Misdiagnosing Extremism

Lisa Nelson

Current Issue

The Biden administration has identified domestic violent extremism as an increasingly significant threat to national security. In 2021, the White House called it "the most urgent terrorism threat the United States faces today." News organizations and non-profits have taken it upon themselves to identify and publicize groups and ideologies deserving of the label. What constitutes extremism, however, is far from clear. And its relationship to the rights protected by the First Amendment raises troubling questions.

The Department of Homeland Security (DHS) defines a "domestic violent extremist" as:

an individual based and operating primarily within the United States or its territories without direction or inspiration from a foreign terrorist group or other foreign power who seeks to further political or social goals wholly or in part through unlawful acts of force or violence dangerous to human life.

Without unlawful acts of force or violence, the accusation of extremism by the government runs the risk of limiting individuals' constitutional rights of free speech and association. DHS acknowledges as much, stating in a footnote to its definition that the "mere advocacy of political or social positions, political activism, use of strong rhetoric, or generalized philosophic embrace of violent tactics...may be constitutionally protected."

To sidestep this constitutional hurdle, the Biden administration has expanded its reliance on the private sector to do what it is not otherwise authorized to do. The private sector faces no obligation to protect constitutional rights and liberties because the Constitution's restrictions apply only to government conduct. This distinction is based on the idea that the private sector, with its reliance on liberty of contract and market regulation, serves as the basis of human flourishing in a free society and, as such, helps guarantee individual liberty.

But doesn't our government also have an obligation to protect our core rights as citizens when they are threatened by private actors as well as the state itself? What happens when the government works with the private sector to usurp rights and liberties without constitutional accountability or transparency?

Unfortunately, these are not hypothetical questions.


Examples of the growing abuse of the label "extremist" abound in our contemporary political vocabulary. Yet in recent years, an even more troubling phenomenon has emerged.

Consider the work of an organization called Moonshot, a firm founded in 2015 by Vidhya Ramalingam and Ross Frenett, both of whom worked at a London-based anti-extremism think tank. Moonshot describes its mission as combating disinformation, violent extremism, and other harms online, which it does in part through a strategy known as the "redirect method."

This method works by training search engines and social-media platform moderators to monitor content and identify users who are doing nothing more than conducting an online search. If a user searches for terms found on Moonshot's list of extremist-linked words and concepts, he is flagged. The user is then micro-targeted with ads that encourage him to contact a mental-health expert by phone or text.

Moonshot is working with DHS, along with at least a dozen U.S. states, to identify search terms and concepts linked with extremism. What constitutes "extreme" content for Moonshot's purposes is neither clear nor transparent, nor is the definition especially narrow: The organization is said to have compiled a list of some 20,000 terms that will trigger its ads. While Moonshot, DHS, and the search engines and social-media platforms involved know what these terms are, the user is given no warning; he is only made aware that a line has been crossed when he is encouraged to reach out to receive mental-health services.

A private entity undermining free-speech rights is cause for alarm in itself; that the government is working with private organizations to identify extremist-linked terms online without any transparency or accountability is of constitutional concern.

This Big Brother intervention at the behest of government chills the First Amendment rights of users targeted by Moonshot's ads. Having received the ad indicates to the user not only that he is being watched by the search engine or social-media platform, but that he has also been flagged as having mental-health issues. Identifying users as potentially needing contact with a mental-health professional merely because they are searching for certain content online is reminiscent of socialist regimes' use of psychiatry as a means of repression. Given the decades-long jurisprudential debates that the United States has devoted to drawing lines between legal and illegal forms of threats, obscenity, and the like, it stands to reason that defining extreme content should be subject to the same process of debate and oversight.

Underlying this entire process is the assumption that searching for certain content online is a precursor to engaging in unlawful acts of force or violence. At best, this assumption is highly speculative; at worst, it offers a way for the government to marginalize and stigmatize viewpoints with which it does not agree. The significant body of research into violent extremism that now exists has not revealed a single set of factors that leads to the kind of radicalization that results in violence. As terrorism expert John Horgan observed in his well-known article, "From Profiles to Pathways and Roots to Routes: Perspectives from Psychology on Radicalization into Terrorism," attempts to systematically profile terrorists have "failed resoundingly."

Moreover, there is no proof that online content is a determinative factor in causing someone to engage in violence in the real world. Evidence suggesting a primary role for online content in the radicalization process is even more inconclusive — some studies suggest that there is no perceptible effect at all. Without a demonstrated causal link between online content and offline violence, there is little beyond anecdote that justifies the association between online searches for extremist material and eventual unlawful acts or violence.

