should address the role of the government in such

censorship. The first step, however, is to fully understand the role played in prior years

and to address the deep-seated doubts of many Americans concerning the actions of the

government to stifle or sanction speech.

The Twitter Files and other recent disclosures raise serious questions of whether

the United States government is now a partner in what may be the largest censorship

system in our history. That involvement cuts across the Executive Branch, with

confirmed coordination with agencies ranging from the Homeland Security to the State

Department to the Federal Bureau of Investigation (FBI). Even based on our limited

knowledge, the size of this censorship system is breathtaking, and we only know of a

fraction of its operations through the Twitter Files, congressional hearings, and pending

litigation. Most of the information has come from the Twitter Files after the purchase of

the company by Elon Musk. Notably, Twitter has 450 million active users

but it is still

only ranked 15th in the number of users, after companies such as Facebook, Instagram,

TikTok, Snapchat, and Pinterest.

The assumption is that the government censorship

program dovetailed with these other companies, which continue to refuse to share past

communications or work with the government. Assuming these efforts extended to the

larger platforms, we have a government-supported censorship system that is unparalleled

in history.

We now have undeniable evidence of a comprehensive system of censorship that

stretches across the government, academia, and corporate realms. Through disinformation

offices, grants, and other means, an array of federal agencies has been active

“stakeholders” in this system. This includes Homeland Security, State Department, the

FBI and other federal agencies actively seeking the censorship of citizens and groups.

The partners in this effort extend across social media platforms. The goal is not just to

remove dissenting views, but also to isolate those citizens who voice them. We recently

Speech By Stopping Anarchist Violence: Hearing Before the Subcomm. on the Const. of the S. Comm. on

the Judiciary, 116th Cong. (2020) (statement of Jonathan Turley); Respect for Law Enforcement and the

Rule of Law: Hearing Before the Commission on Law Enforcement and the Administration of Justice,

(2020) (statement of Jonathan Turley); The

Media and The Publication of Classified Information: Hearing

Before the H. Select Comm. on Intelligence, 109

Cong. (2006) (statement of Jonathan Turley).

See U.S. House of Representatives v. Burwell, 185 F. Supp. 3d 165 (D.D.C. 2016),

https://casetext.com/case/us-house-of-representatives-v-capacity-1.

Some of today’s testimony is material include from that earlier hearing. “Hearing on the

Weaponization of the Federal Government,” United States House of Representatives, House Judiciary

Committee, Select Subcommittee on the Weaponization of the Federal Government, February 9, 2023

(statement of Jonathan Turley).

Twitter Revenue and User Statistics, BUSINESS OF APPS, Jan. 31, 2023,

https://www.businessofapps.com/data/twitter-statistics/.

Most Popular Social Networks, STATISTA, https://www.statista.com/statistics/272014/global-

social-networks-ranked-by-number-of-users/.

learned that this effort extended even to companies like LinkedIn.

New emails

uncovered in the Missouri v. Biden litigation reportedly show that the Biden

Administration’s censorship efforts extended to Facebook to censor private

communications on its WhatsApp messaging service.

The effort to limit access, even to

professional sites like LinkedIn, creates a chilling effect on those who would challenge

majoritarian or official views. It was the same chilling effect experienced by scientists

who tried to voice alternative views on vaccines, school closures, masks, or the Covid

origins. The success of this partnership may surpass anything achieved by direct state-run

systems in countries like Russia or China.

The recent disclosures involving the Cybersecurity and Infrastructure Security

Agency (CISA) is chillingly familiar. It is part of an ever-expanding complex of

government programs and grants directed toward the censorship or blacklisting of

citizens and groups. In just a matter of weeks, the size of this complex has come into

greater focus and has confirmed the fears held by many of us over the use of private

actors to do indirectly what the government is prohibited from doing directly. I have

called it “censorship by surrogate” and CISA appears to be the latest agency to enlist

private proxy actors.

The focus of this hearing is particularly welcomed, as it reminds us that the cost

of censorship is not just the loss of the right to free expression. Those costs can include

the impact of reducing needed public debate and scrutiny in areas like public health. For

years, government and corporate figures worked to silence scientists and researchers who

opposed government policies on mask efficacy, universal vaccinations, school closures,

and even the origin of Covid-19. Leading experts Drs. Jayanta Bhattacharya (Stanford

University) and Martin Kulldorff (Harvard University) as well as a host of others, faced

overwhelming attacks for questioning policies or views that later proved questionable or

downright wrong. Those doctors were the co-authors of the Great Barrington

Declaration, which advocated for a more focused Covid response that targeted the most

vulnerable populations, rather than widespread lockdowns and mandates.

Dr. Kulldorff was censored in March 2021 when he tweeted “Thinking that

everyone must be vaccinated is as scientifically flawed as thinking that nobody should.

COVID vaccines are important for older high-risk people and their care-takers. Those

with prior natural infection do not need it. Nor children.” Every aspect of that tweet was

worthy of scientific and public debate. However, with the support of political, academic,

and media figures, such views were suppressed at the very moment in which they could

have made the most difference. For example, if we had a true and open debate, we might

have followed other countries in keeping schools open for young children. Agencies and

the media now recognize that these objections had merit.. We are now experiencing an

educational and mental health crisis associated with a lockdown that might have been

avoided or reduced (as in other countries). Millions died as government agencies enlisted

Jonathan Turley, “Connect to Opportunity”: State Department Pushed LinkedIn to Censor

“Disinformation,” Res Ipsa Blog (www.jonathanturley.org), Apr. 12, 2023,

https://jonathanturley.org/2023/04/12/connect-to-opportunity-new-evidence-shows-state-department-

pushing-linkenin-to-censor-disinformation/.

Jonathan Turley, New Documents Expose Government Censorship Efforts at Facebook and

WhatsApp, Res Ipsa Blog (www.jonathanturley.org), March 26, 2023,

https://jonathanturley.org/2023/03/26/new-documents-expose-government-censorship-efforts-at-facebook-

and-whatsapp/.

companies to silence dissenting viewpoints on best practices and approaches. We do not

know how many of those deaths or costs might have been avoided because this debate

was delayed until after the pandemic had largely subsided.

The purpose of my testimony today is to address the legal question of when

government support for censorship systems becomes a violation of the First Amendment

and, more broadly, when it convenes free speech principles. To that end, I hope to briefly

explore what we know, what we do not know, and why we must know much more about

the government’s efforts to combat speech deemed misinformation, disinformation, and

malinformation (MDM).

Regardless of how one comes out on the constitutional ramifications of the

government’s role in the censorship system, there should be no serious debate over the

dangers that government-supported censorship presents to our democracy. The United

States government may be outsourcing censorship, but the impact is still inimical to the

free speech values that define this country. This should not be a matter that divides our

political parties. Free speech is the core article of faith of all citizens in our constitutional

system. It should transcend politics and, despite our deepening divisions, unite us all in a

common cause to protect what Justice Louis Brandeis once called “the indispensable

right.”

II. MDM AND CENSORSHIP BY SURROGATE

It is a common refrain among many supporters of corporate censorship that the

barring, suspension, or shadow banning of individuals on social media is not a free

speech problem. The reason is that the First Amendment applies to the government, not

private parties. As a threshold matter, it is important to stress that free speech values are

neither synonymous with, nor contained exclusively within, the First Amendment. The

First Amendment addressed the most prevalent danger of the time in the form of direct

government regulation and censorship of free speech and the free press. Yet, free speech

in society is impacted by both public and private conduct. Indeed, the massive censorship

system employed by social media companies presents the greatest loss of free speech in

our history. These companies, not the government, now control access to the

“marketplace of ideas.” That is also a free speech threat that needs to be taken seriously

by Congress. While the Washington Post has shown that the Russian trolling operations

had virtually zero impact on our elections,

the corporate censorship of companies like

Twitter and Facebook clearly had an impact by suppressing certain stories and viewpoints

in our public discourse. It was the response to alleged disinformation, not the

disinformation itself, that manipulated the debate and issues for voters.

The First Amendment addresses actions by the government, but there are certainly

actions taken by these agencies to censor the views of citizens. While one can debate

whether social media executives became effective government agents, public employees

are government agents. Their actions must not seek to abridge the freedom of speech. It is

possible that a systemic government program supporting a privately-run censorship

Whitney v. California, 274 U.S. 357, 375-76 (1927) (Brandeis, J., concurring).

Tim Starks, Russian Trolls on Twitter Had Little Influence on 2016 Election, WASH. POST, Jan. 9,

2023, https://www.washingtonpost.com/politics/2023/01/09/russian-trolls-twitter-had-little-influence-2016-

voters/.

system is sufficient to justify injunctive relief based on the actions of dozens of federal

employees to target and seek the suspension of citizens due to their viewpoints. However,

this program can also run afoul of the First Amendment if the corporate counterparts in

the system are considered effective government agents themselves. The most common

example occurs under the Fourth Amendment where the government is sometimes

viewed as acting through private security guards or snitches performing tasks at its

request.

The same agency relationship can occur under the First Amendment, particularly

on social media. The “marketplace of ideas” is now largely digital. The question is

whether the private bodies engaging in censorship are truly acting independently of the

government. There is now ample reason to question that separation. Social media

companies operate under statutory conditions and agency review. That relationship can

allow or encourage private parties to act as willing or coerced agents in the denial of free

speech. Notably, in 1946, the Court dealt with a town run by a private corporation in

Marsh v. Alabama.

It was that corporation, rather than a government unit, that

prevented citizens from distributing religious literature on a sidewalk. However, the

Court still found that the First Amendment was violated because the corporation was

acting as a governing body. The Court held that, while the denial of free speech rights

“took place, [in a location] held by others than the public, [it] is not sufficient to justify

the State’s permitting a corporation to govern a community of citizens so as to restrict

their fundamental liberties.”

Congress has created a curious status for social media companies in granting

immunity protections in Section 230. That status and immunity have been repeatedly

threatened by members of Congress unless social media companies expanded censorship

programs in a variety of different areas. The demands for censorship have been

reinforced by letters threatening congressional action. Many of those threats have

centered around removing Section 230 immunity, pursuing antitrust measures, or other

vague regulatory responses. Many of these threats have focused on conservative sites or

speakers. The language of the Section itself is problematic in giving these companies

immunity “to restrict access to or availability of material that the provider or user

considers to be obscene, lewd, lascivious, filthy, excessively violent, harassing, or

otherwise objectionable, whether or not such material is constitutionally protected.”

Columbia Law professor Phil Hamburger has noted, the statute appears to permit what is

made impermissible under the First Amendment:

“Congress makes explicit that it is

immunizing companies from liability for speech restrictions that would be

unconstitutional if lawmakers themselves imposed them.”

As Hamburger notes, that

does not mean that the statute is unconstitutional, particularly given the judicial rule

favoring narrow constructions to avoid unconstitutional meanings.

However, there is

Marsh v. Alabama, 326 U.S. 501 (1946).

Id. at 509.

47 U.S.C. § 230(c).

Philip Hamburger, The Constitution Can Crack Section 230, WALL STREET JOURNAL (Jan. 30,

2021).

Congress makes explicit that it is immunizing companies from liability for speech restrictions that

would be unconstitutional if lawmakers themselves imposed them.

Id. See, e.g., Republican Party of Hawaii v. Mink, 474 U.S. 1301, 1302 (1985) (narrowly

interpreting the recall provisions of the Honolulu City Charter).

another lingering issue raised by the use of this power to carry out the clear preference on

“content moderation” of one party.

The Court has recognized that private actors can be treated as agents of the

government under a variety of theories. Courts have found such agency exists when the

government exercises “coercive power” or “provided such significant encouragement,

either overt or covert, that the choice must in law be deemed to be that of the

State.”

The Court has also held that the actions of a private party can be “fairly treated

as that of the State itself” where there exists a “close nexus between the State and the

challenged action” that a private action “may be fairly treated as that of the State itself.”

I will return to the case law below, but first it is useful to consider what is currently

known about the government-corporate coordination revealed by the Twitter Files.

I will not lay out the full array of communications revealed by Twitter and recent

litigation, but some are worth noting as illustrative of a systemic and close coordination

between the company and federal officials, including dozens reportedly working within

the FBI. The level of back-channel communications at one point became so

overwhelming that a Twitter executive complained that the FBI was “probing & pushing

everywhere.” Another official referred to managing the government censorship referrals

as a “monumental undertaking.” At the same time, dozens of ex-FBI employees were

hired, including former FBI General Counsel James Baker. There were so many FBI

employees that they set up a private Slack channel and a crib sheet to allow them to

translate FBI terms into Twitter terms more easily. The Twitter Files have led groups

from the right to the left of our political spectrum to raise alarms over a censorship

system maintained by a joint government-corporate effort.

Journalist Matt Taibbi was

enlisted by Elon Musk to present some of these files and reduced his findings to a simple

header: “Twitter, the FBI Subsidiary.”

As discussed today, these disclosures show that FBI is not alone among the

federal agencies in systemically targeting posters for censorship. Indeed, emails reveal

FBI figures, like San Francisco Assistant Special Agent in Charge Elvis Chan, asking

Twitter executives to “invite an OGA” (or “Other Government Organization”) to an

upcoming meeting. A week later, Stacia Cardille, a senior Twitter legal executive,

indicated the OGA was the CIA, an agency under strict limits regarding domestic

activities. Much of this work apparently was done through the multi-agency Foreign

Influence Task Force (FITF), which operated secretly to censor citizens. Cardille

referenced her “monthly (soon to be weekly) 90-minute meeting with FBI, DOJ, DHS,

ODNI [Office of the Director of National Intelligence], and industry peers on election

threats.” She detailed long lists of tasks sent to Twitter by government officials. The

censorship efforts reportedly included “regular meetings” with intelligence officials. This

included an effort to warn Twitter about a “hack-and-leak operation” by state actors

targeting the 2020 presidential election. That occurred just before the New York Post

Blum v. Yaretsky, 457 U.S. 991, 1004 (1982).

Jackson v. Metro. Edison Co., 419 U.S. 345, 351 (1974).

Compare Yes, You Should be Worried About the Relationship with Twitter, THE FIRE, Dec. 23,

2022, https://www.thefire.org/news/yes-you-should-be-worried-about-fbis-relationship-twitter with Branco

Marcetic, Why the Twitter Files Are In Fact a Big Deal, JACOBIN, Dec. 29, 2022,

https://jacobin.com/2022/12/twitter-files-censorship-content-moderation-intelligence-agencies-surveillance.

story on Hunter Biden’s laptop was published and then blocked by Twitter. It was also

blocked by other social media platforms like Facebook.

The files also show the staggering size of government searches and demands. The

FBI reportedly did key word searches to flag large numbers of postings for possible

referral to Twitter. On November 3, 2020, Cardille told Baker that “[t]he FBI has “some

folks in the Baltimore field office and at HQ that are just doing keyword searches for

violations. This is probably the 10th request I have dealt with in the last 5 days.” Baker

responded that it was “odd that they are searching for violations of our policies.” But it

was not odd at all. Twitter had integrated both current and former FBI officials into its

network and the FBI was using the company’s broadly defined terms of service to target

a wide array of postings and posters for suspensions and deletions.

At one point, the coordination became so tight that, in July 2020, Chan offered to

grant temporary top-secret clearance to Twitter executives to allow for easier

communications and incorporation into the government network.

This close working

relationship also allowed the government use of accounts covertly, reportedly with the

knowledge of Twitter. One 2017 email sent by an official from United States Central

Command (CENTCOM) requested that Twitter “whitelist” Arabic-language Twitter

accounts that the government was using to “amplify certain messages.” The government

also asked that these accounts be granted the “verified” blue checkmark.

The range of available evidence on government coordination with censorship

extends beyond the Twitter Files and involves other agencies. For example, recent

litigation brought by various states over social media censorship revealed a back-channel

exchange between defendant Carol Crawford, the CDC’s Chief of digital media and a

Twitter executive.

The timing of the request for the meeting was made on March 18,

2021. Twitter senior manager for public policy Todd O’Boyle asked Crawford to help

identify tweets to be censored and emphasized that the company was “looking forward to

setting up regular chats.” However, Crawford said that the timing that week was “tricky.”

Notably, that week, Dorsey and other CEOs were to appear at a House hearing to discuss

“misinformation” on social media and their “content moderation” policies. I had

just testified on private censorship in circumventing the First Amendment as a type of

censorship by surrogate.

Dorsey and the other CEOs were asked at the March 25, 2021,

hearing about my warning of a “little brother problem, a problem which private entities

do for the government which it cannot legally do for itself.”

Dorsey insisted that there

was no such censorship office or program.

Mark Zuckerberg has also stated that the FBI clearly warned about the Hunter Biden laptop as

Russian disinformation. David Molloy, Zuckerberg Tells Rogan that FBI Warning Prompted Biden Laptop

Story Censorship, BBC, August 26, 2022, https://www.bbc.com/news/world-us-canada-62688532.

Gadde and Roth both testified that they do not know if anyone took up this offer for clearances.

The lawsuit addresses how experts, including Drs. Jayanta Bhattacharya (Stanford University) and

Martin Kulldorff (Harvard University), have faced censorship on these platforms.

