Free Speech on Trial: How Courts Decide Which Speech Rules Are Fair

Last updated 1 month ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.

A city can tell musicians to lower their volume without gaining the same power to tell them which ideas they may express. The difference lies in the control officials exercise: one rule governs the delivery of speech, while the other selects its message.

Courts generally require a stronger justification for selecting messages than for neutral restrictions on when, where or how people speak. Even a rule treating all messages alike can fail if it shuts down more expression than the government can justify in that setting.

Start with who is restricting speech

The First Amendment’s free-speech protection applies to states through the Fourteenth Amendment, including municipal governments exercising state authority. Its Free Speech Clause prohibits governmental restrictions, rather than ordinary private restrictions on expression.

A workplace rule from a private employer and an ordinance from a city do not enter the same constitutional analysis merely because both limit speech.

Under the state-action doctrine, which asks when a private party counts as the government, merely hosting other people’s speech does not turn a private property owner into a government actor bound by the First Amendment. A privately operated website therefore does not become bound by the First Amendment just by welcoming the public.

When the message determines the rule

A restriction is content-based when it applies because of the topic discussed or the idea or message expressed. Even a rule that looks neutral on its face can be content-based if its justification depends on the speech’s content or it was adopted because government disagreed with the message.

Viewpoint discrimination goes further: within a subject, the government singles out messages for disfavor because of the views they express. A ban on discussing a city project concerns a subject; allowing praise of that project while excluding criticism selects a viewpoint.

The content-based label was the issue in Reed v. Town of Gilbert, decided in 2015, a case about Good News Community Church’s signs announcing worship services at changing locations. The church owned no building and relied on temporary signs to tell people where its next service would be. The town’s sign code allowed ideological signs of up to 20 square feet with no time limits, but limited the size and number of temporary directional signs like the church’s and allowed them only from 12 hours before an event until 1 hour afterward.

The Supreme Court held that the sign code’s classifications were content-based and failed strict scrutiny. Under that demanding test, government must prove that a content-based restriction is narrowly tailored to serve a compelling interest. In plain English, the government needs an especially strong reason, and the rule must closely fit that reason.

In Reed, the Court assumed for argument’s sake that aesthetics and traffic safety were compelling interests, but found the sign distinctions underinclusive, meaning they left too much of the same problem untouched. Other signs created the same aesthetic problems, and the town had not shown that directional signs posed a greater public-safety threat.

In City of Austin v. Reagan National Advertising of Austin, LLC, decided in 2022, the Court rejected the idea that needing to read a sign automatically makes a restriction content-based. Under Austin’s ordinance, off-premises signs advertise things located elsewhere or direct people to offsite locations, and the city’s code at the time prohibited building new ones. An on-premises sign concerns the place where it stands, rather than something elsewhere. Austin’s distinction between the two required reading a sign only to draw neutral location-based lines, not to single out a topic or viewpoint. The Court treated that distinction as content-neutral absent a content-based purpose or justification.

Neutral rules still have to leave room to speak

In a public forum such as a street or park, content-neutral time, place and manner rules must serve a significant government interest, be narrowly tailored and leave ample alternative channels for communication.

At Central Park’s bandshell, New York City required performances to use the high-quality sound equipment and experienced technician it supplied. Performers retained autonomy over the sound mix under the city’s practice. In Ward v. Rock Against Racism, decided in 1989, the Supreme Court upheld the guideline because it was narrowly tailored to neutral interests in controlling excessive volume and providing adequate amplification, while leaving ample communication channels.

Narrow tailoring under this test does not require the least restrictive possible measure. But a neutral rule cannot burden substantially more speech than needed or impose a substantial speech burden that does not advance its goals. On June 26, 2014, the Court held that Massachusetts’s law generally barring people from public ways or sidewalks within 35 feet of certain abortion-clinic entrances or driveways violated the First Amendment. McCullen v. Coakley shows that limit at work: the Court found that Massachusetts had not shown serious use of readily available, less intrusive tools before imposing its clinic buffer zones.

In Clark v. Community for Creative Non-Violence, decided in 1984, demonstrators sought to highlight homelessness through symbolic tent cities in Lafayette Park and the National Mall. The Court upheld the rule preventing sleeping in the demonstration tents, assuming sleeping was expressive conduct and applying the time, place and manner framework.

A permit system also needs limits on official discretion. In Forsyth County v. Nationalist Movement, a county left it to an administrator to decide how much to charge for a permit, or whether to charge at all, and the Court found the ordinance lacked the narrowly drawn, reasonable and definite standards the Constitution requires.

Where speech happens changes the test

Streets and parks are traditional public forums, places historically devoted to assembly and debate. But government ownership alone does not make every space open for public expression.

When government opens a forum for public communication, it must follow the traditional public-forum standards for as long as it keeps the forum open. Those standards permit reasonable time, place and manner rules, while a content-based prohibition must be narrowly drawn to serve a compelling interest.

In a limited forum, government may reserve space for certain groups or topics to serve the forum’s legitimate purpose. The boundaries must be reasonable in light of that purpose, and government cannot discriminate by viewpoint against speech within them. Restricting a comment period to its defined subject is different from allowing only agreement on that subject.

In a nonpublic forum, property not opened by tradition or designation for public communication, government may reserve it for its intended purposes. Speech restrictions there must still be reasonable and cannot suppress expression merely because officials oppose a speaker’s view.