Indeed, Moonshot itself admits that most of those identified as searching for "extremist" content are not threatening violent acts at all. Rather, the organization assumes that these individuals are more likely to be radicalized, potentially leading them to commit violent or otherwise unlawful acts.

Given that these policies are implemented through the intermediary work of private organizations, are there legal avenues for curtailing them? Answering that question requires us to consider both whether the actions of Moonshot and other private actors call for a government response and whether the government's involvement in their actions subject them to First Amendment restrictions.


Even setting aside the federal government's direct involvement in the work of entities like Moonshot, there is a strong case for restricting such work in the name of protecting speech rights.

At a basic level, the reason that private persons and companies are not subject to the constitutional protections is because the Constitution limits only government action — generally referred to as "state action." This state-action doctrine is essential to American law. But there are obvious problems with any attempt to establish a strict demarcation between the public realm and the private, especially when the stakes are high.

Sometimes the trouble presents itself as a failure to protect essential rights from private action — as when the 14th Amendment, which had been drafted to provide a federal mechanism to combat racial discrimination, was deemed inapplicable when private actors were the ones discriminating. In response, the Supreme Court developed an exception to the state-action doctrine, which, in certain circumstances, recognizes that the government can become entangled with private actors in such a way that the Constitution comes to bear on actions that might have otherwise been considered private. This exception proved effective in dismantling segregation and discrimination in the private sector, but could it also be used to support the First Amendment rights of those who are censored by private internet companies?

Some criticisms of the state-action doctrine are relevant to this question. One such critique comes from a 2002 article by Harvard law professor Mark Tushnet that examines the private-sector's impact on the exercise of fundamental rights in what Tushnet calls an "activist state" — one in which the state assumes affirmative obligations — as opposed to a "classical liberal state." As Tushnet argues, in a classical-liberal state, the demarcation between state and private action is relatively clear. But once the state assumes "affirmative obligations," its enforcement of general property and contract laws might well constitute state action for purposes of the state-action doctrine.

While Tushnet focused on the private-sector restrictions of social-welfare rights in his essay, he used the First Amendment in a book chapter expanding on the same subject to illustrate his point:

If free speech is denied to an individual by "private" actors — exercising their background rights to exclude speakers from their property, say, or to impose speech restrictions as a condition of agreeing to employ someone — then a speech opportunity has been burdened, regardless of the source of the burden....The issue is whether such burdens, in light of other free speech opportunities that remain, are substantial enough that one should conclude that such a society is in fact undemocratic. This would require judgments about speech distribution, the degree of restriction, the availability of alternatives, and an evaluation of the significance of particular restrictions.

Harvard's Cass Sunstein expanded on this logic, arguing that a classical-liberal state as Tushnet describes it also assumes affirmative obligations, meaning that state action is in fact omnipresent. In other words, the very existence of constitutional rights engages the state in their protection.

This critique of the state-action doctrine is especially relevant to speech on the internet. Normally, when a platform such as a newspaper publishes third-party content, its publisher can be held liable for that content if it were to include, say, defamatory material, or material that violates intellectual-property rights. With the advent of the internet, Congress carved out certain immunity guarantees for "interactive computer services" — an intentionally broad category defined to include "any information service, system, or access software provider that provides or enables computer access by multiple users to a computer server" — through the Communications Decency Act of 1996 (CDA). Under Section 230 of the CDA, interactive computer services that host third-party content are largely immune from civil liability related to that content.

Here, the state's action — passage and enforcement of Section 230 — gave companies largely free rein to regulate third-party content posted on their online platforms. Lawmakers pointed to the norms of free speech and association to justify the provision — a point made clear when the Supreme Court took up the constitutionality of the CDA in the 1997 case Reno v. American Civil Liberties Union.

While the Court struck down much of the CDA on First Amendment grounds, lower courts have upheld Section 230 largely because of them. In Zeran v. America Online, the U.S. Court of Appeals for the Fourth Circuit reasoned that burdening internet companies with the duty to police all the content posted on their platforms would dampen the market for users and content. Section 230, the court assumed, would remove any obligation or incentive for such services to restrict content posted by their users.

In the early stages of the internet, the logic of this decision held sway. Engaging more users translated into more valuable data for the private sector, so companies were loath to exclude anyone from their platforms. First Amendment norms and values predominated as a result. But as the scope and character of the internet evolved, the incentives and motivations changed. Now that private actors increasingly restrict speech online, the question of government action must be understood anew.