Fanning the Flames: Disinformation and Extremism in the Media: Hearing Before the Subcomm.

on Commc’n & Tech. of the H. Comm. on Energy & Com., 117th Cong. (2021) (statement of Jonathan

Turley, Shapiro Professor of Public Interest Law, The George Washington University Law School).

Misinformation and Disinformation on Online Platforms: Hearing Before the Subcomm. on

Commc’n & Tech. and Subcomm. on Consumer Protection of the H. Comm. on Energy & Com., 117th

Cong. (2021).

The pressure to censor Covid-related views was also coming from the White

House, as they targeted Alex Berenson, a former New York Times reporter, who had

contested agency positions on vaccines and underlying research. Rather than push

information to counter Berenson’s views, the White House wanted him banned. Berenson

was eventually suspended.

These files show not just a massive censorship system but a coordination and

integration of the government to a degree that few imagined before the release of the

Twitter Files. Congressional hearings have only deepened the alarm for many in the free

speech community. At one hearing, former Twitter executive Anika Collier Navaroli

testified on what she called the “nuanced” standard used by her and her staff on

censorship, including the elimination of “dog whistles” and “coded” messaging. She then

said that they balanced free speech against safety and explained that they sought a

different approach:

“Instead of asking just free speech versus safety to say free speech for whom and

public safety for whom. So whose free expression are we protecting at the expense

of whose safety and whose safety are we willing to allow to go the winds so that

people can speak freely?”

The statement was similar to the statement of the former CEO Parag Agrawal. After

taking over as CEO, Agrawal pledged to regulate content as “reflective of things that we

believe lead to a healthier public conversation.” Agrawal said the company would “focus

less on thinking about free speech” because “speech is easy on the internet. Most people

can speak. Where our role is particularly emphasized is who can be heard.”

The sweeping standards revealed at these hearings were defended by members as

necessary to avoid “insurrections” and other social harms. What is particularly distressing

is to hear members repeatedly defending censorship by citing Oliver Wendell Holmes’

famous statement on “shouting fire in a crowded theater.” This mantra has been grossly

misused as a justification for censorship. From statements on the pandemic to climate

change, anti-free speech advocates are claiming that opponents are screaming “fire” and

causing panic. The line comes from Schenck v. United States, a case that discarded the

free speech rights of citizens opposing the draft. Charles Schenck and Elizabeth Baer

were leading socialists in Philadelphia who opposed the draft in World War I. Fliers were

distributed that encouraged men to “assert your rights” and stand up for their right to

refuse such conscription as a form of involuntary servitude. Writing for the Court, Justice

Oliver Wendell Holmes dismissed the free speech interests in protecting the war and the

draft. He then wrote the most regrettable and misunderstood judicial soundbites in

history: “the character of every act depends on the circumstances in which it is done . . .

The most stringent protection of free speech would not protect a man in falsely shouting

fire in a theater and causing a panic.” “Shouting fire in a crowded theater” quickly

became a mantra for every effort to curtail free speech.

Holmes sought to narrow his clear and present danger test in his dissent

in Abrams v. United States. He warned that “we should be eternally vigilant against

attempts to check the expression of opinions that we loath and believe to be frought (sic)

with death, unless they so imminently threaten immediate interference with the lawful

and pressing purposes of the law that at an immediate check is required to save the

country.” Holmes’ reframing of his view would foreshadow the standard

in Brandenburg v. Ohio, where the Supreme Court ruled that even calling for violence is

protected under the First Amendment unless there is a threat of “imminent lawless

action and is likely to incite or produce such action.” However, members are still

channeling the standard from Schenck, which is a curious choice for most Democrats in

using a standard used against socialists and anti-war protesters.

Even more unnerving is the fact that Navaroli’s standard and those referencing

terms like “delegitimization” makes the Schenck standard look like the model of clarity.

Essentially, they add that you also have to consider the theater, movie, and audience to

decide what speech to allow. What could be treated as crying “Fire!” by any given person

or in any given circumstances would change according to their “nuanced” judgment.

III. CISA WITHIN THE GOVERNMENT-CORPORATE ALLIANCE

The role of CISA in this complex of government-corporate programs only

recently came into closer scrutiny. The Department of Homeland Security was previously

the focus of public controversy with the disclosure of the creation of Department’s

Disinformation Governance Board and the appointment of Nina Jankowicz, its head.

Jankowicz was a long advocate for censorship in the name of combating disinformation.

At the time, White House press secretary Jen Psaki described the board as intended “to

prevent disinformation and misinformation from traveling around the country in a range

of communities.”

While the Department ultimately yielded to the public outcry over the

board and disbanded it, the public was never told of a wide array other offices doing

much of the same work in targeting citizens and groups for possible censorship.

In January 2017, the Homeland Security declared that election infrastructure

would be treated as “critical infrastructure.” CISA took a lead in supporting election

infrastructure integrity and countering election misinformation. In 2018, CISA and its

Countering Foreign Influence Task Force (CFITF) reportedly assumed a greater role in

monitoring and counteracting foreign interference in U.S. elections. In 2020, this work

appears to have expanded further to pursue allegations of “switch boarding” by domestic

actors, or individuals thought to be acting as conduits for information undermining

elections or critical infrastructure. Much about this work remains unclear and I am no

expert on CISA or its operational profile. However, the expanding mandate of CISA

follows a strikingly familiar pattern.

The Twitter Files references CISA participation in these coordination meetings.

Given a mandate to help protect election integrity, CISA plunged into the monitoring and

targeting of those accused of disinformation. Infrastructure was interpreted to include

speech. As its director, Jen Easterly, declared “the most critical infrastructure is our

cognitive infrastructure” and thus included “building that resilience to misinformation

and disinformation, I think, is incredibly important.”

She pledged to continue that work

with the private sector including social media companies on that effort. We do not need

the government in the business of building our “cognitive infrastructure.” Like content

Press Briefing by Press Secretary Jen Psaki, April 29, 2022, https://www.whitehouse.gov/briefing-

room/press-briefings/2022/04/28/press-briefing-by-press-secretary-jen-psaki-april-28-2022/.

Maggie Miller, Cyber Agency Beefing Up Disinformation, Misinformation Team, THE HILL, Nov.

10, 2022, https://thehill.com/policy/cybersecurity/580990-cyber-agency-beefing-up-disinformation-

misinformation-team/.

moderation, the use of this euphemism does not disguise the government’s effort to direct

and control what citizens may read or say on public platforms.

Over the years, the range of information deemed harmful has expanded to the

point that even true information is now viewed as harmful for the purposes of censorship.

Some of the recent disclosures from Twitter highlighted the work of Stanford’s Virality

Project which insisted “true stories … could fuel hesitancy” over taking the vaccine or

other measures.

It is reminiscent of the sedition prosecutions under the Crown before

the American revolution where truth was no defense. Even true statements could be

viewed as seditious and criminal. Once the government gets into the business of speech

regulation, the appetite for censorship becomes insatiable as viewpoints are deemed

harmful, even if true. CISA shows the same broad range of suspect speech:

§ Misinformation is false, but not created or shared with the intention of causing

harm.

§ Disinformation is deliberately created to mislead, harm, or manipulate a person,

social group, organization, or country.

§ Malinformation is based on fact, but used out of context to mislead, harm, or

manipulate. An example of malinformation is editing a video to remove important

context to harm or mislead.”

MDM regulations offer the government the maximal space for censorship based on how

information may be received or used. The inclusion of true material used to “manipulate”

others is particularly chilling as a rationale for speech controls.

According to the Election Integrity Partnership (EIP), “tickets” flag material for

investigation that can be “one piece of content, an idea or narrative, or hundreds of URLs

pulled in a data dump.”

These tickets reportedly include those suspected of

“delegitimization,” which includes speech that undermines or spread distrust in the

political or electoral system. The ill-defined character of these categories is by design. It

allows for highly selective or biased “ticketing” of speech. The concern is that

conservative writers or sites subjected to the greatest targeting or ticketing. This pattern

was evident in other recent disclosures from private bodies working with U.S. agencies.

For example, we recently learned that the U.S. State Department funding for the National

Endowment for Democracy (NED) included support for the Global Disinformation

Index (GDI).

The British group sought to discourage advertisers from supporting sites

Jonathan Turley, True Stores …Could Fuel Hesitancy”: Stanford Project Worked to Censor Even

True Stories on Social Media, Res Ipsa Blog (www.jonathanturley.org), March 19, 2023, at

https://jonathanturley.org/2023/03/19/true-stories-could-fuel-hesitancy-stanford-project-worked-to-censor-

even-true-stories-on-social-media/.

Foreign Influence Operations and Disinformation,https://www.cisa.gov/topics/election-

security/foreign-influence-operations-and-disinformation.

ELECTION INTEGRITY PARTNERSHIP, THE LONG FUSE MISINFORMATION AND THE

2020 ELECTION 9 (2021), https://stacks.stanford.edu/file/druid:tr171zs0069/EIP-Final-Report.pdf

Jonathan Turley, Scoring Speech: How the Biden Administration has been Quietly Shaping Public

Discourse, Res Ipsa Blog (www.jonathanturley.org), Feb. 20, 2023,

https://jonathanturley.org/2023/02/20/scoring-speech-how-the-biden-administration-has-been-quietly-

shaping-speech/.

deemed dangerous due to disinformation. Companies were warned by GDI about “risky”

sites that pose “reputational and brand risk” and asked them to avoid “financially

supporting disinformation online.” All ten of the “riskiest” sites identified by the GDI are

popular with conservatives, libertarians, and independents, including Reason, a site

featuring legal analysis of conservative law professors. Liberal sites like HuffPost were

ranked as the most trustworthy. The categories were as ill-defined as those used by CISA.

RealClearPolitics was blacklisted due to what GDI considers “biased and sensational

language.” The New York Post was blacklisted because “content sampled from the Post

frequently displayed bias, sensationalism and clickbait, which carries the risk of

misleading the site’s reader.” After the biased blacklisting was revealed, NED announced

that it would withdraw funding for the organization. However, as with the Disinformation

Board, the Disinformation Index was just one of a myriad of groups being funded or fed

information from federal agencies. These controversies have created a type of “Whack-a-

mole” challenge for the free speech community. Every time one censorship partnership is

identified and neutralized, another one pops up.

EIP embodies this complex of groups working with agencies. It describes itself as

an organization that “was formed between four of the nation’s leading institutions

focused on understanding misinformation and disinformation in the social media

landscape: the Stanford Internet Observatory, the University of Washington’s Center for

an Informed Public, Graphika, and the Atlantic Council’s Digital Forensic Research

Lab.” The EIP has referred to CISA as one of its “stakeholders” and CISA has used the

partnership to censor individuals or groups identified by the agency. We still do not the

full extent of the coordination between CISA and other agencies with private and

academic groups in carrying out censorship efforts. However, the available evidence

raises legitimate questions over an agency relationship for the purposes of the First

Amendment.

IV. OUTSOURCING CENSORSHIP: THE NEED FOR GREATER

TRANSPARENCY AND ACCOUNTABILITY

In recent years, a massive censorship complex has been established with

government, academic, and corporate components. Millions of posts and comments are

now being filtered through this system in arguably the most sophisticated censorship

system in history. This partnership was facilitated by the demands of the First

Amendment, which bars the government from directly engaging in forms of prior

restraint and censorship. If “necessity is the mother of invention,” the censorship complex

shows how inventive motivated people can be in circumventing the Constitution. It has

been an unprecedented challenge for the free speech community. The First Amendment

was designed to deal with the classic threat to free speech of a government-directed

system of censorship. However, the traditional model of a ministry of information is now

almost quaint in comparison to the current system. It is possible to have an effective state

media by consent rather than coercion. There is no question that the work of these

academic and private groups limits free speech. Calling opposing views disinformation,

malinformation, or misinformation does not sanitize the censorship. It is still censorship

being conducted through a screen of academic and corporate entities. It may also

contravene the First Amendment.

The government can violate the Constitution through public employees or private

actors. As I testified recently before the Judiciary Committee, this agency relationship

can be established through consent or coercion. Indeed, the line can be difficult to discern

in many cases. There is an argument that this is a violation of the First Amendment.

Where the earlier debate over the status of these companies under Section 230 remained

mired in speculation, the recent disclosures of government involvement in the Twitter

censorship program presents a more compelling and concrete case for arguing agency

theories. These emails refer to multiple agencies with dozens of employees actively

coordinating the blacklisting and blocking of citizens due to their public statements.

There is no question that the United States government is actively involved in a massive

censorship system. The only question is whether it is in violation of the First

Amendment.

Once again, the Twitter Files show direct action from federal employees to censor

viewpoints and individual speakers on social media. The government conduct is direct

and clear. That may alone be sufficient to satisfy courts that a program or policy abridges

free speech under the First Amendment. Even if a company like Twitter declined

occasionally, the federal government was actively seeking to silence citizens. Any

declinations only show that that effort was not always successful.

In addition to that direct action, the government may also be responsible for the

actions of third parties who are partnering with the government on censorship. The

government has long attempted to use private parties to evade direct limits imposed by

the Constitution. Indeed, this tactic has been part of some of the worst chapters in our

history. For example, in Lombard v. Louisiana,

the Supreme Court dealt with the denial

of a restaurant to serve three black students and one white student at a lunch counter in

New Orleans reserved for white people. The Court acknowledged that there was no state

statute or city ordinance requiring racial segregation in restaurants. However, both the

Mayor and the Superintendent of Police had made public statements that “sit-in

demonstrations” would not be permitted. The Court held that the government cannot do

indirectly what it cannot do directly. In other words, it “cannot achieve the same result by

an official command which has at least as much coercive effect as an ordinance.”

As the Court said in Blum v. Yaretsky (where state action was not found), “a State

normally can be held responsible for a private decision only when it has exercised

coercive power or has provided such significant encouragement, either overt or covert,

that the choice must in law be deemed to be that of the State.”

Past cases (often dealing

with state action under the Fourteenth Amendment) have produced different tests for

establishing an agency relationship, including (1) public function; (2) joint action; (3)

governmental compulsion or coercion; and (4) governmental nexus.

Courts have noted

373 U.S. 267 (1963).

Id. at 273.

Blum v. Yaretsky, 457 U.S. 991, 1004-05 (1982).

Pasadena Republican Club v. W. Justice Ctr., 985 F.3d 1161, 1167 (9th Cir. 2021); Kirtley v.

Rainey, 326 F.3d 1088, 1092 (9th Cir. 2003). Some courts reduce this to three tests.

that these cases “overlap” in critical respects.

I will not go into each of these tests but

they show the highly contextual analysis performed by courts in finding private conduct

taken at the behest or direction of the government. The Twitter Files show a multilayered

incorporation of government information, access, and personnel in the censorship

program. One question is “whether the state has so far insinuated itself into a position of

interdependence with [the private entity] that it must be recognized as a joint participant

in the challenged activity.”

Nevertheless, the Supreme Court noted in Blum that “[m]ere

approval of or acquiescence in the initiatives of a private party is not sufficient to justify

holding the State responsible for those initiatives.”

Courts have previously rejected claims of agency by private parties over social

media.

However, these cases often cited that lack of evidence of coordination and

occurred before the release of the Twitter Files. For example, in Rogalinski v. Meta

Platforms, Inc.,

the court rejected a claim that Meta Platforms, Inc. violated the First

Amendment when it censored posts about COVID-19. However, the claim was based

entirely on a statement by the White House Press Secretary and “all of the alleged

censorship against Rogalinski occurred before any government statement.” It noted that

there was no evidence that there was any input of the government to challenge the

assertion that Meta’s message was “entirely its own.”

There is an interesting comparison to the decision of the United States Court of

Appeals for the Sixth Circuit in Paige v. Coyner, where the Court dealt with the

termination of an employee after a county official called her employer to complain about

comments made in a public hearing.

The court recognized that “[t]his so-called state-

actor requirement becomes particularly complicated in cases such as the present one

where a private party is involved in inflicting the alleged injury on the plaintiff.”

However, in reversing the lower court, it still found state action due to the fact that a

government official made the call to the employer, which prompted the termination.

Likewise, in Dossett v. First State Bank, the United States Court of Appeals for

the Eighth Circuit ruled that the termination of a bank employee was the result of state

action after school board members contacted her employer about comments made at a

public-school board meeting.

The Eighth Circuit ruled that the district court erred by

instructing a jury that it had to find that the school board members had “actual authority”

to make these calls. In this free speech case, the court held that you could have state

action under the color of law when the “school official who was purporting to act in the

performance of official duties but was acting outside what a reasonable person would

believe the school official was authorized to do.”

In this case, federal officials are

clearly acting in their official capacity. Indeed, that official capacity is part of the concern

Rogalinski v. Meta Platforms, Inc., 2022 U.S. Dist. LEXIS 142721 (August 9, 2022).

Gorenc v. Salt River Project Agr. Imp. & Power Dist., 869 F.2d 503, 507 (9th Cir. 1989).

Blum, 457 U.S. at 1004-05.

O’Handley v. Padilla, 579 F. Supp.3d 1163 1192-93 (N.D. Cal. 2022).

2022 U.S. Dist. LEXIS 142721 (August 9, 2022).

Id.

Paige v. Coyner, 614 F.3d 273, 276 (6th Cir. 2010).