Some speech follows a different rulebook

Speech does not lose First Amendment protection merely because its ideas are hateful or offensive. The narrower exceptions concern specific legal categories, not a general power to ban disagreeable messages.

Incitement requires advocacy directed toward producing imminent lawless action and likely to produce it.

To punish a statement as a true threat, the government must show at least recklessness: that the speaker consciously disregarded a substantial risk that the communication would be viewed as threatening violence.

The Miller obscenity test asks whether an average person applying contemporary community standards would find that a work, taken as a whole, appeals to a prurient sexual interest. Prurient means a shameful or morbid interest in sex, as distinct from a normal sexual interest. The test also requires that the work depict or describe, in a patently offensive way, sexual conduct specifically defined by applicable state law, and that the work, taken as a whole, lacks serious literary, artistic, political or scientific value.

Defamation also has constitutional limits: a public official claiming a defamatory falsehood about official conduct must prove actual malice, meaning knowledge of falsity or reckless disregard of whether it was false. That meaning concerns the speaker’s treatment of truth, not simply personal hostility.

Even unprotected categories are not outside the Constitution: under R.A.V. v. City of St. Paul, government may not regulate them based on hostility or favoritism toward an otherwise protected message they contain.

Truthful, nonmisleading commercial speech about lawful activity, such as advertising, follows a test from the Supreme Court’s Central Hudson decision. The government must identify a substantial interest, directly advance it and avoid a restriction more extensive than necessary.

Public-school student speech has its own context: Tinker v. Des Moines protects expression that does not materially disrupt school work or collide with others’ rights. Morse v. Frederick permits a school principal to restrict speech at a school event that is reasonably viewed as promoting illegal drug use. That ruling distinguished such promotion from political debate about drug laws.

Adult-theater zoning is another special case: in Renton v. Playtime Theatres, decided in 1986, a city ordinance barred adult movie theaters within 1,000 feet of any residential zone, home, church, park or school. The Court treated it as a time, place and manner rule because it did not ban adult theaters altogether and the district court found the city’s predominant concerns were secondary effects rather than film content. Here, secondary effects means the claimed consequences for the surrounding community, rather than disagreement with the films’ message. Not every justice agreed: Justice William Brennan’s dissent argued that the circumstances suggested an effort to suppress expression and criticized relying on similar restrictions elsewhere to shield the city’s motives from scrutiny.

How the same questions work online

In 2021, Florida and Texas enacted laws that curtail large platforms’ content moderation, their filtering, prioritizing and labeling of users’ posts. The Supreme Court’s 2024 decision in Moody v. NetChoice explained that platforms receive First Amendment protection to the extent they create expressive products, including certain curated feeds of others’ speech. Selecting, excluding, organizing and prioritizing third-party posts can produce a platform’s own compilation of expression.

The Court vacated the lower-court judgments and sent the Florida and Texas cases back because the courts had not properly analyzed the facial challenges to the statutes. In a First Amendment facial challenge, one aimed at a law as written rather than at one use of it, a plaintiff must show that a substantial number of the law’s applications are unconstitutional, judged against its legitimate reach. The decision was neither a blanket approval nor a blanket rejection of every application of those laws.

In Lindke v. Freed in 2024, the Court said an official’s social-media speech is attributable to the state only when the official has actual authority to speak for it and purports to exercise that authority.

Officials can also violate the First Amendment by coercing a third party: the claim requires conduct that, viewed in context, could reasonably be understood as a threat of adverse government action to punish or suppress the plaintiff’s speech. In National Rifle Association of America v. Vullo, decided in 2024, the NRA alleged that a former New York financial regulator coerced regulated companies to punish or suppress its gun advocacy. Taking the complaint’s factual allegations as true, as courts do at the pleading stage, the Court concluded that the NRA had plausibly alleged a First Amendment violation; that was not a finding that the alleged conduct occurred.

In Murthy v. Missouri, decided in 2024, the plaintiffs sought an injunction barring federal agencies and officials from pressuring or encouraging social-media platforms to suppress speech. An injunction is a court order requiring or prohibiting a specific act. The Court held that neither the individual nor state plaintiffs had established standing, the kind of injury that entitles a party to sue, to seek one. The Court did not reach the merits and expressed no view on the lower court’s standard for turning private conduct into state action. So the Court never decided whether the alleged pressure was constitutional.

What to check before challenging a restriction

Start by collecting the exact rule, the person or institution enforcing it, the affected speech and the setting. Then document how the rule was applied: which messages were allowed, what reason officials gave and what opportunities remained to reach the intended audience.

A federal statute known as Section 1983 permits a civil lawsuit over constitutional rights violations by someone acting under the authority of a state, a territory or the District of Columbia. Possible remedies include financial compensation, orders requiring or forbidding conduct, and declarations of the parties’ rights, depending on the claim and defendant.

One limit is written into the statute: when the defendant is a judge acting in a judicial capacity, it bars an injunction unless a declaratory decree was violated or declaratory relief was unavailable. Declaratory relief is a court’s determination of the parties’ rights, rather than an order directing conduct.

Federal standing generally requires a concrete, particularized actual or imminent injury that is fairly traceable to the challenged action and can be redressed by a favorable ruling.