What Is Incitement?

Incitement refers to speech that urges or encourages others to commit illegal acts — and under American law, it is one of the few narrow categories of expression that the First Amendment does not protect. The concept sounds simple but is legally demanding: not all speech that advocates illegal action qualifies as incitement. The speaker must intend to produce imminent lawless action, and the speech must actually be likely to do so. Abstract advocacy of revolution, generalized expressions of anger toward authority figures, and political rhetoric that hints at violence do not meet this demanding standard.

The practical effect of this demanding standard is that American law protects a very wide range of speech that many other democracies would prohibit. A speaker who tells an angry crowd that a certain government official deserves to be harmed is likely protected unless the context makes it clear that violence is being directed at a specific target immediately. This may feel counterintuitive, but the logic is well-grounded: if the government could punish speech simply because it expressed ideas that could, in some chain of causation, lead to harmful action, the power to suppress political dissent would be nearly limitless.

The modern incitement standard was not always the law. Early 20th century courts applied a far weaker standard that permitted the prosecution of speech far removed from actual violence. The pivotal World War I era cases — Schenck v. United States (1919) and Debs v. United States (1919) — established the 'clear and present danger' test, which Justice Oliver Wendell Holmes articulated with his famous example: shouting fire in a crowded theater (a line frequently taken out of context today).

The clear and present danger test, whatever its theoretical merits, was applied in practice to sustain convictions for anti-war pamphlets and socialist speeches that posed no realistic threat of imminent violence. Eugene V. Debs was sentenced to ten years in federal prison for delivering a speech expressing sympathy for draft resisters. The idea that his words created a clear and present danger of imminent lawless action is barely plausible. The cases illustrate how speech-protective legal standards can collapse under political pressure.

Over the following decades, the Court struggled with the test, applying it inconsistently and permitting convictions of Communist Party members for advocacy in Dennis v. United States (1951) based on a diluted version that barely required any immediacy of harm at all. The 'clear and present danger' test had become, in practice, a license for prosecuting unpopular political speech. The stage was set for a fundamental revision.

Brandenburg v. Ohio (1969) is the landmark case that established the current incitement doctrine. Clarence Brandenburg, a Ku Klux Klan leader in Ohio, had been convicted under a state criminal syndicalism statute after appearing at a Klan rally where a cross was burned and speeches were made advocating 'revengeance' against Black Americans and Jews. Ohio courts upheld his conviction.

The Supreme Court reversed unanimously in a per curiam opinion, holding that the First Amendment does not allow government to punish advocacy of illegal action unless that advocacy is 'directed to inciting or producing imminent lawless action and is likely to incite or produce such action.' The Brandenburg test has three elements: the speech must be directed at inciting or producing (intent), imminent (timing), lawless action (subject matter), and must be likely to produce such action (probability). All three elements must be satisfied.

The test dramatically raises the bar for incitement prosecutions compared to earlier standards. Abstract advocacy of illegal action — even of violent revolution — is protected. The government must show that the speaker specifically intended to cause immediate lawlessness and that such lawlessness was genuinely likely to result. This standard reflects a core First Amendment judgment: in a democracy, the proper response to dangerous ideas is argument, not prosecution.

Understanding what Brandenburg protects is as important as understanding what it prohibits. The standard is narrow — but it protects a very wide range of speech that might seem, superficially, to urge illegal conduct. Abstract advocacy of illegal action is the clearest example: a speaker who argues that violent revolution would be morally justified under certain circumstances, or that property destruction can be a legitimate form of political protest, is expressing a political view — not inciting imminent lawless action.

Political rhetoric that uses violent metaphors — 'fight back against this administration,' 'we need to take our country back by any means necessary' — is generally protected absent far more specific targeting and immediacy. Hyperbolic expressions of anger — 'I could kill him for what he did' — are not true threats and not incitement. Rap lyrics, fiction, and other artistic expression that depicts violence, even in graphic terms, is protected.

The imminence requirement is particularly important. Speech delivered at a rally about an event occurring next week, or urging people to 'remember this' for a future election, does not create the immediacy that Brandenburg requires. Courts have consistently held that future-directed advocacy — 'we should march on Congress when the time comes' — does not incite imminent lawless action regardless of how inflammatory the surrounding rhetoric may be.

The January 6, 2021 attack on the U.S. Capitol brought incitement doctrine to the center of American political and legal debate. In the days and months after the attack, scholars, commentators, and politicians debated whether the speech delivered by President Trump at the Ellipse immediately before the Capitol breach constituted incitement under the Brandenburg standard.

Most First Amendment scholars concluded that Trump's January 6 speech, while inflammatory and irresponsible, likely did not meet the Brandenburg standard for criminal incitement. The speech did not direct the crowd to enter the Capitol specifically; it did not call for violence against particular individuals with the specificity Brandenburg requires; and the connection between the speech and the breach involved planning and action by specific groups that goes beyond the usual spontaneous response that the immediacy prong contemplates. This legal conclusion was separate from moral and political judgments about the speech's role in the events.

The Senate impeachment trial on an incitement-related article produced acquittal on a 57-43 vote — not the two-thirds supermajority required for conviction. The episode illustrates the important distinction between legal incitement under the First Amendment and political accountability for dangerous speech. A speaker can engage in profoundly irresponsible rhetoric that falls short of legal incitement while still bearing moral and political responsibility for foreseeable consequences.

The internet and social media have created new dimensions to incitement doctrine that courts and scholars are only beginning to work through. Online speech spreads instantly across jurisdictions and audiences, making the geographic and contextual analysis that courts traditionally applied to incitement claims far more complicated. A post made in one city might reach audiences in dozens of countries, with different legal contexts for incitement in each.

The immediacy requirement poses particular challenges online. The Brandenburg test requires that speech be 'imminent' — directed at producing lawless action now, not later. Online speech often operates differently: a post might be shared millions of times over days or weeks, building a sense of grievance and normalizing violent imagery gradually rather than producing immediate action. Some scholars argue the imminence requirement needs rethinking for the internet context; others argue that loosening it would permit the government to use incitement law as a broad tool for suppressing radical political speech.

Platform moderation has become a de facto first line of response to potential incitement online, with platforms removing accounts, posts, and live streams that they determine to be inciting violence. This private response operates outside the First Amendment — platforms are not bound by Brandenburg — but raises its own concerns about consistency, transparency, and the concentration of speech-control power in private hands.

Generative AI has introduced new questions about incitement that existing doctrine does not cleanly answer. AI tools can now produce personalized, targeted content at scale — content that could, in theory, be designed to incite specific individuals or groups to violence more effectively than a mass broadcast. If someone uses AI to generate thousands of personalized messages directing individuals to commit specific acts of violence at specific times and places, the Brandenburg elements might be satisfied in ways that a general public speech could not achieve.

On the defensive side, AI content moderation systems are now central to platforms' efforts to detect and remove inciting speech before it spreads. These systems, trained on examples of prior incitement, are faster and operate at greater scale than human review. But they also produce false positives — removing legitimate political speech that pattern-matches to prohibited content — and can be gamed by determined bad actors who learn the detection systems' weaknesses.

The deeper question is whether society needs different legal tools for AI-assisted incitement — tools designed for content generated or amplified at machine scale rather than human scale. The First Amendment framework was built around the constraints of human speech production. An AI system that generates millions of targeted, personalized incitement messages is a qualitatively different threat that the Brandenburg test was not designed to address.