Id.

399 F.3d 940 (8th Cir. 2005).

Id. at 948.

raised by the Twitter Files: the assignment of dozens of federal employees to support a

massive censorship system.

Courts have also ruled that there is state action where government officials use

their positions to intimidate or pressure private parties to limit free speech. In National

Rifle Association v. Vullo, the United States Court of Appeals for the Second Circuit

ruled that a free speech claim could be made on the basis of a state official’s pressuring

companies not to do business with the NRA.

The Second Circuit held “although

government officials are free to advocate for (or against) certain viewpoints, they may not

encourage suppression of protected speech in a manner that ‘can reasonably be

interpreted as intimating that some form of punishment or adverse regulatory action will

follow the failure to accede to the official’s request.’”

It is also important to note that

pressure is not required to establish an agency relationship under three of the prior tests.

It can be based on consent rather than coercion.

We have seen how censorship efforts began with claims of foreign interference

and gradually expanded into general efforts to target harm or “delegitimizing” speech.

The Twitter Files show FBI officials warning Twitter executives that their platform was

being targeted by foreign powers, including a warning that an executive cited as a basis

for blocking postings related to the Hunter Biden laptop. At the same time, various

members of Congress have warned social media companies that they could face

legislative action if they did not continue to censor social media. Indeed, after Twitter

began to reinstate free speech protections and dismantle its censorship program, Rep.

Schiff (joined by Reps. André Carson (D-Ind.), Kathy Castor (D-Fla.) and Sen. Sheldon

Whitehouse (D-R.I.)) sent a letter to Facebook, warning it not to relax its censorship

efforts. The letter reminded Facebook that some lawmakers are watching the company

“as part of our ongoing oversight efforts” — and suggested they may be forced to

exercise that oversight into any move by Facebook to “alter or rollback certain

misinformation policies.” This is only the latest such warning. In prior hearings, social

media executives were repeatedly told that a failure to remove viewpoints were

considered “disinformation.” For example, in a November 2020 Senate hearing, then-

Twitter CEO Jack Dorsey apologized for censoring the Hunter Biden laptop story. But

Sen. Richard Blumenthal, D-Conn., warned that he and his Senate colleagues would not

tolerate any “backsliding or retrenching” by “failing to take action against dangerous

disinformation.”

Senators demands increased censorship in areas ranging from the

pandemic to elections to climate change.

These warnings do not necessarily mean that a court would find that executives

were carrying out government priorities. An investigation is needed to fully understand

the coordination and the communications between the government and these companies.

In Brentwood Academy v. Tennessee Secondary School Athletic Assn.,

the Supreme

Court noted that state action decisions involving such private actors are highly case

specific:

National Rifle Association of America v. Vullo, 49 F.4th 700, 715 (2d Cir. 2022).

Id. (quoting Hammerhead Enters., Inc. v. Brezenoff, 707 F.2d 33, 39 (2d Cir. 1983)).

Misinformation and Disinformation on Online Platforms: Hearing Before the Subcomm. on

Commc’n & Tech. and Subcomm. on Consumer Protection of the H. Comm. on Energy & Com., 117th Cong.

(2021).

531 U.S. 288 (2001).

What is fairly attributable is a matter of normative judgment, and the criteria lack

rigid simplicity. From the range of circumstances that could point toward the

State behind an individual face, no one fact can function as a necessary condition

across the board for finding state action; nor is any set of circumstances

absolutely sufficient, for there may be some countervailing reason against

attributing activity to the government…

Our cases have identified a host of facts that can bear on the fairness of such an

attribution. We have, for example, held that a challenged activity may be state

action when it results from the State’s exercise of “coercive power,” …when the

State provides “significant encouragement, either overt or covert,” … or when a

private actor operates as a ‘willful participant in joint activity with the State or its

agents,” … We have treated a nominally private entity as a state actor when it is

controlled by an “agency of the State,” … when it has been delegated a public

function by the State, … when it is “entwined with governmental policies,” or

when government is “entwined in [its] management or control.”

Obviously, many of these elements appear present. However, the Twitter Files also show

executives occasionally declining to ban posters targeted by the government. It also

shows such pressure coming from the legislative branch. For example, the Twitter Files

reveal that Twitter refused to carry out censorship requests from at least one member

targeting a columnist and critic. Twitter declined and one of its employees simply wrote,

“no, this isn’t feasible/we don’t do this.”

There were also requests from Republicans to

Twitter for action against posters, including allegedly one from the Trump White House

to take down content.

We simply do not know the extent of what companies like Twitter “did do,” nor

for whom. We do not know how demands were declined when flagged by the CISA, FBI,

or other agencies. The report from Twitter reviewers selected by Elon Musk suggests that

most requests coming from the Executive Branch were granted. That is one of the areas

that could be illuminated by this select subcommittee. The investigation may be able to

supply the first comprehensive record of the government efforts to use these companies

to censor speech. It can pull back the curtain on America’s censorship system so that both

Congress and the public can judge the conduct of our government.

Whether the surrogate censorship conducted by social media companies is a form

of government action may be addressed by the courts in the coming years. However,

certain facts are well-established and warrant congressional action. First, while these

companies and government officials prefer to call it “content moderation,” these

companies have carried out the largest censorship system in history, effectively

Id. at 296.

Jonathan Turley, “We Don’t Do This”: Twitter Censors Rejected Adam Schiff’s Censorship

Request, THE HILL, Jan. 5, 2023, https://thehill.com/opinion/judiciary/3800380-we-dont-do-this-even-

twitters-censors-rejected-adam-schiffs-censorship-request/.

This included the Trump White House allegedly asking to take down derogatory tweets from the

wife of John Legend after the former president attacked the couple. Moreover, some Trump officials

supported efforts to combat foreign interference and false information on social media. It has been reported

that Twitter has a “database” of Republican demands. Adam Rawnsley and Asawin Suebaeny, Twitter Kept

Entire “Database” of Republican Requests to Censor Posts, ROLLING STONE, Feb. 8, 2023,

https://www.rollingstone.com/politics/politics-news/elon-trump-twitter-files-collusion-biden-censorship-

1234675969/.

governing the speech of billions of people. The American Civil Liberties Union, for

example, maintains that censorship applies to both government and private actions. It is

defined as “the suppression of words, images, or ideas that are ‘offensive,’ [and] happens

whenever some people succeed in imposing their personal political or moral values on

others.”

Adopting Orwellian alternative terminology does not alter the fact that these

companies are engaging in the systemic censoring of viewpoints on social media.

Second, the government admits that it has supported this massive censorship

system. Even if the censorship is not deemed government action for the purposes of the

First Amendment, it is now clear that the government has actively supported and assisted

in the censorship of citizens. Objecting that the conduct of government officials may not

qualify under the First Amendment does not answer the question of whether members

believe that the government should be working for the censorship of opposing or

dissenting viewpoints. During the McCarthy period, the government pushed blacklists for

suspected communists and the term “fellow travelers” was rightfully denounced

regardless of whether it qualified as a violation of the First Amendment. Even before Joe

McCarthy launched his un-American activities hearings, the Justice Department created

an effective blacklist of organizations called “Attorney General’s List of Subversive

Organizations” (AGLOSO) that was then widely distributed to the media and the public.

It became the foundation for individual blacklists.

The maintenance of the list fell to the

FBI. Ultimately, blacklisting became the norm with both legislative and executive

officials tagging artists, writers, and others. As Professor Geoffrey Stone observed,

“Government at all levels hunted down ‘disloyal’ individuals and denounced them.

Anyone so stigmatized became a liability to his friends and an outcast to society.”

the time, those who raised the same free speech objections were also attacked as “fellow

travelers” or “apologists” for communists. It was wrong then and it is wrong now. It was

an affront to free speech values that have long been at the core of our country. It is not

enough to say that the government is merely seeking the censorship of posters like any

other user. There are many things that are more menacing when done by the government

rather than individuals. Moreover, the government is seeking to silence certain speakers

in our collective name and using tax dollars to do so. The FBI and other agencies have

massive powers and resources to amplify censorship efforts. The question is whether

Congress and its individual members support censorship whether carried out by corporate

or government officials on social media platforms.

Third, the government is engaged in targeting users under the ambiguous

mandates of combating disinformation or misinformation. These are not areas

traditionally addressed by public affairs offices to correct false or misleading statements

American Civil Liberties Union, What is Censorship?, https://www.aclu.org/other/what-

censorship.

Robert Justin Goldstein, Prelude to McCarthyism, PROLOGUE MAGAZINE, Fall 2006,

https://www.archives.gov/publications/prologue/2006/fall/agloso.html. Courts pushed back on the listing to

require some due process for those listed.

Geoffrey R. Stone, Free Speech in the Age of McCarthy: A Cautionary Tale, 93 CALIF. L. REV.

1387, 1400 (2005).

The distinction between these companies from other corporate entities like the NFL or Starbucks

is important. There is no question that businesses can limit speech on their premises and by their own

employees. However, these companies constitute the most popular communication platforms in the

country. They are closer to AT&T than Starbucks in offering a system of communication.

made about an agency’s work. The courts have repeatedly said that agencies are allowed

to speak in their voices without viewpoint neutrality.

As the Second Circuit stated,

“[w]hen it acts as a speaker, the government is entitled to favor certain views over

others.”

This was an effort to secretly silence others. Courts have emphasized that “[i]t

is well-established that First Amendment rights may be violated by the chilling effect of

governmental action that falls short of a direct prohibition against speech.”

These public

employees were deployed to monitor and target user spreading “disinformation” on a

variety of subjects, from election fraud to government corruption. The Twitter Files show

how this mandate led to an array of abuses, from targeting jokes to barring opposing

scientific views.

These facts already warrant bipartisan action from Congress. Free speech

advocates have long opposed disinformation mandates as an excuse or invitation for

public or private censorship. I admittedly subscribe to the view that the solution to bad

speech is better speech, not speech regulation.

Justice Brandeis embraced the view of

the Framers that free speech was its own protection against false statements: “If there be

time to discover through discussion the falsehood and the fallacies, to avert the evil by

the processes of education, the remedy to be applied is more speech not enforced

silence.”

We have already seen how disinformation was used to silence dissenting

views of subjects like mask efficacy and Covid policies like school closures that are now

being recognized as legitimate.

We have also seen how claims of Russian trolling operations may have been

overblown in their size or their impact. Indeed, even some Twitter officials ultimately

concluded that the FBI was pushing exaggerated claims of foreign influence on social

media.

The Twitter Files refer to sharp messages from the FBI when Twitter failed to

find evidence supporting the widely reported foreign trolling operations. One Twitter

official referred to finding “no links to Russia.” This was not for want of trying. Spurred

on by the FBI, another official promised “I can brainstorm with [redacted] and see if we

can dig even deeper and try to find a stronger connection.” The pressure from the FBI led

Roth to tell his colleagues that he was “not comfortable” with the agenda of the FBI and

said that it reminded him of something “more like something we’d get from a

congressional committee than the Bureau.”

The danger of censorship is not solely a concern of one party. To his great credit,

Rep. Ro Khanna (D., Cal.) in October 2020, said that he was appalled by the censorship

Pleasant Grove City v. Summum, 555 U.S. 460, 467-68 (2009); Johanns v. Livestock Mktg.Ass’n,

544 U.S. 550, 553 (2005).

Wandering Dago, Inc. v. Destito, 879 F.3d 20, 34 (2d Cir. 2018).

Zieper v. Metzinger, 474 F.3d 60, 65 (2d Cir. 2007).

See generally Jonathan Turley, Harm and Hegemony: The Decline of Free Speech in the United

States, 45 HARV. J.L. & PUB. POL’Y 571 (2022).

Whitney, 274 U.S. at 375, 377.

In his testimony, Roth stated that they found substantial Russian interference impacting the

election. Protecting Speech from Government Interference and Social Media Bias, Part 1: Twitter’s Role in

Suppressing the Biden Laptop Story: Hearing Before the H. Comm. on Oversight & Accountability, 118th

Cong. (2023) (statement of Yael Roth, Former Head of Trust and Safety, Twitter). That claim stands in

conflict with other studies and reports, but it can also be addressed as part of the investigation into these

communications.

and was alarmed by the apparent “violation of the 1st Amendment principles.”

Congress can bar the use of federal funds for such disinformation offices. Such

legislation can require detailed reporting on agency efforts to ban or block public

comments or speech by citizens. Even James Baker told the House Oversight Committee

that there may be a need to pass legislation to limit the role of government officials in

their dealings with social media companies.

Legislation can protect the legitimate role

of agencies in responding and disproving statements made out its own programs or

policies. It is censorship, not disinformation, that has damaged our nation in recent years.

Free speech like sunshine can be its own disinfectant. In Terminiello v. City of Chicago,

the Supreme Court declared that:

The right to speak freely and to promote diversity of ideas . . . is . . . one of the

chief distinctions that sets us apart from totalitarian regimes . . . [A] function

of free speech under our system of government is to invite dispute. . . . Speech is

often provocative and challenging. . . [F]reedom of speech, though not absolute, is

nevertheless protected against censorship.

Disinformation does cause divisions, but the solution is not to embrace government-

corporate censorship. The government effort to reduce speech does not solve the problem

of disinformation. It does not change minds but simply silences voices in national

debates.

V. CONCLUSION

There is obviously a deep division in Congress over censorship, with many

members supporting the efforts to blacklist and remove certain citizens or groups from

social media platforms. That is a debate that many of us in the free speech community

welcome. However, let it be an honest and open debate. The first step in securing such a

debate is to support transparency on the full extent of these efforts by federal agencies.

The second step is to allow these questions to be discussed without attacking journalists

and witnesses who come to Congress to share their own concerns over the threat to both

free press and free speech values. Calling reporters “so called journalists” or others

“Putin lovers” represent a return to the rhetoric used against free speech advocates during

the Red Scare.

We are better than that as a country and our Constitution demands more

from this body. If members want to defend censorship, then do so with the full record

before the public on the scope and standards of this government effort.

Democratic Rep. Ro Khana Expressed Concerns Over Twitter’s Censorship of Hunter Biden

Laptop, FOX NEWS, Dec. 2, 2022, https://www.foxnews.com/politics/democratic-rep-ro-khanna-expressed-

concerns-twitters-censorship-hunter-biden-laptop-story.

Protecting Speech from Government Interference and Social Media Bias, Part 1: Twitter’s Role in

Suppressing the Biden Laptop Story: Hearing Before the H. Comm. on Oversight & Accountability, 118th

Cong. (2023) (statement of James Baker, Former General Counsel, FBI.

Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949) (citations omitted).

Jonathan Turley, Is the Red Scare Turning Blue?, Res Ipsa Blog (www.jonathanturley.org), Feb. 12,

2023, https://jonathanturley.org/2023/02/12/is-the-red-scare-going-blue-democrats-accuse-government-

critics-of-being-putin-lovers-and-supporting-insurrectionists/.

The public understands the threat to free speech and strongly supports an

investigation into the FBI’s role in censoring social media. Despite the push for

censorship by some politicians and pundits, most Americans still want free-speech

protections. It is in our DNA. This country was founded on deep commitments to free

speech and limited government – and that constitutional tradition is no conspiracy theory.

Polls show that 73% of Americans believe that these companies censored material for

political purposes.

Another poll showed that 63% want an investigation into FBI

censorship allegations.

Adlai Stevenson famously warned of this danger: “Public confidence in the

integrity of the Government is indispensable to faith in democracy; and when we lose

faith in the system, we have lost faith in everything we fight . . . for.” Senator

Stevenson’s words should resonate on both sides of our political divide and that we

might, even now, find a common ground and common purpose. The loss of faith in our

government creates political instabilities and vulnerabilities in our system. Moreover,

regardless of party affiliation, we should all want answers to come of these questions. We

can differ on our conclusions, but the first step for Congress is to force greater

transparency on controversies involving bias to censorship. One of the greatest values of

oversight is to allow greater public understanding of the facts behind government actions.

Greater transparency is the only course that can help resolve the doubts that many have

over the motivations and actions of their government. I remain an optimist that it is still

possible to have a civil and constructive discussion of these issues. Regardless of our

political affiliations and differences, everyone in this room is here because of a deep love

and commitment to this country. It was what brought us from vastly different

backgrounds and areas in our country. We share a single article of faith in our

Constitution and the values that it represents. We are witnessing a crisis of faith today

that must be healed for the good of our entire nation. The first step toward that healing is

an open and civil discussion of the concerns that the public has with our government. We

can debate what measures are warranted in light of any censorship conducted with

government assistance. However, we first need to get a full and complete understanding

of the relationship between federal agencies and these companies in the removal or

suspension of individuals from social media. At a minimum, that should be a position that

both parties can support in the full disclosure of past government conduct and

communications with these companies.

Once again, thank you for the honor of appearing before you to discuss these

important issues, and I would be happy to answer any questions from the Committee.

Jonathan Turley

J.B. & Maurice C. Shapiro Chair of Public Interest Law

George Washington University

Sean Burch, Nearly 75% of Americans Believe Twitter, Facebook Censor Posts Based on

Viewpoints, Pew Finds, THE WRAP, Aug. 19, 2020, https://www.thewrap.com/nearly-75-percent-twitter-

facebook-censor/.

