Your freedom of speech rights and limits under the U.S. Constitution come down to a simple frame with complicated edges: the First Amendment stops the government from punishing you for what you say, think, or express, but it does not stop private employers, landlords, or online platforms from doing so, and it leaves several narrow categories of speech, such as true threats and defamation, entirely unprotected. Knowing where those lines fall is the difference between a right you can actually invoke and a protection you only assumed you had.
Who the First Amendment Actually Binds
The speech clause is short: “Congress shall make no law…abridging the freedom of speech, or of the press.”1Congress.gov. U.S. Constitution – First Amendment Read literally, it binds only Congress. In practice it binds every level of American government, because the Supreme Court has interpreted the Fourteenth Amendment’s due process clause to apply most of the Bill of Rights to the states.2Constitution Annotated. Fourteenth Amendment Section 1 – Due Process Generally City councils, school boards, state legislatures, and federal agencies all face the same constitutional ceiling.
What the amendment does not touch is private conduct. Your employer can fire you over a social media post. A restaurant can throw you out for a political T-shirt. A platform can delete your comment or close your account. Lawyers call this the state action doctrine: if the entity restricting your speech is not part of the government, the First Amendment does not apply.3Legal Information Institute. State Action Doctrine and Free Speech Because most private employment in the United States is at-will, a private employer can discipline or terminate a worker over speech the employer dislikes, and the worker has no constitutional claim against the company.
There is a narrow exception when a private entity performs a function that has traditionally been the exclusive job of the government. The Supreme Court applied this reasoning to a company that owned and operated an entire town, holding that residents kept their speech rights because the company was functioning as a municipality. Courts rarely find that a private entity qualifies as a government actor under this test, and no court has yet extended it to social media platforms.
What Counts as Speech
The word “speech” in the First Amendment covers much more than talking or writing. Conduct intended to communicate a specific message qualifies for protection when observers would likely understand what is being communicated.4Constitution Annotated. Overview of Symbolic Speech This is sometimes called symbolic speech. Political donations, art, music, silent protest, and even some computer code have been recognized as forms of expression the government cannot freely suppress.
Two cases anchor the doctrine. In 1969, the Supreme Court held that students who wore black armbands to school to protest the Vietnam War were engaged in protected expression, writing that students do not “shed their constitutional rights to freedom of speech or expression at the schoolhouse gate.”5Justia. Tinker v. Des Moines Independent Community School District Twenty years later, the Court held that burning an American flag at a political demonstration is also protected, even though many people find it deeply offensive.6Cornell Law School. Texas v. Johnson The government cannot punish expressive conduct simply because the message provokes outrage.
Expressive conduct does not get identical treatment to pure speech. Courts ask whether a restriction targets the message itself or the non-communicative side of the behavior. A law banning all open fires in public parks is easier to justify than a law banning only flag-burning, because the fire ban targets conduct rather than the political message behind it.
Speech the Government Can Still Punish
Free speech is broad, but it is not absolute. The Supreme Court has identified several narrow categories that receive no First Amendment protection. Each has its own test, and courts apply them strictly to keep an exception from swallowing the rule.
Incitement to Imminent Lawless Action
Advocating illegal activity in the abstract is protected. Telling a crowd “the system is corrupt and ought to be torn down” is lawful political rhetoric. Speech crosses the line only when it is both directed at producing immediate illegal action and likely to succeed in doing so. The Supreme Court drew this boundary in Brandenburg v. Ohio, replacing earlier tests that had allowed the government to punish mere advocacy of revolution.7Justia. Brandenburg v. Ohio A speaker urging a crowd to burn a specific building down right now can be prosecuted; someone writing an essay arguing that property destruction is morally justified cannot.
Fighting Words
Words spoken face-to-face that are so provocative they are likely to trigger an immediate violent reaction fall outside the First Amendment. The Supreme Court recognized this category in 1942.8Justia. Chaplinsky v. New Hampshire Courts have narrowed the category substantially since. It applies only to face-to-face encounters where the specific words used would provoke a reasonable person to immediate violence.
True Threats
A statement communicating a serious intent to commit violence against a specific person or group is not protected. In Virginia v. Black, the Court defined true threats as statements where the speaker communicates a serious expression of intent to commit unlawful violence against an identifiable target.9Legal Information Institute. Virginia v. Black The key word is “serious.” Political hyperbole, venting frustration, and dark humor typically do not qualify, which is why context matters so much in these cases.
In 2023 the Supreme Court clarified what prosecutors must prove about a speaker’s mindset. The government must show the speaker was at least reckless, meaning the speaker consciously disregarded a substantial risk that the recipient would perceive the statement as a genuine threat.10Constitution Annotated. True Threats Someone who genuinely had no idea their words could be read as threatening cannot be convicted under this standard.
