Which Speech Falls Outside the First Amendment? The Lines Most People Get Wrong

Freedom & Constitution · Evergreen civics guide

Hate speech, lies and offensive opinions are not blanket exceptions. Incitement, true threats, obscenity, defamation, fraud and criminal solicitation each have distinct tests—and the government cannot create a new exception just by changing the label.

Start with the question most arguments skip

Before asking whether words are protected, ask who is restricting them. The Constitution Annotated’s state-action guide explains that the Free Speech Clause applies to federal, state and local government. A private platform or employer generally may set its own speech rules, subject to contracts and other federal or state laws. A private organization can become a government actor only in limited circumstances, such as when government compels the challenged action or the two act jointly.

A government permit denied because of viewpoint raises a constitutional question. A private forum’s moderation usually does not.

“Hate speech” is not a free-standing exception

Government may punish true threats, assaults or unlawful discriminatory conduct. It may not ban an idea merely because officials or listeners consider it hateful. In Matal v. Tam, the Supreme Court rejected a federal disparagement rule because speech may not be prohibited simply for expressing offensive ideas. The Constitution Annotated’s hate-speech discussion also explains that government cannot single out disfavored subjects even within some otherwise regulable categories.

The same caution applies to lies. Fraud, perjury and defamation can be punished under their own rules, but the Court rejected a general “false statement” exception in United States v. Alvarez. The official false-statements analysis is the useful corrective: identify the legally recognized harm instead of assuming every lie loses protection.

The narrow lines government may enforce

The Congressional overview lists a small set of historically recognized categories. Their names are only the beginning; each has a demanding test.

Incitement

The old “fire in a crowded theater” phrase is not today’s incitement rule. Under Brandenburg v. Ohio, advocacy may be punished only when it is directed to inciting or producing imminent lawless action and is likely to produce that action. The current-doctrine guide distinguishes an immediate, likely call to lawbreaking from abstract advocacy—even advocacy of illegal conduct.

True threats

Political hyperbole is not automatically a true threat. Context must show a serious threat of unlawful violence rather than bluster, joke or rhetoric. In Counterman v. Colorado, the Court added a mental-state floor: the prosecution must show at least that the speaker consciously disregarded a substantial risk that the communication would be understood as threatening. The Court’s 2023 opinion and the Constitution Annotated summary explain that recklessness standard.

Obscenity and child sexual-abuse material

Sexually explicit expression is not automatically obscene. Under Miller v. California, a work, taken as a whole, must appeal to a prurient interest under community standards; depict specifically defined sexual conduct in a patently offensive way; and lack serious literary, artistic, political or scientific value. Serious value uses a reasonable-person standard, not local taste. The official obscenity guide lays out the test. Visual depictions of actual minors engaged in sexually explicit conduct, when they meet the governing statutory and constitutional definition, are a distinct category because their production and distribution exploit and harm children.

Defamation, fraud and crime-facilitating speech

Defamation law can reach false factual statements that injure reputation, but constitutional safeguards vary with the plaintiff and subject. A public official suing over official conduct must prove “actual malice”—knowledge of falsity or reckless disregard for truth—not merely hostility or carelessness. The Constitution Annotated’s defamation guide explains the higher bar.

Words may also constitute fraud, solicitation, conspiracy or intentional assistance to a crime. In United States v. Hansen, the Court read a federal law’s “encourages or induces” language as criminal-law terms for solicitation and facilitation—conduct requiring intent to bring about a particular unlawful act—not a ban on broad political advocacy. The official opinion shows why “speech related to crime” is too broad a shortcut.

Fighting words

This category survives, but it is exceptionally narrow: personally directed words likely to provoke an immediate violent response. Later cases sharply limited the doctrine and struck down vague, overbroad and viewpoint-based laws. The official fighting-words history is a warning against treating every insult as unprotected speech.

Protected speech can still face neutral rules

A city may regulate parade routes, sound levels, traffic obstruction and competing uses of a park through content-neutral time, place and manner rules. Such restrictions must serve a significant interest, be narrowly tailored and leave open ample alternative channels of communication. They cannot be a disguised veto of the speaker’s message. The public-forum guide explains the distinction.

Public schools, government workplaces, prisons and the military use additional doctrines. Truthful advertising for lawful activity receives protection, while misleading commercial speech and proposals for illegal transactions may be prohibited. Those are context rules, not a general license to suppress opinions.

In 2026, the Supreme Court reinforced that point in Chiles v. Salazar: government cannot evade serious First Amendment review merely by calling speech “conduct” or inventing a broad new category. The opinion reiterates that recognized exceptions are few, narrow and historically grounded. The Court held only that Colorado’s law, as applied to Chiles’s talk therapy, discriminated by viewpoint and that the lower courts used insufficiently rigorous First Amendment scrutiny; it reversed and remanded. Traditional qualification-based licensing, ordinary informed-consent rules and regulations of conduct that only incidentally burden speech remain subject to their own doctrines, while other content-based but viewpoint-neutral professional-speech rules were left for another day.

The Daily Fix lens

Analysis: A serious free-speech position has to defend two rules at once. Government needs authority to punish true threats, fraud and intentional participation in crime. It also needs a hard boundary against converting “harmful,” “hateful” or “misinformation” into open-ended permission to suppress political dissent.

That is not softness on intimidation or lawlessness. It is the rule of law: name the exception, prove its elements and apply it without favoring one political camp.

The strongest limitation

These categories are not immunity cards. Context, statutory wording, state law and evidence matter, and private discipline may remain possible when the Constitution blocks government punishment. Critics reasonably argue that the recklessness requirement makes some dangerous threats harder to prosecute. The Counterman Court acknowledged that cost but required breathing room so ambiguous or heated speech is not criminalized by mistake.

A five-question reader test

1. Is the restriction imposed by government or by a genuinely private actor?
2. Is the government targeting a viewpoint, or a separately defined harm or unlawful act?
3. Which recognized category or context rule supposedly applies?
4. What evidence satisfies every element—imminence, likelihood, mental state, falsity, intent or legally defined obscenity?
5. Is the rule narrow and evenhanded, or does it give officials discretion to punish the ideas they dislike?

If a claim about “illegal speech” cannot answer those questions, it is probably a slogan rather than a constitutional analysis.

---

Sources and update note

• Categorical approach
• State action
• Hate speech
• False statements
• Incitement
• Counterman v. Colorado (2023)
• True threats
• Obscenity
• Child sexual-abuse material
• Defamation
• United States v. Hansen (2023)
• Fighting words
• Public forums
• Commercial speech
• Chiles v. Salazar (2026)

Sources last checked August 19, 2026 at 11:50 AM ET. This is a general civics guide, not legal advice. First Amendment outcomes are highly fact-dependent, and new Supreme Court decisions or legislation can change the analysis.

The Morning Fix

One consequential story. Five useful minutes.

Get the facts, the center-right case, the strongest counterargument, and the primary-source receipts every weekday morning.

Send me The Morning Fix

By subscribing, you agree to receive The Morning Fix and occasional clearly labeled offers from Daily Fix. Unsubscribe anytime. See our Privacy Policy.