63% Want FBI’s Social Media Activity Investigated, RASMUSSEN REPORTS, Dec. 26, 2022,

https://www.rasmussenreports.com/public_content/politics/partner_surveys/twittergate_63_want_fbi_s_soc

ial_media_activity_investigated.

CENSORSHIP CAN BE DEADLY

!"#$%&$'()*+,#*-"*.",./01*2&"".%3*%4*$)""+'*($*,/5,0$*(3)%&6,#67*-86*(6*($*"$)"+(,//0*(3)%&6,#6*.8&(#9*,*

#,:%#,/*"3"&9"#+0*$8+'*,$*,*),#."3(+1*;%*,86'%&(60*($*(#4,//(-/"7*,#.*5'"#*,*#"5*<(&8$*"3"&9"$7*(6*($*

(3)%$$(-/"*4%&*)%/(:+(,#$* ,#.* )8-/(+* '",/6'*%=+(,/$* 6%* 9"6* 6'(#9$* &(9'6* 5(6'%86* /($6"#(#9* 6%* .($+8$$(%#$*

-"65""#*,*5(."*+,$6*%4*$+("#:$6$*5(6'*.(>"&"#6*,&",$*%4*"?)"&:$"*,#.*6'%89'6$1**

@*,3*,#*")(."3(%/%9($67*,*-(%$6,:$:+(,#*,#.*,*)&%4"$$%&*%4*3".(+(#"*,6*A,&<,&.7*%#*/",<"1*2%&*%<"&*65%*

."+,."$*@*',<"*.%#"*&"$",&+'*%#*6'"*."6"+:%#*,#.*3%#(6%&(#9*%4*(#4"+:%8$*.($",$"*%86-&",B$*,#.*%#*6'"*

$,4"60*"<,/8,:%#*%4*<,++(#"$*,#.*.&89$1*@*'"/)".*-8(/.*6'"*#,:%#$*.($",$"*$8&<"(//,#+"*$0$6"3$1*C"$)(6"*

6'($7*@*5,$*+"#$%&".*,#.*-/,+B*/($6".*.8&(#9*6'"*),#."3(+7*-0*D5(E"&7*F(#B".@#7*G%8D8-"*,#.*2,+"-%%B1**

@#*",&/0*HIHI7*5"*,/&",.0*B#"5*4&%3*J8',#*.,6,*6',6*6'"&"*($*3%&"*6',#*,*6'%8$,#.K4%/.*.(>"&"#+"*(#*

!%<(.*3%&6,/(60*-"65""#*6'"*%/.*,#.*6'"*0%8#9

1*C8&(#9*6'"*),#."3(+7*5"*4,(/".*6%*,."L8,6"/0*)&%6"+6*

%/."&* M3"&(+,#$* 5'(/"* $+'%%/* +/%$8&"$* ,#.* %6'"&* /%+B.%5#* 3",$8&"$* 9"#"&,6".* "#%&3%8$* +%//,6"&,/*

)8-/(+*'",/6'*.,3,9"*6',6*5"*#%5*38$6*/(<"*5(6'7*,#.*.("*4&%37*4%&*0",&$*6%*+%3"1*

C8&(#9*6'"*$)&(#9*%4*HIHI7*N5"."#*5,$*6'"*%#/0*3,O%&*J"$6"&#*+%8#6&0*6%*B"")*$+'%%/$*,#.*.,0+,&"*%)"#*

4%&*+'(/.&"#*,9"$*P*6%*PQ1*M3%#9*6'%$"*P1R*3(//(%#*+'(/.&"#7*6'"&"*5"&"*S"&%*+%<(.*.",6'$*,#.*6'"*+%<(.*

&($B*4%&*6",+'"&$*5"&"*/"$$*6',#*6'"*,<"&,9"*%4*%6'"&*)&%4"$$(%#$

1*D'($*$'%5".*6',6*(6*5,$*$,4"*6%*B"")*

$+'%%/$*%)"#1*@6*5,$*(3)%&6,#6*4%&*M3"&(+,*6%*B#%5*6',67*-86*,*T8/0*HIHI*;"5*U#9/,#.*T%8&#,/*%4*V".(+(#"*

,&:+/"*%#*$+'%%/*+/%$8&"*.(.*#%6*"<"#*3"#:%#*N5"."#

1*D',6W$*/(B"*&")%&:#9*%#*,*#"5*3".(+,/*6&",63"#6*

5(6'%86*(#+/8.(#9*(#4%&3,:%#*4&%3*6'"*+%3),&($%#*+%#6&%/*9&%8)1**

X#,-/"*6%*)8-/($'*30*6'%89'6$*,-%86*6'"*),#."3(+*(#*XN*U#9/($'*/,#98,9"*3".(,7*(#*6'"*$833"&*%4*HIHI*

@*8$".*30*D5(E"&*,++%8#6*6%*$',&"*6'"*N5".($'*.,6,*,#.*,&98"*4%&*%)"#*$+'%%/$1*Y86*(#*T8/0*HIHI7*D5(E"&*

)86*3"*%#*6'"(&*Z6&"#.$*-/,+B/($6[*6%*/(3(6*6'"*&",+'*%4*30*%)"#*$+'%%/*)%$6$

@#*\+6%-"&*HIHI7*@*,86'%&".*6'"*]&",6*Y,&&(#96%#*C"+/,&,:%#*5(6'*65%*4"//%5*")(."3(%/%9($6$7*C&1*N8#"6&,*

]8)6,*,6*\?4%&.*,#.*C&1*T,0*Y',E,+',&0,*,6*N6,#4%&.

1*J"*,&98".*4%&*-"E"&*)&%6"+:%#*%4*'(9'K&($B*%/."&*

)"%)/"*5'(/"* B"")(#9*$+'%%/$* %)"#* ,#.* /"^#9*0%8#9*)"%)/"* /(<"* 3%&"* #%&3,/* /(<"$1* D'($* 5,$* $',.%5*

-,##".*-0*]%%9/"*,#.*+"#$%&".*-0*_"..(6

1*M`"&*)%$:#9*(#*4,<%&*%4*)&(%&(:S(#9*6'"*"/."&/0*4%&*<,++(#,:%#7*

%8&*2,+"-%%B*),9"*5,$*“unpublished”.*a2(98&"*Pb*

@#*V,&+'*HIHP7*D5(E"&*+"#$%&".*,*)%$6*5'"#*@*5&%6"*6',6*“Thinking that everyone must be vaccinated is

as scien:ﬁcally ﬂawed as thinking that nobody should. COVID vaccines are important for older high-risk

people and their care-takers. Those with prior natural infec:on do not need it. Nor children.”*D5(E"&*4,/$"/0*

+/,(3".*6',6*6'"*65""6*5,$*3($/",.(#97*,#.*(6*+%8/.*#%6*-"*&")/(".*6%7*$',&".*%&*/(B".1*J"*',<"* B#%5#*

,-%86*(#4"+:%#K,+L8(&".*(338#(60*$(#+"*6'"*M6'"#(,#*c/,98"*(#*deI*Y!7*,#.*6'"*L8"$:%#(#97*."#(,/*,#.*

+"#$%&(#9*%4*$8+'*#,68&,/*(338#(60*($*6'"*3%$6*$68##(#9*."#(,/*%4*$+("#:f+*4,+6$*.8&(#9*6'"*),#."3(+1*

!"#$$%&'(!)*!+,-./012!+&#345'!)567#'57!89&#$%!:5!;<5!8=5>?@>*!A?3B5%.3C!;='?$!1DC!EDED*!

!F#:$?>!G56$49!;<53>H!&I!8J5%53*!+&K?%012!?3!7>9&&$>9?$%'53!L!;!>&M=6'?7&3!:54J553!N?3$63%!63%!8J5%53C!O#$H!PC!

EDED*!

!A5K?37&3!)C!+5K?B!)C!A?=7?4>9!)*!Q5&=53?3<!F'?M6'H!7>9&&$7!%#'?3<!495!F63%5M?>*!R5J!S3<$63%!O&#'36$!&I!

)5%?>?35C!O#$H!E2C!EDED*!

*T96U6>96'H6!O*!V964!.!%?7>&K5'5%!64!WJ?U5'!956%X#6'45'7C!Y3G5'%C!/5>5M:5'!EEC!EDEE*!

!T96U6>96'H6!OC!Z#=46!8C!"#$$%&'(!)*!Z'564!T6''?3<4&3!/5>$6'6[&3C!,>4&:5'!\C!EDED*!

!]&#3<!W*!!V9H!>63^4!J5!46$B!6:&#4!495!Z'564!T6''?3<4&3!/5>$6'6[&3_!W95!8=5>464&'C!,>4&:5'!1PC!EDED*!

J(6'* .",./0* +%#$"L8"#+"$1* M6* ,* :3"* 5'"#* <,++(#"$* 5"&"* (#* $'%&6* $8))/07* 5"* 5"&"*<,++(#,:#9* 0%8#9*

,.8/6$*,#.*)"%)/"*5(6'*#,68&,/*(338#(607*5'%*.(.*#%6*#"".*(67*-"4%&"*3,#0*%/."&*M3"&(+,#$*5'%*5'%$"*

/(<"$*+%8/.*',<"*-""#*$,<".*-0*(61*a2(98&"*Hb*

D'&%89'* &,#.%3(S".* $68.("$* ,#.* &"<("5$

7* 5"* B#%5* 6',6* 4,+"* 3,$B$* )&%<(."* %#/0* 3,&9(#,/* %&* #%*

)&%6"+:%#* ,9,(#$6*!%<(.1* M* &,#.%3(S".* $68.0* (#* C"#3,&B* $'%5".* #%* $(9#(f+,#6* -"#"f6

* 5'(/"* ,* G,/"*

X#(<"&$(60*$68.0* +%#.8+6".* (#* Y,#9/,."$'* $'%5".* ,* &".8+:%#* -"65""#*I* ,#.* PR* )"&+"#6

1* @6* ($* 6'"#*

.,#9"&%8$*6%*3,B"*%/."&*'(9'K&($B*M3"&(+,#$*-"/("<"*6',6*3,$B$*5(//*)&%6"+6*6'"3*5'"#*6'"0*5(//*#%67*,$*

6'"0*3,0*9%*6%*+&%5.".*&"$6,8&,#6$*%&*$8)"&*3,&B"6$*6'(#B(#9*6',6*6'"(&*3,$B*($*B"")(#9*6'"3*$,4"1*@#*

V,0*HIHP*@*5,$*6"3)%&,&(/0*$8$)"#.".*-0*D5(E"&*4%&*6'&""*5""B$*4%&*5&(:#9*6',6g*“Naively fooled to think

that masks would protect them, some older high-risk people did not socially distance properly, and some

died from Covid19 because of it. Tragic. Public health oﬃcials/scien:sts must always be honest with the

public”1*a2(98&"*eb*

D5(E"&* ,/$%* +"#$%&".* 3"* 4%&* L8%:#9* ,#.* /(#B(#9* 6%* ,#* ,&:+/"* ,-%86* 3,$B$* 5&(E"#* -0* ,#* M$$%+(,6"*

c&%4"$$%&*%4*Y/,+B*N68.("$*,6*6'"*X#(<"&$(60*%4*!,/(4%&#(,1*a2(98&"*db*

@#* M)&(/* HIHP* @* ),&:+(),6".* (#* ,* $+("#:f+* &%8#.6,-/"* '%$6".* -0* 2/%&(.,* 9%<"&#%&* _%#* C"N,#:$1* D'($*

&%8#.6,-/"*5,$*+"#$%&".*-0*G%8D8-"*,`"&*-"(#9*)%$6".*-0*,*!YN*,=/(,6".*6"/"<($(%#*$6,:%#*(#*2/%&(.,

G%8D8-"*($*%5#".*-0*]%%9/"1*

V,#0*#8&$"$*5"&"*(#4"+6".*5'(/"*'"&%(+,//0*6,B(#9*+,&"*%4*+%<(.*),:"#6$7*,#.*$%3"*%4*6'"3*5"&"*/,6"&*

f&".*4%&*#%6*6,B(#9*,*<,++(#"*"<"#*6'%89'*6'"0*',.*$6&%#9"&*(338#(60*6',#*6'"*<,++(#,6".1*@#*\+6%-"&*

HIHP7*@*5&%6"*,#*,&:+/"*8&9(#9*'%$)(6,/$*6%*'(&"*(#$6",.*%4*f&"*#8&$"$*5(6'*#,68&,/*(338#(607*,$*6'"0*,&"*

6'"* /",$6* /(B"/0* 6%* (#4"+6* %/."&* 4&,(/* ),:"#6$

1* D',6* 5,$* +"#$%&".* -0* F(#B".@#7* 5'(+'* ($* %5#".* -0*

V(+&%$%`

1*A%$)(6,/$*,#.*#8&$(#9*'%3"$*+%8/.*',<"*-"E"&*)&%6"+6".*),:"#6$*(4*6'"0*',.*,+:<"/0*'(&".*

)"&$%##"/*5(6'*(#4"+:%#*,+L8(&".*(338#(601*D',6*5%8/.*',<"*$,<".*/(<"$1*

F(#B".@#*+"#$%&".*3"*38/:)/"*:3"$7*%#"*%6'"&*"?,3)/"*-"(#9*,*&")%$6*%4*,#*(#6"&<("5*5(6'*6'"*+'("4*

")(."3(%/%9($6*%4*@+"/,#.

*a2(98&"$*QKhb1*D',6*($7*F(#B".@#*.(.*#%6*%#/0*+"#$%&*)8-/(+*'",/6'*,+,."3(+$*

-86*,/$%*9%<"&#3"#6*)8-/(+*'",/6'*%=+(,/$*6',6*.(.*#%6*+%#4%&3*6%*F(#B".@#W$*<("5*%#*6'"*),#."3(+1**

Jeﬀerson T, et al. Physical interven[ons to interrupt or reduce the spread of respiratory viruses. Cochrane Library,

January 30, 2023.

Liu AT, Prasad V, Darrow JJ. Evidence for Community Cloth Face Masking to Limit the Spread of SARS-CoV-2: A

Cri[cal Review, Cato Working Paper, November 8, 2021.

Bundgaard H, et al. Eﬀec[veness of Adding a Mask Recommenda[on to Oher Public Health Measures to prevent

SARS-CoV-2 Infec[on in Danish Mask Wearers: A Randomized Controlled Trial. Annals of Internal Medicine,

November 18, 2020.

Abaluck J, et al. Impact of community masking on COVID-19: A cluster-randomized trial in Bangladesh. Science,

December 2, 2021.

Wilson K, Ross A. YouTube removes video of DeSan[s coronavirus roundtable. Tampa Bay Times, April 9, 2021.

Kulldorﬀ M. Hospitals Should Hire, not Fire, Nurses with Natural Immunity. Brownstone Ins[tute, October 1, 2021.

Harvard Epidemiologist Censored by LinkedIn for Defending Healthcare Jobs. Brownstone Ins[tute, October 4,

2021.

LinkedIn Censors Harvard Epidemiologist Mar[ne Kulldorﬀ, Brownstone Ins[tute, August 12, 2021.

Tucker JA. Kulldorﬀ Deleted: Famed Epidemiologist and Early Opponent of Lockdowns Banned by LinkedIn,

Brownstone Ins[tute, January 28, 2022.