Obscenity
Sexual content that qualifies as legally obscene is unprotected, but the legal definition is far narrower than most people assume. The Supreme Court’s test from Miller v. California requires all three of the following: the average person, applying local community standards, would find the work appeals to a prurient interest in sex; the work depicts sexual conduct in a patently offensive way; and the work, taken as a whole, lacks serious literary, artistic, political, or scientific value.11Justia. Miller v. California All three prongs must be met. Material with genuine artistic or political value is protected no matter how explicit it is.
Child Sexual Abuse Material
The production, distribution, and possession of sexual images depicting minors is categorically illegal and receives no First Amendment protection.12Office of the Law Revision Counsel. 18 U.S. Code 2251 – Sexual Exploitation of Children13Office of the Law Revision Counsel. 18 USC 2252 – Certain Activities Relating to Material Involving the Sexual Exploitation of Minors Federal penalties are severe, and the category stands apart from obscenity because the state’s interest is in protecting the children depicted, not in evaluating the material’s content.
Defamation
Publishing a false statement of fact that damages someone’s reputation can trigger civil liability. Defamation includes both libel (written) and slander (spoken). The statement must be presented as fact, not opinion, and must be provably false.
Public officials and public figures face a much higher bar. Under New York Times Co. v. Sullivan, they must prove “actual malice,” meaning the speaker either knew the statement was false or published it with reckless disregard for whether it was true.14Justia. New York Times Co. v. Sullivan Getting some facts wrong in good faith while covering a public controversy is not enough to lose a defamation case brought by a public figure. That is deliberate.
Hate Speech Is Not a Category
“Hate speech” is not a legal category in the United States. There is no hate-speech exception to the First Amendment, and the Supreme Court has said so directly. In Matal v. Tam, a unanimous Court struck down a federal law that denied trademark registration to names considered disparaging, holding that “speech may not be banned on the ground that it expresses ideas that offend.”15Justia. Matal v. Tam
That does not mean hateful expression can never be punished. If speech independently falls into one of the unprotected categories, such as a true threat directed at a specific person, incitement to immediate violence, or fighting words in a face-to-face confrontation, the government can act. What the government cannot do is single out speech for punishment based on the viewpoint it expresses. The Supreme Court reinforced this in R.A.V. v. City of St. Paul, striking down an ordinance that banned fighting words only when they targeted someone based on race, religion, or gender. Even within an unprotected category, the government cannot pick which messages to punish based on the topic.16Justia. R.A.V. v. City of St. Paul
This is one of the sharpest differences between American free speech law and the law of most other democracies. Many European countries criminalize hate speech. The United States has taken a different position: the risks of giving the government power to decide which ideas are too offensive outweigh the harms of hateful rhetoric.
Time, Place, and Manner Rules
Even fully protected speech can be regulated in terms of logistics. The government can control when, where, and how you express yourself, so long as the rules do not target the content of your message. These restrictions must be content-neutral, narrowly tailored to serve a significant government interest such as public safety or traffic management, and leave open alternative channels of communication.17Justia. Perry Education Association v. Perry Educators Association A city can require parade permits to manage traffic. It can enforce noise limits in residential neighborhoods at night. It can ban amplified sound near a hospital. It cannot grant permits only to groups whose message the mayor agrees with.
If a restriction turns out to be content-based, courts apply strict scrutiny. The government must prove the restriction serves a compelling interest and is the least restrictive means of doing so. Very few content-based restrictions survive that test.
Where You Are Matters
How much protection you have on government property depends on the type of space. The Supreme Court recognizes three categories.17Justia. Perry Education Association v. Perry Educators Association
- Traditional public forums are streets, sidewalks, and public parks. The government’s power to restrict speech is sharply limited, and content-based exclusions must survive strict scrutiny.
- Designated public forums are government-owned spaces the government has voluntarily opened for public expression, such as a university meeting hall opened to student groups. Once opened, the space must be run under the same rules as a traditional public forum for as long as it stays open.
- Nonpublic forums are government properties not opened for general expression, such as military bases, government office buildings, or jails. The government can restrict speech here as long as the restrictions are reasonable and are not aimed at suppressing a particular viewpoint.
Banning political leafleting on a public sidewalk would face strict scrutiny and almost certainly fail. Banning it inside a secure government facility would face a much lower bar and would likely be upheld.
Speech at School and on the Government Payroll
Students and public employees keep their First Amendment rights, but the rights are narrower than what a private citizen enjoys on a street corner.