Tushnet's critique of the state-action doctrine argued that, if the 14th Amendment was going to deliver on the promise of non-discrimination, it was dependent on the courts to provide a remedy for private discrimination in the name of democracy. In that case, the state-action doctrine posed a legal and ideological obstacle to undoing segregation and discrimination. That impediment was overcome by showing how entanglement brought the state to bear on private action, allowing for a judicial solution to private discrimination.

The same logic should curtail efforts to undermine the rights of today's internet users by restricting their speech and invading their privacy online. Indeed, even if the government were not working behind the scenes with groups like Moonshot to identify and flag extremist content, its role in upholding background rights and liberties should evoke constitutional concerns.


But the government is working behind the scenes to abridge users' rights. That renders the basic constitutional question even simpler.

Five decades ago, sociologists Seymour Martin Lipset and Earl Raab observed that accusations of extremism are used to suppress political differences, shut down the marketplace of ideas, and downplay substantive ideological conflict. Labeling ideas "extreme" thus robs those views of legitimacy in the democratic political process upon which a liberal order depends. This is something every American should find troubling.

Indeed, history is replete with examples of government labeling groups extremist, fanatical, or irrational as a way to silence dissent. With the advent of World War I, for instance, Congress passed the Espionage Act of 1917, the most pivotal provisions of which targeted individuals who "willfully cause or attempt to cause insubordination, disloyalty, mutiny, or refusal of duty, in the military or naval forces of the United States." World War II brought with it a second challenge to First Amendment freedoms with the infamous Special Committee on Un-American Activities (later the House Un-American Activities Committee), which investigated both aliens and American citizens for alleged acts of disloyalty or subversion.

The anti-communist fervor that swept the United States during the Cold War prompted additional restrictions of free speech. The Internal Security Act of 1950, popularly known as the McCarran Act, posited the existence of a "world Communist movement" that operated by "treachery, deceit, infiltration...espionage, sabotage, [and] terrorism," and declared that the movement presented "a clear and present danger to the security of the United States and to the existence of free American institutions." That same year, U.S. senator Joseph McCarthy gained notoriety for alleging that communist spies had infiltrated powerful American institutions, including the government. Organizations such as the NAACP, the Southern Christian Leadership Conference, and new-left feminist, environmentalist, and anti-war groups all found themselves under surveillance by the FBI at one point or another.

Finally, in 1969, the Supreme Court put a stop to the criminalization of political dissent in Brandenburg v. Ohio. That case involved Clarence Brandenburg, who was filmed at a Ku Klux Klan rally in Ohio making a speech threatening, among other things, that if the president, Congress, and the Supreme Court continued "to suppress the white, Caucasian race, it's possible that there might have to be some revengeance [sic] taken." For this speech, Brandenburg was convicted of violating Ohio's criminal prohibition against advocating crime or unlawful conduct as a means of accomplishing political reform.

Rejecting the limits such laws had placed on political dissent, the Court held that "constitutional guarantees of free speech and free press do not permit a State to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action." In other words, the First Amendment protects the advocacy of unlawful conduct unless the speaker is speaking with the aim of provoking imminent lawlessness. DHS's definition of domestic violent extremist quoted above nods to this protection.

Federal, state, and local governments' cooperation with entities like Moonshot to curtail Americans' free-speech rights offer a direct analogy to the line of cases preceding the Brandenburg ruling. Such similarity raises grave constitutional concerns that we cannot ignore.


Constitutional protection for speech, ideas, advocacy, and philosophies with which we disagree is a central reason that state and federal governments have resorted to private-sector interventions to do what they cannot do. The resulting alliance, however, is far from constitutional.

Whether in the abridgement of First Amendment rights by the private sector or the role being played by DHS behind the scenes, there are pressing reasons to be concerned about the long-term consequences of this activity not only for speech on the internet, but for the American political order.

Accusations of extremism and the suggestion of mental-health problems without any associated unlawful acts or use of force undermine Americans' freedoms of speech and association. Worse still, they eat away at the pluralism that sustains a liberal democracy. Americans cannot allow this activity to continue. As Justice Oliver Wendell Holmes, Jr., once wisely observed, if we are to maintain our free society, we must be "eternally vigilant against attempts to check the expression of opinions that we loathe."

Lisa Nelson is associate professor in the Graduate School of Public and International Affairs at the University of Pittsburgh, a fellow at the Center for Philosophy of Science, and an affiliated faculty member of the University of Pittsburgh School of Law.


from the


A weekly newsletter with free essays from past issues of National Affairs and The Public Interest that shed light on the week's pressing issues.


to your National Affairs subscriber account.

Already a subscriber? Activate your account.


Unlimited access to intelligent essays on the nation’s affairs.

Subscribe to National Affairs.