M$*,*/",.(#9*"?)"&6*%#*<,++(#"*$,4"607*!C!*,$B".*3"*6%*$"&<"*%#*6'"(&*!\i@CKPh*i,++(#"*N,4"60*D"+'#(+,/*

J%&B* ]&%8)1* @#* M)&(/* HIHP7* !C!* f&".* 3"* 4&%3* 6',6* 9&%8)

1*@4*0%8*6'(#B*@*5,$*f&".*4%&*L8"$:%#(#9*6'"*

<,++(#"$7*0%8*,&"*5&%#91*@*,3*)&%-,-/0*6'"*%#/0*)"&$%#*f&".*-0*!C!*4%&*-"(#9*6%%*)&%K<,++(#"1*\#*M)&(/*

Pe7*HIHP7*!C!*),8$".*6'"*T%'#$%#*j*T%'#$%#*<,++(#"*,`"&*&")%&6$*%4*-/%%.*+/%6$*(#*,*4"5*5%3"#*8#."&*

,9"*QI1*D'"&"*5"&"*#%*&")%&6".*+,$"$*,3%#9*%/."&*)"%)/"7*5'%*-"#"f6*6'"*3%$6*4&%3*6'"*<,++(#"$1*N(#+"*

6'"&"*5,$*,*9"#"&,/*<,++(#"*$'%&6,9"*,6*6'"*:3"7*@*,&98".*(#*,#*%)K".*(#*D'"*A(//*6',6*6'"*<,++(#"*$'%8/.*

#%6*-"*),8$".*4%&*%/."&*'(9'K&($B*M3"&(+,#$

1*D',6*9%6*3"*f&".7*,/6'%89'*!C!*.(.*/(`*6'"*),8$"*4%8&*.,0$*

/,6"&1*D&,9(+,//07*$%3"*%/."&*M3"&(+,#$*.(".*-"+,8$"*%4*6'($*<,++(#"*Z),8$"[1*

D'"*)&(3,&0*<(+:3*%4*+"#$%&$'()*($*#%6*3"*,#.*%6'"&$*-"(#9*+"#$%&".7*-86*6'"*)8-/(+1*M$*)%/(:+(,#$7*6%*

)&%)"&/0*$"&<"*0%8&*+%#$:68"#6$7*0%8*',<"*-%6'*6'"*&(9'6*6%*'",&*4&%3*,#.*,*.860*6%*/($6"#*6%*,*&,#9"*%4*

$+("#:$6$1*D'"*)8-/(+* ,/$%* ',$* 6',6* &(9'61*2%&*"?,3)/"7*'%5*3,#0*%4* 0%8*B#"5*6',6* N5"."#*B")6* 6'"(&*

$+'%%/$*%)"#*(#*6'"*$)&(#9*%4*HIHI*5(6'%86*,*$(#9/"*+%<(.*3%&6,/(60*,3%#9*(6$*P1R*3(//(%#*+'(/.&"#k*A%5*

3,#0*%4*0%8*B#%5*#%57*6',6*4%&*HIHIKHIHIH*N5"."#*4%+8$".*)&%6"+:%#*$6&,6"90*/".*6%*6'"*/%5"$6*"?+"$$*

3%&6,/(60*,3%#9*5"$6"&#*+%8#6&("$k

*!"#$%&$'()*.")&(<"$*-%6'*0%8*,#.*6'"*)8-/(+*4&%3*<(6,/*(#4%&3,:%#*

#"".".*6%*$,<"*/(<"$1**

!"#$%&$'()*(#"<(6,-/0*/",.$*6%*$"/4K+"#$%&(#91*N%3"*%4*30*)8-/(+*'",/6'*+%//",98"$*.(.*#%6*$)",B*8)*4%&*

4",&* %4* -"(#9* +"#$%&".7* $(/"#+".* %&* $/,#."&".7* /(B"* @* 5,$1* @* .%#W6* -/,3"*6'"31*@*5,$*,/$%*4%&+".* 6%*$"/4K

+"#$%&7*6%*,<%(.*-"(#9*)"&3,#"#6/0*-,##".*4&%3*$%+(,/*3".(,1*a2(98&"*PIb**

@*',<"*,*L8"$:%#*4%&*0%8g*C%*5"*',<"*4&"".%3*%4*$)""+'*-"+,8$"*%4*6'"*2(&$6*M3"#.3"#6*%&*.%*5"*',<"*

6'"*2(&$6*M3"#.3"#6*-"+,8$"*4&"".%3*%4*$)""+'*($*(3)%&6,#6*6%*)&"$"&<"*$%+("60*,#.*/(4"k*

J'"#*6'"*Y(//*%4*_(9'6$*5,$*5&(E"#7*M3"&(+,#$*',.*/(<".*6'&%89'*6&%8-/".*:3"$1*D'"0*.(.*#%6*8$"*6',6*,$*

,#*"?+8$"*4%&*+"#$%&$'()1*@*6'(#B*6'"0*8#."&$6%%.*6',6*4&"".%3*%4*$)""+'*($*"$)"+(,//0*(3)%&6,#6*.8&(#9*

.(=+8/6*:3"$*5'"#*.(=+8/6*."+($(%#$*38$6*-"*3,."1*@*'%)"*6',6*6'"*PPR6'*!%#9&"$$*($*O8$6*,$*5($"*,$*6'"*

*!%#9&"$$*5,$*5'"#*(6*,.%)6".*6'"*2(&$6*M3"#.3"#6*,$*),&6*%4*6'"*Y(//*%4*_(9'6$1**

D',#B*0%8*4%&*/($6"#(#91*

V,&:#*l8//.%&>*

* *

Pullman J. CDC Punishes ‘Superstar’ Scien[st For COVID Vaccine Recommenda[on The CDC Followed 4 Days

Later, The Federalist, April 28, 2021.

Kulldorﬀ M. The dangers of pausing the J&J vaccine. The Hill, April 17, 2021.

Simmons M. Sweden, Covid and ‘excess deaths’: a look at the data. The Spectator, March 10, 2023.

_________________________________________________________________________________

FIGURES

Figure 1: Tweet with Screenshot of “Unpublished” Facebook Page.

Figure 2: TwiLer, March 2021

Figure 3: TwiLer, May 2021

Figure 4: TwiLer, November 2021

Figure 5: LinkedIn, August 2021

Figure 6: LinkedIn, August 2021

Figure 7: LinkedIn, January 2022

Figure 8: LinkedIn, January 2022

012345*Z7*.1=I5F(=M*.H>:*)?>:*,5@?45*+<<?3=:*$3>C5=>1?=M*XH=3H4Q*KOKK*

012345*6O7*891L54M*NH4<;*6UM*KOKK*

“Censorship Laundering: How the U.S. Department of Homeland Security Enables the

Silencing of Dissent”

Subcommittee on Oversight, Investigations, and Accountability

Committee on Homeland Security

United States House of Representatives

May 11, 2023

Statement for the Record

Benjamin Weingarten

Investigative Journalist & Columnist

I. Introduction

Chairman Bishop, Ranking Member Ivey, and members of the Subcommittee, thank you for the

opportunity to testify today.

It is an honor and a privilege to appear before you to discuss the

Department of Homeland Security’s (DHS) enabling of the silencing of dissent.

Government’s first charge is to defend the life and limb of the governed. DHS generally, and the

Cybersecurity and Infrastructure Security Agency (CISA) specifically, have vital roles to play in

this regard. Given the criticality of their mission to protect the homeland, these agencies must be

held to exacting standards. Should they experience mission creep, in so doing wielding powers in

ways violative of the constitutional rights they are meant to secure for all Americans, it compels

good faith scrutiny. I offer today’s testimony in this spirit.

Our republic rests on the inalienable right to free speech. That right is currently under assault by

those working to consign their political foes to the digital gulag in defense of “our democracy.”

Disturbingly, the federal government itself appears to be a key culprit. Overwhelming evidence

suggests that federal agencies – led by, among others, CISA

– buoyed by senior executive

branch officials and lawmakers, colluding with Big Tech, and a coterie of often government-

coordinated and government-funded

“counter-disinformation” organizations, have imposed

nothing less than a mass public-private censorship

regime on the American people.

I appear today on my own behalf, and my views do not necessarily reflect those of the media or other

organizations with which I am affiliated.

See generally Missouri v. Biden and Special Assistant Attorney General for the Louisiana

Department of Justice D. John Sauer’s related testimony before the House Judiciary Committee’s

Weaponization Subcommittee at https://judiciary.house.gov/sites/evo-subsites/republicans-

judiciary.house.gov/files/2023-03/Sauer-Testimony.pdf; Hines v. Stamos; and

https://report.foundationforfreedomonline.com/11-9-22.html.

DHS’ Inspector General has reported that the agency’s Office of Intelligence and Analysis (I&A) was also

involved in counter-disinformation efforts during the 2020 election season. Other DHS components in the

last several years have also worked to “counter disinformation originating from foreign and domestic

sources.” [Emphasis mine] See https://www.oig.dhs.gov/sites/default/files/assets/2022-08/OIG-22-58-

Aug22.pdf#page=7. These efforts extend to other agencies including the Federal Bureau of Investigations

(FBI), Department of Justice (DOJ), and Office of the Director of National Intelligence (ODNI). Senior

executive branch officials and federal lawmakers have also publicly and privately exerted pressure on

social media companies to more aggressively police speech – at times under threat of adverse regulatory or

legislative action. See generally Missouri v. Biden; https://www.wsj.com/articles/save-the-constitution-

from-big-tech-11610387105; https://www.newsweek.com/taxpayer-dollars-must-not-fund-government-led-

censorship-regime-opinion-1792828.

https://foundationforfreedomonline.com/dhs-censorship-agency-had-strange-first-mission-banning-

speech-that-casts-doubt-on-red-mirage-blue-shift-election-events/.

https://foundationforfreedomonline.com/bidens-national-science-foundation-has-pumped-nearly-40-

million-into-social-media-censorship-grants-and-contracts/.

I use censorship herein broadly to encompass “terminating speakers’ accounts, deplatforming speakers,

temporarily suspending accounts, imposing warnings or strikes against accounts to chill future disfavored

speech, ‘shadow banning’ speakers, demonetizing content or speakers, adjusting algorithms to suppress or

de-emphasize speakers or messages, deboosting speakers or content, promoting or demoting content,

placing warning labels or explanatory notes on content, suppressing content in other users’ feeds,

promoting negative comments on disfavored content, and requiring additional click-through(s) to access

content, and other methods,” as plaintiffs in Hines v. Stamos define it. See: https://aflegal.org/wp-

content/uploads/2023/05/Doc-1-Complaint.pdf#page=9.

Authorities, led by the federal government, tell us this censorship is for our own good – that we

suffer from a pandemic of “mis-, dis-, and mal-information” (MDM);

that MDM fuels domestic

terrorism;

and therefore that America must undertake a whole-of-society effort to combat

MDM.

For its part, the censorship regime has equated MDM with Wrongthink – dissenting

opinions from its orthodoxy, and even facts

inconvenient to its agenda, on an ever-growing

number of subjective and contentious issues.

It conflates, cynically and purposefully, genuine

political difference with “extremism,” which it links to danger and violence to justify speech

policing.

In turn, the regime has surveilled the wide expanse of the digital public square to

identify such Wrongthink, and proceeded to suppress it under guise of national security and

public health.

Notably, the public-private speech police have targeted, for example, skepticism about the

integrity of mass mail-in balloting that used to be shared on a bipartisan basis and was never

linked to “domestic violent extremism;”

and skepticism about COVID-19 mitigation efforts that

often proved not only justified, but which in some instances, if more widely heard and

understood, might have saved lives and liberties. Given authorities have asserted, but not

necessarily established a clear and compelling nexus between the mere expression of such views

and widespread or dire threats of violence – and certainly not threats justifying suspension of the

First Amendment to quell them, for which this non-lawyer witness finds little precedent; and

given that authorities show little equivalent concern or zeal for suppressing a virtually limitless

array of other views that can be linked to violence – anti-cop sentiment to attacks on law

CISA has defined “Misinformation” as that which “is false, but not created or shared with the intention of

causing harm.” It has defined “Disinformation” as that which “is deliberately created to mislead, harm, or

manipulate a person, social group, organization, or country.” It has defined “Malinformation” as that which

“is based on fact, but used out of context to mislead, harm, or manipulate.” See:

https://www.cisa.gov/sites/default/files/publications/mdm-incident-response-guide_508.pdf. Setting aside

the question of who is to be the arbiter of truth in CISA’s MDM paradigm, on what grounds, and whether

and to what extent government ought to intervene accordingly, the matter of intent baked into these

definitions makes MDM a largely subjective concept.

https://www.dhs.gov/ntas/advisory/national-terrorism-advisory-system-bulletin-february-07-2022.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.268.0.pdf#pag

e=88.

See for example the Biden administration’s “National Strategy for Countering Domestic Terrorism” at

https://www.whitehouse.gov/wp-content/uploads/2021/06/National-Strategy-for-Countering-Domestic-

Terrorism.pdf.

https://nypost.com/2023/03/17/private-federal-censorship-machine-targeted-true-misinformation/.

The targeting began largely with a focus on skepticism of the integrity and outcome of the 2020 election;

it expanded to encompass derogatory views to those of federal authorities – including those ultimately

proving true and even known to be true contemporaneously – concerning virtually every aspect of COVID-

19, and particularly around mitigation efforts and their efficacy; since, federal officials have shown their

intent to expand such targeting to cover “abortion, climate-related speech, ‘gendered disinformation,’

economic policy, the financial services industry, the U.S. withdrawal from Afghanistan, the war in Ukraine,

and other[]” topics, per recent testimony from litigation counsel in Missouri v. Biden, Special Assistant

Attorney General for the Louisiana Department of Justice D. John Sauer. See:

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/2023-03/Sauer-

Testimony.pdf.

https://www.newsweek.com/biden-domestic-terror-strategy-codifies-woke-war-wrongthink-opinion-

1605341.

https://foundationforfreedomonline.com/dhs-censorship-agency-had-strange-first-mission-banning-

speech-that-casts-doubt-on-red-mirage-blue-shift-election-events/.

https://www.wsj.com/articles/heed-jimmy-carter-on-the-danger-of-mail-in-voting-11586557667.

enforcement and widespread riots,

pro-abortion sentiment to attacks on pro-life centers and

threats to judges,

environmentalist sentiment to attacks on relevant targets by eco-terrorists, etc.

– this indicates the speech-muzzling is rooted in politics, not the public good. Understood in this

light, the censorship regime’s efforts start to look like they are intended more for its own benefit,

than ours.

The regime’s systematic speech-stifling, targeting core political speech and intensifying during

recent federal election cycles, seems tantamount to a conspiracy to violate the First Amendment,

viewpoint discrimination, and running domestic election interference – ironically borne of claims

of foreign election interference.

In short, Americans have unknowingly and unwittingly been paying unelected and unaccountable

bureaucrats to, directly and by proxy, silence ourselves.

CISA has been described as a “nerve center” of these federal government-led censorship efforts.

It has served as a key facilitator of, and participant in, meetings between federal authorities and

technology companies aimed at encouraging the latter to combat purported misinformation and

disinformation. It has served as a clearinghouse for social media content flagged for censorship

by third parties – governmental and non-governmental – relaying the parties’ censorship requests

on to social media companies, and flagged perceived problematic speech for the platforms

directly.

And it has helped foster the development of the broader public-private censorship

architecture through consulting, partnering with, and networking often government-linked third-

party organizations to themselves serve as First Amendment-circumventing,

mass-surveillance

and mass-censorship clearinghouses for content flagged by, among others, government partners.

It is perhaps incalculable how many people have been bereft of their right to speak, and listen, by

way of these censorship efforts – and at what cost.

Remarkably, we would know little of such efforts were it not for a billionaire’s decision to

purchase a social media platform,

and then empower a handful of journalists to expose the

government-tied censorship efforts in which it had been implicated;

and the legal action of

vigilant state attorneys general, who, alongside the silenced, sued implicated federal authorities,

and through discovery began to untangle this twisted censorship web.

https://archive.is/SA9H1.

https://www.realclearinvestigations.com/articles/2023/01/25/frustrated_by_police_inaction_the_pro-

life_movement_takes_up_the_work_of_law_enforcement_877348.html.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.268.0.pdf#pag

e=7.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.214.1_1.pdf#p

age=278.

As Justice Clarence Thomas wrote in his concurring opinion in Biden v. Knight, a private entity violates

the First Amendment “if the government coerces or induces it to take action the government itself would

not be permitted to do, such as censor expression of a lawful viewpoint.” Further, “The government cannot

accomplish through threats of adverse government action what the Constitution prohibits it from doing

directly.” See: https://www.supremecourt.gov/opinions/20pdf/20-197_5ie6.pdf#page=11.

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2.

https://www.wsj.com/articles/elon-musk-completes-twitter-takeover-11666918031.

https://www.racket.news/p/capsule-summaries-of-all-twitter.

See Missouri v. Biden.

As its role in the censorship regime has started to come into focus, CISA has gone about

scrubbing evidence of its associated efforts;

it has reorganized related entities;

and it has

stonewalled congressional investigators

– while maintaining that, as the agency’s Director, Jen

Easterly put it in recent congressional testimony, “We don’t censor anything” or “flag anything

for social media organizations at all.”

It is hard to fully square this position with what we have learned to date. Congress can and should

help resolve this seeming dispute. At minimum, the troubling evidence suggests the national

security apparatus’s apparent interest in Americans’ speech warrants oversight, without which, if

merited, there can be no accountability and reform. This Subcommittee’s engagement, therefore,

alongside other committees with relevant jurisdiction,

is most welcome and necessary. It is also

most urgent, with the 2024 elections looming, censorship tools becoming more sophisticated and

powerful,

and the censorship regime’s ambitions only growing – alongside its footprint.

To help inform this Subcommittee’s efforts, I will briefly address how CISA came to take on a

pivotal role in this censorship regime, detail its associated actions, and offer recommendations for

further oversight.

II. How CISA Became a “Nerve Center” of America’s Censorship Regime

The plaintiffs in the landmark pending case, Missouri v. Biden, allege, and have revealed a trove

of information substantiating the claim that there is a “massive, sprawling federal ‘Censorship

Enterprise,’ which includes dozens of federal officials across at least eleven federal agencies and

components, who communicate with social-media platforms about misinformation,

disinformation, and the suppression of private speech on social media—all with the intent and

effect of pressuring social-media platforms to censor and suppress private speech that federal

officials disfavor,” in violation of the First Amendment.

The plaintiffs identify CISA

specifically as a “nerve center” of federal government-led speech policing, which began in

earnest in the run-up to the 2020 election.

Several key developments help to explain how a DHS sub-agency tasked with preventing

cyberattacks and defending physical infrastructure would come to occupy a central role in this

censorship effort. Among them are that: (i) Donald Trump won the 2016 presidential election. (ii)

His victory came to be seen by many as being enabled by (a) Social media and (b) Russian

https://foundationforfreedomonline.com/flash-report-dhs-quietly-purges-cisa-mis-dis-and-

malinformation-website-to-remove-domestic-censorship-references-2/.

https://www.racket.news/p/homeland-security-reorganizes-appearing.