Students
The armband case remains the starting point: students do not lose their speech rights at the schoolhouse gate, and schools cannot ban student expression simply because administrators dislike the message or fear disruption without evidence.18United States Courts. Facts and Case Summary – Tinker v. Des Moines The Supreme Court later gave schools more room over school-sponsored activities like student newspapers and assemblies, letting administrators exercise editorial control when their decisions are reasonably related to legitimate educational goals. The practical line is between a student’s independent expression, which is heavily protected, and expression that carries the school’s name and endorsement, which is more regulable.
Government Workers
If you work for the government, your speech rights turn on two questions. Were you speaking as a citizen on a matter of public concern, or as part of your job duties? And if you were speaking as a citizen, does your interest in speaking outweigh your employer’s interest in running an efficient workplace?
When public employees make statements as part of their official duties, the First Amendment provides no protection at all. The government, acting as employer, can discipline workers for on-the-job speech the way any other employer can.19Justia. Garcetti v. Ceballos When a government employee speaks as a private citizen about a matter of public concern, courts weigh the employee’s interest in commenting on public issues against the employer’s interest in workplace discipline and effectiveness.20Congress.gov. Pickering Balancing Test for Government Employee Speech A teacher writing a letter to the newspaper about the school board’s budget gets strong protection. A prosecutor circulating an internal memo questioning a supervisor’s case strategy gets almost none.
Commercial Speech
Advertising and other speech that proposes a commercial transaction is protected, but less strongly than political or artistic expression. Courts evaluate restrictions on commercial speech through a four-part test. The speech must concern lawful activity and not be misleading; if it fails that threshold, the government can ban it outright. If it passes, the government must show a substantial interest in the restriction, that the restriction directly advances that interest, and that it is no more extensive than necessary.21Justia. Central Hudson Gas and Electric Corp. v. Public Service Commission of New York
The government can ban false advertising, require health-risk disclosures on product labels, and restrict marketing of illegal products. It cannot prohibit truthful advertising about legal products because it disapproves of them or wants to suppress demand.
Online Speech and Social Media
The internet has not changed the core rules, but it has created predictable confusion. When Facebook, X, YouTube, or TikTok removes a post or suspends an account, users often call it censorship. Legally, it is not. These are private companies exercising editorial control over their own platforms, the same way a newspaper chooses which letters to publish.
Federal law reinforces the arrangement. Section 230 of the Communications Decency Act provides that online platforms are not treated as the publisher or speaker of content posted by their users.22Office of the Law Revision Counsel. 47 USC 230 – Protection for Private Blocking and Screening of Offensive Material A platform generally cannot be held liable for what its users say, and it can moderate content, including removing posts or banning users, without losing that legal shield. Section 230 has faced sustained political criticism from both parties, and multiple reform bills have been introduced in Congress, though none had been enacted as of early 2026.
Several states have tried to pass laws prohibiting large platforms from removing content based on political viewpoint. In 2024 the Supreme Court reviewed challenges to these laws in Moody v. NetChoice and sent the cases back to the lower courts, but its reasoning was pointed. The Court stated that a platform’s choices about which content to display or remove are themselves a form of protected expression, and that a state “cannot prohibit speech to rebalance the speech market.”23Supreme Court of the United States. Moody v. NetChoice LLC The fight over state regulation of platform moderation is unresolved, but the Court’s language suggests forcing platforms to carry speech they want to remove raises serious First Amendment problems.
Two Structural Protections
Most speech restrictions work as punishments after the fact. You speak, and if what you said falls into an unprotected category, the government can prosecute. Prior restraint flips the sequence: the government tries to prevent speech from happening in the first place, through court injunctions, licensing, or prepublication censorship. Courts treat prior restraints as presumptively unconstitutional. The Supreme Court established this in Near v. Minnesota, holding that the government generally cannot censor a publication in advance, even if the content could be punished after it appears.24Justia. Near v. Minnesota ex rel. Olson Narrow exceptions exist for national security secrets and certain obscenity, but the burden on the government is extraordinarily heavy.
Speech laws can also be attacked for being written too broadly. Under the overbreadth doctrine, a person charged under a speech restriction can argue not just that the law is unconstitutional as applied to them, but that it would chill protected speech by others who might stay silent out of fear.25Constitution Annotated. Overbreadth Doctrine The overbreadth must be substantial relative to the law’s legitimate reach. A statute banning all public demonstrations would fall on its face; a statute narrowly targeting incitement to immediate violence would survive. The doctrine forces legislatures to write speech restrictions with precision instead of painting broadly and trusting prosecutors to hold back.