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/evo-media-

document/2023-04-28-jdj-to-easterly-cisa-subpoena-cover-letter.pdf.

https://www.youtube.com/watch?v=JnbWb5ZFN8s&t=4673s.

https://judiciary.house.gov/media/press-releases/chairman-jordan-subpoenas-cdc-cisa-and-gec-

documents-and-communications.

https://foundationforfreedomonline.com/the-national-science-foundations-convergence-accelerator-track-

f-is-funding-domestic-censorship-superweapons/.

https://twitter.com/shellenberger/status/1651355243722973186?s=20.

https://twitter.com/DFRLab/status/1654500447816654849?s=20.

https://theintercept.com/2023/05/05/foreign-malign-influence-center-disinformation/.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.268.0.pdf/

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2.

interference on social media aimed at elevating Trump’s candidacy. These developments would

both escalate to a matter of national security “content moderation” – a euphemism for speech

regulation up to and including deplatforming – and fuel the creation of America’s mass public-

private censorship regime.

(iii) In partial response, in January 2017 outgoing DHS Secretary Jeh

Johnson designated election infrastructure as a critical infrastructure subsector, putting elections

ultimately under CISA’s purview.

(iv) That same year, the State Department established the

Global Engagement Center (GEC), tasked with leading federal efforts to “counter foreign state

and non-state propaganda and disinformation efforts aimed at undermining United States national

security interests.”

The FBI also established its Foreign Influence Task Force (FITF) to

“identify and counteract malign foreign influence operations targeting the United States,” with an

explicit emphasis on voting and elections.

(v) Following suit, in 2018 DHS stood up a

Countering Foreign Influence Task Force comprised of CISA’s Election Security Initiative

division, and Office of Intelligence and Analysis (I&A) staff. Its purpose, according to a recent

DHS Office of Inspector General (OIG) report, was to focus on “election infrastructure

disinformation.”

(vi) On top of this counter-disinformation mobilization, certain federal

lawmakers increasingly chided social media platforms for dithering on “content moderation,”

including but not exclusively pertaining to foreign adversaries.

(vii) Amid the government’s

growing counter-disinformation push, a constellation of sometimes state-funded non-

governmental counter-disinformation organizations grew alongside it.

This by no means exhaustive list of developments, combined with two shifts in the posture of key

players within the looming censorship regime, would create the conditions for, and leave CISA

uniquely positioned to serve as a linchpin of it. First, federal authorities and their future private-

sector partners

would train their sights increasingly on domestic Wrongthinkers over foreign

adversaries as key disinformation threat actors – or at minimum focus on the content of speech

over the country of origin of the speaker. Second, they would begin to treat words critical of

institutions as threats to those institutions.

In CISA’s case, under its first Director Chris Krebs, who served through the 2020 election cycle,

that meant targeting speech dubious of election administration and outcomes as a threat to

election infrastructure. Under his successor, infrastructure would come to comprise nearly every

For a more comprehensive treatment on both the theory and practice of our censorship regime, see

https://www.tabletmag.com/sections/news/articles/guide-understanding-hoax-century-thirteen-ways-

looking-disinformation.

See https://www.dhs.gov/news/2017/01/06/statement-secretary-johnson-designation-election-

infrastructure-critical and https://www.intelligence.senate.gov/sites/default/files/documents/os-jjohnson-

032118.pdf. In the designation, Sec. Johnson describes election infrastructure as “storage facilities, polling

places, and centralized vote tabulations locations used to support the election process, and information and

communications technology to include voter registration databases, voting machines, and other systems to

manage the election process and report and display results on behalf of state and local governments.”

https://www.cisa.gov/topics/election-security.

https://www.congress.gov/bill/114th-congress/senate-bill/2943/text.

https://www.fbi.gov/news/press-releases/the-fbi-launches-a-combating-foreign-influence-webpage.

https://www.oig.dhs.gov/sites/default/files/assets/2022-08/OIG-22-58-Aug22.pdf#page=7.

See https://twitter.com/mtaibbi/status/1610372352872783872?s=20 and

https://www.wsj.com/articles/save-the-constitution-from-big-tech-11610387105.

For an extensive accounting of the theory and practice behind this burgeoning disinformation industrial

complex, see https://www.tabletmag.com/sections/news/articles/guide-understanding-hoax-century-

thirteen-ways-looking-disinformation and https://judiciary.house.gov/sites/evo-subsites/republicans-

judiciary.house.gov/files/evo-media-document/shellenberger-testimony.pdf#page=8.

https://rumble.com/v1gx8h7-dhss-foreign-to-domestic-disinformation-switcheroo.html/.

reporting on the contents of Hunter Biden’s laptop – indicating Biden family

influence peddling – weeks from the 2020 presidential election, on false grounds

that it was the product of such a “hack-and-leak.”

o It has been reported that government warnings about “hack-and-leaks” led

platforms to change their terms of service in the run-up to the 2020 election to

suppress related content. In CISA-convened industry meetings, content

moderation policies are a regular topic, and CISA regularly communicates with

social media platforms about such policies.

• “Switchboard” reports of purported misinformation and disinformation from state and

local authorities, among others, beginning in 2018 and expanding through the 2020

election. Switchboarding entails receiving and then forwarding reports of offending

content to social media platforms for censorship. Officials did so without assessing

whether the content came from foreign or domestic speakers. Among other notable points

about these efforts:

o CISA staff switchboarded misinformation reports, for example, flagging tweets

for censorship alleging election fraud, that ballots were not counted, and mail-in

voting was implemented to benefit Democrats. One such report forwarded by a

CISA official to Twitter called for “swift removal of…posts and continued

monitoring of the user’s account” because said user had “claimed…that mail-in

voting is insecure,” and that “conspiracy theories about election fraud are hard to

discount.” Twitter reported back to CISA it had taken action pursuant to its

policy on Civic Integrity.

o Staffers also switchboarded misinformation reports flagging obviously satirical

social media accounts for censorship, including one Colorado Twitter account

with 56 followers “UnOfficialCOgov.” The user’s biographical information read:

“dm us your weed store location (hoes be mad, but this is a parody account).”

o A CISA switchboard tracking spreadsheet from 2020 suggests that in certain

instances, officials from both CISA and DHS I&A were the originators of

flagged content ultimately conveyed by CISA staff to social media companies for

review.

o Switchboarding efforts at times would even touch on private postings on social

media platforms.

o Social media companies would often report that they would “escalate” CISA-

switchboarded requests and revert to CISA once addressed.

• Brief state officials about content CISA considers misinformation, which those officials

often then flag for social media platforms for censorship; fact-check “misinformation”

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=274.

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/2023-03/Sauer-

Testimony.pdf#page=18.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.214.1_1.pdf#p

age=267.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.209.15.pdf#pa

ge=11.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.214.35.pdf.

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=271

Ibid.

significant institution, and now, even our brains. Director Easterly would argue that the American

mind – “our cognitive infrastructure” – is “the most critical infrastructure,” obligating authorities

to “protect” such infrastructure.

One way to do so would be through controlling the information

space by suppressing disfavored narratives – hence the efforts she would take to “grow and

strengthen my misinformation and disinformation team.”

Accordingly, CISA would come to equate first the American public’s skeptical tweets on subjects

like mail-in voting with attacks on election infrastructure, and later a growing list of dissident

views on other issues as threats to relevant infrastructure, and arrogate unto itself the power to

neutralize the threats through helping orchestrate a public-private censorship regime.

III. CISA’s Leadership in the Censorship Regime

In fact, CISA would not only help orchestrate widespread censorship efforts, but would actively

participate in them. During the 2020 election, and in some instances continuing and expanding

thereafter, findings from Missouri v. Biden and additional supporting evidence demonstrate that

CISA officials contribute to censorship efforts directly and by proxy.

CISA’s Direct Censorship-Related Efforts

Among other direct actions CISA officials have taken with respect to countering MDM,

personnel:

• Convene and coordinate meetings between national security and law enforcement

agencies, and technology companies – including not just social media platforms

Facebook/Meta, Google, Twitter, and Reddit, but also Microsoft, Verizon Media,

Pinterest, LinkedIn, and Wikimedia Foundation

– aimed at combating purported

misinformation and disinformation. These meetings occur more frequently in the run-up

to elections.

CISA is party to at least five sets of recurring confabs with social media

platforms touching on MDM and/or policing of speech on said platforms, separate and

apart from the many bilateral such meetings CISA hosts.

o In 2020 meetings with social media companies, CISA and other officials warned

of potential foreign “hack-and-leak” operations to come during the election.

Major social media companies would proceed to censor the New York Post’s

https://thehill.com/policy/cybersecurity/580990-cyber-agency-beefing-up-disinformation-

misinformation-team/.

Ibid.

The Biden administration in fact would incorporate this view into its first-of-its-kind National Strategy

for Countering Domestic Terrorism, in calling for government to “accelerat[e] work to contend with an

information environment that challenges healthy democratic discourse” as part of its effort to “confront

long-term contributors to domestic terrorism.” See: https://www.whitehouse.gov/wp-

content/uploads/2021/06/National-Strategy-for-Countering-Domestic-Terrorism.pdf#page=29/

Since much of the available record details CISA activities prior to the 2022 midterm elections, it is not

entirely clear in some instances whether certain activities persist. This only further underscores the need for

congressional oversight.

See https://twitter.com/MSFTIssues/status/1293623288262987777?s=20. While much of this testimony

focuses on the actions of social media platforms, the inclusion of other technology companies in

conversations with U.S. government agencies about MDM suggests oversight need be done on the actions

of these companies in conjunction with the federal government as well.

Officials from CISA, DHS’s I&A division, ODNI, FITF, and other agencies attend the meetings. See:

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=218.

reports for social media platforms;

and publish “debunks of social-media narratives,

knowing...platforms will use this information to censor,” per litigation counsel in

Missouri v. Biden.

• Coordinate with public and private sector partners, including social media companies “on

a variety of projects to build resilience against malicious information activities,” as well

as supporting “private sector partners’ COVID-19 response efforts via regular reporting

and analysis of key pandemic-related MDM trends.”

This is part and parcel of what

CISA’s Cybersecurity Advisory Committee has described as a “burgeoning MDM effort”

that includes “directly engaging with social media companies to flag MDM.”

The coordination referenced above comes from a bulletin CISA posted on its website detailing

the work of its MDM team – the successor to its Countering Foreign Influence Task Force. The

creation of that team formally codified the transition that had already taken place during the 2020

election cycle, from a focus on foreign to domestic speech.

In February 2023, CISA pulled

down that site, redirecting viewers to a “Foreign Influence Operations and Disinformation” page

that makes no mention of domestic actors. One can only speculate as to why CISA made this

change.

CISA would also expand its focus to encompass not just MDM around elections, or COVID-19

vaccine efficacy under banner of defending public health infrastructure,

but “all types of

disinformation, to be responsive to current events,” according to an official quoted in an August

2022 DHS OIG Report.

Evidence collected in Missouri v. Biden indicates CISA has been

involved in combatting “misinformation” with respect to the ongoing Russo-Ukrainian War,

and on an initiative in conjunction with the Treasury Department to address MDM regarding the

financial services industry.

In a January 2023 deposition taken in connection with Missouri v. Biden, the chief of CISA’s

MDM Team, Brian Scully, asserted that his team had a mandate that was almost limitless, in

pursuing MDM that could affect “critical infrastructure in a number of ways,” including causing

“reputational risk [that] could come about if the integrity or the public confidence in a particular

sector was critical to that sector’s functioning.”

It is also possible CISA’s efforts have extended beyond social media companies, and perhaps the

other technology companies with which it and other federal agencies have regularly met in

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=269.

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/2023-03/Sauer-

Testimony.pdf#page=19.

https://web.archive.org/web/20211231181148/https://www.cisa.gov/mdm.

https://s3.documentcloud.org/documents/23175380/dhs-cybersecurity-disinformation-meeting-

minutes.pdf.

https://web.archive.org/web/20211231181148/https://www.cisa.gov/mdm.

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/2023-03/Sauer-

Testimony.pdf#page=25.

https://www.oig.dhs.gov/sites/default/files/assets/2022-08/OIG-22-58-Aug22.pdf#page=9.

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=280.

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=283.

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=282.

Election Integrity Partnership (EIP), and a successor organization folding in additional partners,

the Virality Project (VP).

CIS is a nonprofit that collects and forwards reports of disinformation from state and local

government officials to social media platforms, and which continued to do so during the 2022

election cycle.

As CISA’s switchboarding activities became too labor-intensive for it, CISA

would direct election officials to report content to be flagged for social media platforms to CIS.

CISA would also help connect CIS, and various election official groups, with EIP.

EIP is a non-governmental “anti-disinformation” consortium that was conceived by and created in

consultation with CISA officials in the run-up to the 2020 election. Its stated purpose was to fill

the “critical gap” created by the fact no federal agency “has a focus on, or authority regarding,

election misinformation originating from domestic sources within the United States.”

That lack

of “authority” may have included both an inability for government agencies, to access social

media platform data – as EIP did – as well as “very real First Amendment questions” regarding

EIP’s work, as a key player in the consortium, Renee DiResta, would acknowledge.

EIP’s four partner organizations, “leading institutions focused on understanding misinformation

and disinformation in the social media landscape,”

sharing pervasive ties to the federal

government, include the:

• Stanford Internet Observatory (SIO) – Founded in June 2019 by former Facebook chief

security officer Alex Stamos, several of SIO’s students came up with the idea for EIP

while serving as CISA interns.

Stamos serves on CISA’s Cybersecurity Advisory

Committee. He and Chris Krebs, CISA’s director through the 2020 election, formed a

consultancy in late 2020 called the Krebs/Stamos Group. CISA’s top election official

through 2020, Matt Masterson, who was involved in the establishment of EIP, joined SIO

as a fellow after leaving CISA in January 2021. SIO’s Research Manager, the

aforementioned DiResta, served as a Subject Matter Expert for CISA’s Cybersecurity

Advisory Committee’s since-abolished MDM Subcommittee.

• University of Washington’s Center for an Informed Public – Founded in December 2019,

its cofounder Kate Starbird served as the chairperson of the since-abolished MDM

Subcommittee – serving incidentally alongside former Twitter executive Vijaya Gadde, a

leader of its censorship efforts prior to her ouster under new owner Elon Musk.

UW’s

Center, along with SIO, would share in a $3 million National Science Foundation grant

awarded in August 2021 to “study ways to apply collaborative, rapid-response research to

mitigate online disinformation.”

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/2023-03/Sauer-

Testimony.pdf#page=19.

https://stacks.stanford.edu/file/druid:tr171zs0069/EIP-Final-Report.pdf#page=9.

https://ago.mo.gov/docs/default-source/press-releases/212-3-proposed-findings-of-

fact.pdf?sfvrsn=739f8cbf_2#page=265.

https://stacks.stanford.edu/file/druid:tr171zs0069/EIP-Final-Report.pdf#page=20.

Ibid.

https://www.racket.news/p/homeland-security-reorganizes-

appearing?r=5mz1&utm_campaign=post&utm_medium=web.

https://www.politico.com/news/magazine/2020/10/28/twitter-vijaya-gadde-free-speech-policies-

technology-social-media-429221.

https://www.cip.uw.edu/2021/08/15/national-science-foundation-uw-cip-misinformation-rapid-response-

research/.

connection with combatting MDM. A June 2022 report from CISA’s Cybsersecurity Advisory

Committee Subcommittee on Protecting Critical Infrastructure from Misinformation and

Disinformation (“MDM Subcommittee”) suggests that CISA should approach the mis- and dis-

information problem “with the entire information ecosystem in view.” This means focusing not

just on social media platforms, but “mainstream media, cable news, hyper partisan media, talk

radio, and other online resources.”

CISA would, as with its MDM webpage, scrap its MDM

Subcommittee, as first publicized in a late 2022 summary of an advisory board meeting.

As significant as CISA’s MDM efforts have been, DHS’ counter-disinformation operations

spread far beyond the sub-agency. According to the aforementioned August 2022 DHS OIG

report, numerous components inside DHS have in recent years been targeting MDM foreign and

domestic. What’s more, the report details that DHS planned to target “inaccurate information” on

myriad topics including “the origins of the COVID-19 pandemic and the efficacy of COVID-19

vaccines, racial justice, U.S. withdrawal from Afghanistan, and the nature of U.S. support to

Ukraine.”

Corroborating the OIG Report, one document revealed in connection with congressional inquiries

into DHS’ stunted Disinformation Governance Board (DGB) indicated that myriad “DHS

components are already engaged in countering disinformation,” alongside “excellent work being

done by interagency partners, the private sector, and academia—particularly concerning

identifying and analyzing disinformation,” which “DHS should leverage.”

A subsequent

memorandum would indicate that the DGB would “support and coordinate…MDM work with

other departments and agencies, the private sector, and non-government actors.” The purpose of

creating the DGB, in other words, was not so much to establish a “Ministry of Truth,” but, as

plaintiffs in Missouri v. Biden aptly describe it, “to impose a bureaucratic structure on the

enormous censorship activities already occurring involving dozens of federal officials and many

federal agencies” – that is, to oversee many such ministries.

CISA’s Proxy Censorship-Related Efforts

Not all of these ministries are to be found within the federal government. CISA officials

coordinate and partner with non-governmental entities who both mass-surveil social media

content for purported MDM, and serve as clearinghouses for receipt of flagged content, which

they then relay to social media platforms for censorship – in an apparent bid to circumvent the

First Amendment via cutout.

CISA has primarily partnered with three non-governmental entities, beginning during the 2020

election cycle, to facilitate the flow of problematic content for potential censorship to social

media platforms: The Center for Internet Security (CIS) and its CISA-funded Election

Infrastructure – Information Sharing & Analysis Center (EI-ISAC); and two consortia: The

https://www.cisa.gov/sites/default/files/publications/June%202022%20CSAC%20Recommendations%20%

E2%80%93%20MDM_0.pdf#page=2.

https://www.racket.news/p/homeland-security-reorganizes-appearing?utm_source=post-email-

title&publication_id=1042&post_id=110070633&isFreemail=true&utm_medium=email.

https://theintercept.com/2022/10/31/social-media-disinformation-dhs/.

https://www.hawley.senate.gov/sites/default/files/2022-06/2022-06-

07%20DOCS%20ONLY%20CEG%20JH%20to%20DHS%20%28Disinformation%20Governance%20Bo

ard%29%5B1%5D.pdf.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.268.0.pdf#pag

e=106.

• The Atlantic Council’s Digital Forensics Research Lab – Founded in 2016, it receives

substantial taxpayer funding from a variety of agencies.

• Graphika – Founded in 2013, it reportedly has historically received funding from

DARPA and the Defense Department’s Minerva Initiative.

Collectively, these groups sought to “fill the gap” by creating a mass-surveillance and censorship-

flagging platform aimed at “content intended to suppress voting, reduce participation, confuse

voters as to election processes, or delegitimize election results without evidence.”

In practice,

this meant targeting for suppression speech dubious of an unprecedented election given the

sweeping, pandemic-driven changes made to the voting system that cycle, whereby the razor-thin

final results in key states did not materialize for days.

EIP did so in part through lobbying social

media platforms to adopt more aggressive content moderation policies around election rhetoric,

and flagging relevant content including entire narratives via “tickets” for suppression by social

media platforms under their often EIP-influenced terms. EIP analysts – some 120 of whom

worked on the project in the waning days of the 2020 election – both identified content for

flagging via tickets, and incorporated requests from “trusted external stakeholders.”

It lists three

such governmental stakeholders: CISA,

CISA-backed EI-ISAC, and the State Department’s

GEC. EIP in fact connected “government partners” with “platform partners” – understood to be

the social media companies – to enable the former to debunk flagged content directly for the

latter.

Some raw numbers concerning EIP’s efforts during the 2020 election cycle alone illustrate the

size and scope of its effort. EIP:

• Collected 859 million tweets for “misinformation” analysis.

• Flagged for Twitter tweets shared 22 million times ultimately labeled “misinformation,” a

disproportionate percentage of which were dinged for “delegitimization,”

which Twitter

adopted as a standard for suppression.

https://www.google.com/url?q=https://foundationforfreedomonline.com/dhs-censorship-agency-had-

strange-first-mission-banning-speech-that-casts-doubt-on-red-mirage-blue-shift-election-

events/&source=gmail&ust=1683295935068000&usg=AOvVaw0KqPf31iTuGK3TncCZZwjc.

https://foundationforfreedomonline.com/dhs-censorship-agency-had-strange-first-mission-banning-

speech-that-casts-doubt-on-red-mirage-blue-shift-election-events/.

https://stacks.stanford.edu/file/druid:tr171zs0069/EIP-Final-Report.pdf#page=23.

Former Trump State Department Cyber official Mike Benz would observe that CISA, “tasked with

election security,” via EIP “also gained the power to censor any questions about election security.” See:

https://foundationforfreedomonline.com/dhs-censorship-agency-had-strange-first-mission-banning-speech-

that-casts-doubt-on-red-mirage-blue-shift-election-events/.

https://judiciary.house.gov/sites/evo-subsites/republicans-judiciary.house.gov/files/2023-03/Sauer-

Testimony.pdf#page=19.

It is worth noting that CISA and EIP’s relationship went both ways. At times, evidence suggests, CISA

would forward reports of misinformation received directly from EIP on to social media platforms for their

review.

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.209.2.pdf#pag

e=47.

EIP cites as an example of delegitimization “Claims of fraud or malfeasance with inaccurate or missing

evidence.” See:

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.209.2.pdf#pag

e=25.

Alex Stamos has challenged this characterization in terms of tweets EIP ensnared. The competing

arguments can be seen here: https://twitter.com/MikeBenzCyber/status/1644110224150736897?s=20.

• Influenced platforms to take action on 35% of all URLs flagged – 21% slapped with a

warning label where content remained visible, 13% removed, and 1% “soft-blocked”

with a warning one would have to bypass to view the content.

• Pushed platforms to target dozens of “misinformation narratives” for throttling.

• Impacted hundreds of millions of posts and videos across major social media platforms

via the terms of service policy changes for which EIP lobbied. EIP members openly

boasted that technology companies would never have modified their terms accordingly

without EIP’s insistence and “huge regulatory pressure” from government.

Further demonstrating the interconnection between EIP and CISA, the group featured former

CISA Director Chris Krebs at the launch seminar associated with the report in which it divulged

some of these figures.

Of note, EIP coded less than one percent of its tickets for having an element of foreign

interference. EIP characterized all 21 of the “most prominent repeat spreaders” of election

integrity “misinformation” on Twitter as “conservative or right-wing.”

Of the civil society

groups that submitted tickets to the EIP, many had a left-leaning bent – including the DNC

itself.

None appear to have been right-leaning.

Mike Benz, a former State Department Cyber official during the Trump administration, has found

that many principals in EIP leadership were heavily invested in the idea that Russia interfered in

the 2016 presidential election, to President Trump’s benefit, and that they or the organizations

with which they were affiliated were critical generally of Trump and Western populist

movements. In an associated report, he concludes that given the backgrounds of EIP’s principals,

when originally conceived in June 2020 it should have been understood to be “a partisan,

powerfully connected political network, panicked that Americans might push back on the use of

mail-in ballots months in the future,” convened “to stop that pushback from happening by

unleashing censorship of the Internet on a scale never before seen in American history.”

Though the EIP’s efforts would re-emerge in the 2022 election, in the interim it also launched a

successor effort called the Virality Project, targeting MDM spreading in relation to COVID-19,

such as “narratives that questioned the safety, distribution, and effectiveness of the vaccines.” Its

leaders, including Stamos communicated with CISA officials about their efforts, as they did

during the original EIP operation. DiResta would serve as principal Executive Editor of its final

April 2022 report, and contributors included herself alongside Kate Starbird and Matt Masterson.

Several current and former CISA interns are also listed as “researchers and analysts” who

monitored social media platforms in connection with the project.

The VP’s stakeholders included federal health agencies, working alongside social media

platforms to combat, for example, vaccine-related “misinformation.” All told, the Virality Project

tracked content with 6.7 million engagements on social media per week – or over 200 million

during the seven months over which the project transpired.

https://stacks.stanford.edu/file/druid:tr171zs0069/EIP-Final-Report.pdf#page=57.

https://foundationforfreedomonline.com/dhs-censorship-agency-had-strange-first-mission-banning-

speech-that-casts-doubt-on-red-mirage-blue-shift-election-events/.

https://stacks.stanford.edu/file/druid:tr171zs0069/EIP-Final-Report.pdf#page=205.

https://oversight.house.gov/wp-content/uploads/2022/11/DHS-Censorship-Letter-11022022.pdf/.

https://www.google.com/url?q=https://foundationforfreedomonline.com/dhs-censorship-agency-had-

strange-first-mission-banning-speech-that-casts-doubt-on-red-mirage-blue-shift-election-

events/&source=gmail&ust=1683295935068000&usg=AOvVaw0KqPf31iTuGK3TncCZZwjc.

Much of what the VP cast as “misinformation” included true facts to the extent they portrayed

narratives with which the project’s leaders – and certainly its government partners – disapproved

of, from reports of vaccine injuries to discussion of “breakthrough” cases and “natural

immunity,” to discussion of potential then-hypothetical vaccine mandates. VP particularly

targeted the speech of “health freedom” groups, and like EIP, overwhelmingly targeted right-

leaning figures.

IV. Conclusion

We may find much of the speech that social media platforms have suppressed in recent years

under government coercion, cajoling, and/or collusion to be wrongheaded or objectionable. But

infinitely more wrongheaded, objectionable, and indeed dangerous for a free society than the

proliferation of “bad ideas” is perhaps the worst idea of all: That government should be the arbiter

of what we are allowed to think and speak.

The notion that to ensure the health and safety of the country, the public and private sectors must

work together to silence those who express unauthorized opinions, that such opinions are to be

treated as threats to an infinitely flexible definition of “critical infrastructure,” and those who hold

them as actual or would-be domestic terrorists, is the stuff of tyranny.

That the state itself has treated as dangerous MDM that which ultimately often has become settled

science – indicating government officials and their partners en masse should have been

deplatformed themselves by their own standards – illustrates the folly of this project.

To turn over to the state and its private sector ancillaries a monopoly on narrative would

ultimately give these partners a monopoly on power, reducing us from citizens with agency to

hapless subjects.

We are a free people capable of evaluating information and ideas for ourselves to discern fact

from fiction, and separate good ideas from bad.

Historically, we would have held in utter contempt authorities who would suggest we are

incapable of thinking for ourselves, and that for our own benefit, since the authorities know best,

that they will do the thinking for us – while silencing those who dare dissent.

No American should stand for it today.

If, as the foregoing suggests, CISA, and perhaps other DHS components, have played an integral

role in imposing a mass public-private censorship regime on the American people, it is incumbent

upon this and other relevant congressional bodies to get to the bottom of it.

This Subcommittee can help develop a comprehensive picture of the “public” side of the regime

within DHS by using its oversight powers to, over a timeline beginning from CISA’s inception in

November 2018, pursue the following questions:

• Which offices and personnel within CISA

are or have been engaged in social media

censorship efforts, or related efforts to impact any other part of the “information

ecosystem” as CISA has defined it?

Plaintiffs in Missouri v. Biden assert that “On information and belief, CISA maintains a number of

• Which other DHS agencies, and/or federal, state, county, and local government entities

have CISA coordinated with in connection with social media censorship efforts, or

related efforts to impact any other part of the “information ecosystem” as CISA has

defined it?

• Which entities within DHS, independent of CISA, if any,

engaged in social media

censorship efforts, or related efforts to impact any other part of the “information

ecosystem” as CISA has defined it?

• What specific policies and practices has each DHS entity developed and undertaken in

connection with each respective censorship effort?

• Is there a comprehensive list of all communications, technology, media, educational, non-

profit, and any other non-governmental agency with which DHS broadly engaged in

fostering its censorship efforts?

• What level of federal funding has each DHS entity received to carry out such censorship

efforts?

• What level of federal funding has each private sector entity with which DHS interacted in

its censorship efforts received?

• What have been the qualitative and quantitative impacts of such censorship efforts during

periods leading up to and immediately following the 2020 and 2022 elections?

• What censorship efforts are CISA and/or any other DHS agencies engaging in at present,

and/or planning for in anticipation of the 2024 elections?

Only with full transparency can Congress and the American people understand the full size and

scope of this portion of the censorship regime and determine what if anything Congress ought to

do about it – be it in terms of withholding funding, curtailing operations, and/or holding

malefactors to account.

If indeed we have had a mass public-private censorship regime foisted upon us, defunding,

dismantling, and deterring government officials from participating in, or funding such an

apparatus ever again, would seem to be of the utmost importance.

Congress should be commended for efforts already underway to prevent such behavior.

I hope it

will do more.

Once again, thank you for the honor of appearing before you to discuss these important issues,

and I would be happy to answer any questions from the Committee.

task forces, working groups, and similar organizations as joint government-private enterprises,

which provide avenues for government officials to push for censorship of disfavored viewpoints

and speakers online.” See:

https://storage.courtlistener.com/recap/gov.uscourts.lawd.189520/gov.uscourts.lawd.189520.268.0.pdf#pag

e-87.

An August 10, 2022 DHS Office of Inspector General report indicates that several other DHS

components have engaged over “the last three years to counter disinformation originating from foreign and

domestic sources.” [Emphasis mine] See https://www.oig.dhs.gov/sites/default/files/assets/2022-08/OIG-

22-58-Aug22.pdf#page=10.

https://www.congress.gov/bill/118th-congress/house-bill/140/text.

Congressional Testimony

Statement of Dr. Cynthia Miller-Idriss

Professor, School of Public Affairs and School of Education, American University

Founding Director, Polarization and Extremism Research and Innovation Lab (PERIL)

Washington, D.C.

Hearing

“Censorship Laundering:

How the U.S. Department of Homeland Security Enables the Silencing of Dissent”

The House Committee on Homeland Security

Subcommittee on Oversight, Investigations, and Accountability

U.S. House of Representatives

310 Cannon House Office Building

Washington, D.C. 20515

Thursday, May 11, 2023, 2:00 PM ET

Chairman Green, Ranking Member Thompson, and Members of the Committee: I would

like to thank you for your service to our country and for calling attention to the critical issue of

disinformation. My name is Cynthia Miller-Idriss, and I am a Professor in the Department of

Justice, Law, and Criminology and in the School of Education at the American University in

Washington, D.C., where I also direct the Polarization and Extremism Research and Innovation

Lab (PERIL)— an applied research lab in the School of Public Affairs. I have been studying

education-based solutions to the prevention of violent extremism, including through early

prevention related to disinformation and propaganda– for over twenty years. I want to

acknowledge the support of my research team at PERIL, whose assistance was invaluable in

preparing my testimony today.

The Polarization and Extremism Research and Innovation Lab, PERIL, develops

evidence-based initiatives- such as, short-form videos, trainings and train-the-trainer programs,

research studies, community toolkits and guides- to build social cohesion, reverse political

polarization, and prevent violent extremism. Utilizing a public health framework and multi-

disciplinary, pre-preventative approaches, we design, test, and scale up evidence-based tools and

intervention strategies to help people recognize and reject harmful online and offline content,

propaganda, supremacist ideologies, conspiracy theories, misinformation, and disinformation

while safeguarding their freedom of speech. As widely recognized experts and leaders in the

field of preventing extremism and radicalization, we have created effective, evidence-based

resources to inoculate against propaganda and extremist content, as well as empower individuals

to intervene and interrupt early radicalization and keep their loved ones safe from online

manipulation– all as an alternative to security-based approaches that rely on surveillance,

monitoring, censorship, or banning.

PERIL’s work focuses specifically on equipping people with tools to recognize online

manipulation in order to protect themselves and their loved ones from disinformation that seeks

to harm them (see below for definitions of disinformation and related terms). We do not teach

people what to think; our work is nonpartisan and rooted in evidence. Our focus is on responding

to community needs and on providing resources to help people understand the kinds of

persuasive techniques that bad actors often use to manipulate others. For example, foreign

influence operations, domestic and international extremist and terrorist groups, and scammers

seeking a profit will often use rhetorical strategies, propaganda, and emotional tactics that are

designed and used to convince others to believe, think, or act in a specific manner. These

persuasive techniques manipulate observers for the purposes of grooming, recruiting, and

building support for violent ideologies, tactics, strategies, or actions. Our research has

demonstrated with consistent statistical significance that people can learn to recognize persuasive

and manipulative tactics in order to make more informed choices in their lives, especially online.

SCOPE AND SCALE: COMMUNITIES IN NEED

The national crisis facing communities across the country is all too evident. Over the past

three years, PERIL has fielded a constant stream of emails and calls from individuals and

communities across the country– all asking for help confronting the impacts of disinformation

and propaganda in their lives. In Michigan, a grandfather and military veteran wrote to ask what

he could do about his grandson, who had joined an armed militia. In Texas, faith leaders asked

for ways to support pastors whose congregations were torn apart by partisan polarization and

conspiracy theories. In Washington State, a local government needed training for city employees

to prevent polarization and reject online manipulation. In Vermont, a local entrepreneur asked if

the school system could do more to ensure that his future employees– most of whom he hired

straight from the local high school– would stop espousing so much propaganda and conspiracy

theories, which had become a problem for his business. A local mom wanted help with her

middle school son, who during the pandemic had consumed so much online misogyny that he

said he did not need to respect her authority as a parent, because she is a woman.

These stories illustrate what research evidence has also demonstrated: we face a national

crisis rooted in the rampant circulation of propaganda, dis/mis and malinformation, and other

harmful online content. American communities are coming to us because they feel threatened by

online disinformation. Some fall prey to hostile foreign influence operations by people who try

to manipulate Americans for profit or to disrupt our democratic process. People give their bank

information to scammers pretending to be from the IRS. Teenagers share intimate details of their

lives with people online who they think are friends their own age, but who are not. Others come

to believe propaganda and disinformation that lures them into what they think is heroic action to

save their racial or ethnic group after going down rabbit holes of antisemitic conspiracies about

demographic change and a supposed orchestrated replacement of white people.

The data on this is clear. The pace, scope, and scale of violent extremism have probably

increased and are escalating rapidly. The Anti-Defamation League reports that white supremacist

propaganda efforts are at the highest level they have ever recorded, jumping 38% above 2021

levels to 6,751 reported cases in 2022.

These incidents include distribution of racist, antisemitic,

and anti-LGBTQIA+ fliers, graffiti and posters, stickers, banners, and laser projections that have

heavily targeted houses of worship and other community institutions.

The repercussions of so much circulation of propaganda, conspiracy theories, and

disinformation are abundantly clear. Between 2013 and 2021, the number of open domestic

terrorism-related cases in the U.S. jumped 357% to 9,049 cases, with the most violent incidents

being committed by racially or ethnically-motivated violent extremists during the same years.

Of the 444 people killed by extremism in the U.S. between 2013 and 2022, the significant

majority of deaths were at the hands of right-wing extremists (335 deaths, or 75%).

Of those

killed by right-wing extremists in 2021, 73% were affiliated with white supremacy, 5% with

incel/toxic masculinity extremism, and 17% with anti-government extremism.

The 2022 racist

shooting that killed 10 people in a grocery store in a predominantly Black neighborhood in

Buffalo, motivated by the false Great Replacement conspiracy theory, is just one tragic recent

example.

Non-lethal attacks have also risen significantly. More than 50 bomb threats were made to

HBCUs (Historically Black Colleges and Universities) and predominantly Black churches in

2022. And the problem goes well beyond white supremacist extremism.

Antisemitism,

conspiracy theories, anti-LGBTQ+ hate, and misogynistic content has spiked across online

platforms. Before he was banned from social media platforms in mid 2022, violent and deeply

misogynistic videos from one content creator were viewed 12 billion times on TikTok alone.

Violent outcomes often show a toxic mix of ideological hatred. Just this week, eight people lost

their lives at a Texas shopping mall at the hands of a man with a swastika tattoo who had posted

both violent misogynistic and neo-Nazi content online.

In October 2020, the U.S. Department of Homeland Security under President Trump

issued a threat assessment report declaring domestic violent extremism in general and white

supremacist extremists (WSEs) in particular the ‘most persistent and lethal threat in the

Homeland.’

The Biden administration issued a similar assessment in spring 2021,

followed by

the first ever national strategy to counter domestic terrorism, noting the rising threat from white

supremacist extremism and anti-government and unlawful militias that threaten civilians, elected

officials, and democratic institutions.

Much of this violence is motivated by disinformation,

propaganda, and conspiracy theories. According to the Global Terrorism Database, terrorist

attacks motivated by conspiracy theory extremists were responsible for 119 attacks in 2020– a

jump from 6 attacks the year before– in Australia, New Zealand, the United States, Canada,

United Kingdom and Germany.

Meanwhile, hate crimes in the U.S. are at the highest level in

decades,

despite persistent underreporting. In sum, the U.S. and our allies have seen rising

violent extremism and hate-fueled and political violence fueled by antisemitism, conspiracy

theories, propaganda, disinformation, and other harmful online content as a pattern of violence

that has been escalating for years.

WHAT WORKS TO PREVENT AND COUNTER DISINFORMATION

The good news is there is a growing body of evidence about what works to equip the

public with tools that shore up their capacity to intervene in pathways to violent extremism,

while protecting their right to free speech and reducing the need for security-based approaches.

We have found that it only takes 7-12 minutes of reading one of our intervention guides for its

audience to be significantly better informed about harmful online content and the risks of

radicalization to violence; to feel more empowered and confident about intervening; to build

their own capability to intervene; and to know where to get more help.* This is the case across

our research with parents and caregivers, including grandparents, uncles, and cousins; with

educators and youth mentors; with local governments and small businesses, and more. For

example, in just 12 minutes of reading one of our intervention tools, 85% of our participants

understood the process by which youth become radicalized, and 83% felt that they knew where

to get help if they suspect a young person to be engaging in extremist ideas.

Importantly, across our work, we found that both prior to and after reading our parents

and caregivers guide, Democrats and Republicans did not significantly differ in their knowledge

of extremism. Republicans scored significantly better (5% better) than Democrats did in terms of

knowledge of extremism after having read the guide, and members of both political parties

reported being satisfied with the guide’s contents and equally willing to intervene with a young

person they suspect is coming into contact with radicalizing content. We have also found that

education alone doesn’t solve our problem of disinformation. Our research has shown that higher

levels of education do not necessarily mean people have the skills to consistently recognize

harmful manipulation tactics online. On the contrary- we found that parents with higher levels of

education were overconfident in their ability to help children distinguish trustworthy and

untrustworthy news sources. After reading our guide, their confidence went down as they

realized how tricky online disinformation and harmful content can be.

There is also strong emerging evidence that even short interventions can have a lasting

impact on local communities. We are currently studying a group of 1500 parents and caregivers

in three-month intervals for a full year after reading our intervention tool. Three months after

reading our guide for parents and caregivers, over 11% (135 individuals) of respondents said that

after the intervention, they either joined or created a group that discusses issues of youth

radicalization and extremism. Six percent of our participants, or about 75 people, told us that

within the three months after reading our guide, they used what they learned to take direct action

to prevent youth from radicalizing further or being recruited through additional online

manipulation. Overall, three months after reading our guide, parents and caregiver retained the

vast majority of the knowledge and skills they had learned. Seventy five percent of participants

reported understanding the process by which youth become radicalized online - a 23% increase

from the initial survey - and 70% felt prepared to talk with youth about online extremism - only a

5% drop from the initial survey. Over a third of participants told us they had shared or used the

information with their biological children, while nearly 13% shared it with other young people in

their family, including grandchildren, nephews and nieces, and cousins.

Taken together, our evidence shows that it is possible to provide communities with tools

to be safer online. Parents, grandparents, teachers, coaches, mental health professionals, and

others deserve help confronting an unprecedented amount of harmful online content and being

more confident and capable to keep their families safe and protected from harmful online

content. All communities need information and tangible action steps for how to help their loved

ones resist manipulative rhetoric, propaganda, conspiracy theories, and disinformation they are

exposed to online and offline in ways that help them make better choices while avoiding

censorship, surveillance, monitoring, or other security-based approaches.

ADOPTING A PUBLIC HEALTH APPROACH

PERIL advocates for a holistic public-health mode of prevention consisting of

investments at the primary, secondary, and tertiary prevention levels to prevent violent

extremism and the components that contribute to it, including disinformation and propaganda.

Primary prevention refers to efforts to address radicalization before it takes root, including

through broad civic education and media literacy focused on helping the public build resilience

in ways that do not infringe on their right to free speech or free association, and that work as an

alternative to security-based approaches that surveil, monitor, censor, or ban content. Secondary

prevention refers to efforts to mitigate the impacts of already radicalized people and groups,

primarily through surveillance, monitoring, arrests, and interruption of plots. Tertiary prevention

refers to focused deradicalization efforts, including through prison deradicalization programs and

“exit”-type counseling services that help radicalized individuals disengage from extremism.

An effective public health approach to countering disinformation builds prevention and

intervention across all three of those levels—with the significant majority of efforts and

resources on the primary prevention side—and would require four things. First, it must be nimble

and responsive to communities’ needs depending on regional areas of concern. Second, it must

be holistic and whole-of-community in ways that broaden engagement of a wide range of

government offices, agencies, and organizations beyond the security and law enforcement

sectors, such as the education, health and human services, and mental health sectors. It would

include primary prevention efforts through the arts, community organizations, faith communities,

or other community-based non-profits. Third, an effective public-health prevention model rests

on evidence at all levels of intervention. This means moving beyond outcome evaluations that

describe only outputs, or the numbers of people trained, the numbers of downloads of a

particular tool, or other descriptive metrics that do not actually provide evidence of impact.

Finally, a holistic public health approach focuses on building resilient systems as well as resilient

individuals. Resilience to propaganda and disinformation is not merely a technical skill, in other

words: it is also rooted in national and community values and commitment to an inclusive

democracy that must be reinforced, emphasized, and modeled in all aspects of life across the life

course. The aim is to reduce the fertile ground in which disinformation, propaganda, hate and

anti-democratic ideas thrive.

This is a vision of a public health-style prevention system that works to prevent violence

and counter harm while simultaneously promoting concrete steps toward inclusive equity,

respect, coexistence, and real and symbolic recognition of difference. Such a prevention system

gives us the best chance of building community social cohesion, reducing violent outcomes, and

strengthening our democracies.

POLICY RECOMMENDATIONS

1. Invest in a holistic, community-based, public-health approach to preventing the spread of

supremacist ideologies, mis/dis/malinformation, conspiracy theories, and propaganda.

This includes creating avenues to fund both pilot testing of innovative approaches,

followed by national scale-up of what is proven to be effective in primary-level

prevention strategies, including digital literacy and civic education that equips educators,

parents and caregivers, youth mentors, faith leaders, coaches, mental health counselors,

and others with better tools to recognize and “offramp” individuals who are persuaded by

disinformation from further radicalization to violence. The federal government can

support the creation of impact driven networks that bring together government agencies

from well beyond the security sector; civil society institutions like schools, mental health

professionals, sports leagues and after school programs; local NGOs and advocacy efforts

that enhance community wellness; and others. At the local level, people need to hear and

see pathways for their own engagement, to spark their imaginations about ways to act, to

be moved to change their behaviors, to know there are resources to support their learning,

and to want to know more in ways that make a difference in their families’ and

communities’ wellbeing.

2. Incentivize and prioritize rigorous impact assessment and evaluation frameworks to

ensure policies and programs are implemented as intended and are effective beyond

descriptive metrics. Evaluation frameworks and results funded with public dollars should

be made publicly available to ensure transparency and reduce the need for every initiative

to reinvent the wheel.

3. Ensure that prevention initiatives focus on equipping the public with better tools for their

own decision-making, while not interfering with any individual’s freedom of speech,

conscience, or association. We cannot repeat the mistakes of historical civil liberties

violations or promote censorship as a solution to disinformation.

4. Continue to work with the tech sector to remove harmful and dangerous content, while

understanding that banning and content removal is an after-the-fact solution that does not,

on its own, solve the crisis of disinformation and propaganda. Addressing the problem of

disinformation must begin with upstream prevention that reduces the production of and

receptivity to harmful content to begin with.

5. Create a central, national, nonpartisan center for prevention to provide federal, state and

local governments and all local communities with tools, resources, training, capacity-

building, and evidence about what works.

DEFINITIONS

We define disinformation as false, untrue, or incorrect information spread to

intentionally deceive, manipulate, misinform and erode an individual or group’s belief of

established facts, often with a specific interest or goal. This includes efforts from hostile foreign

influence operations, profiteers, and international extremist and terrorist groups who aim to harm

American democracy, U.S. elections, or scam unsuspecting Americans for profit. It also includes

domestic efforts that undermine inclusive democracy, such as antisemitism or anti-immigrant

conspiracy theories, or compromise the physical health and wellbeing of communities.

Disinformation is similar but distinct from misinformation, which is the unintentional sharing of

false or incorrect information or untrue claims spread without the aim to deceive, manipulate, or

harm. It also differs from malinformation, which refers to true claims spread with the intent to

deceive, manipulate, or harm. Propaganda refers to manipulative persuasive techniques that

seek to make people believe true or untrue information, or values and opinions, sometimes using

dis/mis/malinformation, persuasive narratives (stories that help audiences imagine themselves as

heroes, villains, victors, or victims), or rhetoric (emotionally-stirring language, image and

sounds), which lend manipulative power. Extremism is the belief that one group of people is in

dire conflict with other groups who don’t share the same racial or ethnic, gender or sexual,

religious, or political identity. This “us” versus “them” framework positions the ‘other’ as an

existential threat and calls for total separation, domination, or other forms of violence.

Notably, terrorist violence from domestic violent extremists does not usually link back to

specific groups. Instead, it’s most often perpetrated by individuals who have experienced

networked online radicalization through exposure to propaganda. Groups are still an important

source of much of the propaganda that circulates in extremist scenes and subcultures, including

online. Finally, it is important to note that the spread of online propaganda and disinformation is

fueled by how people spend time online. Online radicalization happens in part when people

spend time in echo chambers, where extreme content is self-reinforcing across platforms. There

is also significant algorithmic radicalization through recommendation systems that suggest

content that is related, but more salacious or more extreme than the content the viewer just

watched. This can lead to “rabbit holes” of disinformation, conspiracy theories and propaganda

consumption that are difficult to climb out of.

CONCLUSION

The crisis of domestic violent extremism that is fueled by disinformation and propaganda

cannot be solved by law enforcement and security-based approaches. We must invest in

upstream strategies to keep communities safe from online harms. We seek a world in which

every community is equipped with the tools they need to reject harmful online propaganda,

conspiracy theories, and manipulative content without the need for censorship, surveillance,

banning, or security-based solutions. Thank you for your attention and I look forward to your

questions.

ENDNOTES

1. With gratitude to researchers and staff at American University’s Polarization and

Extremism Research and Innovation Lab (PERIL) who helped prepare this written

testimony: Dr. Pasha Dashtgard, Dr. Brian Hughes, Laura Kralicky, Wyatt Russell, and to

colleagues at the Southern Poverty Law Center for their support and partnership on many

of the intervention tools and guides cited above.

2. Anti-Defamation League, Center on Extremism. “Audit of Antisemitic Incidents, 2022.”

March, 2023. https://www.adl.org/sites/default/files/pdfs/2023-03/ADL-2022-Audit-of-

Antisemitic-Incidents-2021.pdf

3. United States Government Accountability Office, Report to the Ranking Member,

Committee on Homeland Security, House of Representatives. “DOMESTIC

TERRORISM Further Actions Needed to Strengthen FBI and DHS Collaboration to

Counter Threats.” GAO-23-104720. February, 2023. https://www.gao.gov/assets/gao-23-

104720.pdf

4. Anti-Defamation League, a Report from the ADL Center on Extremism. “Murder &

Extremism in the United States in 2022.” February, 2023.

https://www.adl.org/sites/default/files/pdfs/2023-02/Murder-and-Extremism-in-the-

United-States-in-2022.pdf

5. Anti-Defamation League, a Report from the ADL Center on Extremism. “Murder &

Extremism in the United States in 2022.” February, 2023.

https://www.adl.org/sites/default/files/pdfs/2023-02/Murder-and-Extremism-in-the-

United-States-in-2022.pdf

6. Weissman, Sara. “Suspect Identified in Bomb Threats Against HBCUs.” Inside Higher

Ed., November 16, 2022. https://www.insidehighered.com/news/2022/11/17/fbi-says-

most-bomb-threats-against-hbcus-made-minor

7. Miller-Idriss, Cynthia. “How to Counter Andrew Tate’s Growing Subculture of Violent

Toxic Masculinity.” MSNBC, February 7, 2023. https://www.msnbc.com/opinion/msnbc-

opinion/counter-andrew-tates-growing-subculture-violent-toxic-masculinity-rcna69411.

8. U.S. Department of Homeland Security. “Homeland Threat Assessment.” October, 2020.

https://www.dhs.gov/sites/default/files/publications/2020_10_06_homeland-threat-

assessment.pdf

9. Office of the Director of National Intelligence, “Domestic Violent Extremism Poses

Heightened Threat in 2021.” March 1, 2021.

https://www.dhs.gov/sites/default/files/publications/21_0301_odni_unclass-summary-of-

dve-assessment-17_march-final_508.pdf.

10. Executive Office of the President, National Security Council. “National Strategy for

Countering Domestic Terrorism.” June, 2021. https://www.whitehouse.gov/wp-

content/uploads/2021/06/National-Strategy-for-Countering-Domestic-Terrorism.pdf

11. START (National Consortium for the Study of Terrorism and Responses to Terrorism).

(2022). Global Terrorism Database 1970-2020 [data file]. https://www.start.umd.edu/gtd

12. U.S. Department of Justice, Federal Bureau of Investigation, Criminal Justice

Information Services Division. “Supplemental Hate Crime Statistics, 2021”. March,

2023. https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/explorer/crime/hate-crime.

*Findings from the lab are included in the following works:

PERIL (Polarization & Extremism Research & Innovation Lab) & SPLC (Southern Poverty

Law Center). “Empowered to Intervene: An Impact Report on the PERIL/SPLC Guide

to Youth Radicalization. 2021. https://www.splcenter.org/peril-assessments-

impact#impact-parents

PERIL & SPLC. “Building Networks & Addressing Harm: A Community Guide to Online

Youth Radicalization, Impact Study.” November, 2023.

https://www.splcenter.org/peril-assessments-impact#impact-networks

PERIL & SPLC. “Parents & Caregivers Longitudinal Impact Study.”

PERIL & Rosa Luxemburg Stiftung. “Resources to Combat Extremism: Impact Study

Report.” 2022.

PERIL & The Lumina Foundation. “Building Resilient & Inclusive Communities of

Knowledge.” July, 2022. https://perilresearch.com/wp-content/uploads/2022/12/PERIL-

Building-Inclusive-and-Resilient-Campuses-v15.2-FINAL.pdf