Seven Things Americans Get Wrong About the First Amendment

The most confident sentence in American argument is "You can't say that, it's not protected by the First Amendment." It gets delivered at parties, on talk shows and in comment threads, with the serene confidence of a man who has never read the First Amendment, never read a single case, and learned everything he knows from a sitcom.

I am going to let you in on a secret. The two groups of people who get the First Amendment most wrong are the ones who think it protects nothing they dislike and the ones who think it protects everything they like. The Supreme Court has been correcting both for more than a century, and the corrections are worth knowing, because they are the difference between a country where you can argue and a country where somebody decides what you may argue.

Here are seven popular claims. For each one I will state the claim, test it against the cases, and give a verdict. I will tell you which parts are fact and which parts are my own view. Start with the text, which is only forty-five words long and says, in the part that matters, that "Congress shall make no law... abridging the freedom of speech, or of the press." Keep that sentence in your head. It is a limit on the government.

Myth one: "Hate speech is not protected"

The claim: There is a category of speech called hate speech, and the Constitution doesn't protect it.

The test: There is no hate speech exception to the First Amendment. The Supreme Court has said so repeatedly, including in cases that conservatives and liberals both cite.

In R.A.V. v. City of St. Paul (1992), a teenager burned a cross on a Black family's lawn. St. Paul had an ordinance banning symbols that arouse anger on the basis of race, color, creed, religion or gender. Justice Scalia, writing for the Court, held the ordinance unconstitutional. The city had singled out certain viewpoints and punished them, and that is not allowed, even for speech that falls into an unprotected category such as fighting words. The cross burner could be prosecuted under other laws, and he could not be prosecuted for the viewpoint the ordinance targeted.

In Matal v. Tam (2017), an Asian-American band called The Slants tried to register its name as a trademark. The government refused, citing a provision that barred registration of marks that disparage people. The Court struck down the provision. Justice Alito's opinion made the point that speech that demeans on the basis of race, religion or other characteristics is offensive, but the law protects the freedom to express "the thought that we hate," a phrase from Justice Holmes. All eight justices who took part agreed on the result.

The honest caveats: Some conduct that happens to involve hateful words is not protected. Threats, harassment that crosses a legal line, and incitement to imminent violence can be punished, and in Virginia v. Black (2003) the Court held that a state may ban cross burning done with the intent to intimidate. Private employers, colleges and platforms can have their own rules. But those are narrow categories defined by what the speaker does, not by how offensive the message is.

Verdict: False. The government cannot ban speech because it is hateful.

Myth two: "You can't yell fire in a crowded theater"

The claim: The First Amendment doesn't protect dangerous speech, and the classic example is shouting fire in a crowded theater.

The test: The line comes from Schenck v. United States (1919), written by Justice Oliver Wendell Holmes. But look at the case. Charles Schenck was a Socialist Party official who had mailed leaflets urging men to resist the draft in World War I. The Court upheld his conviction under the Espionage Act. The line about shouting fire was a side remark, a rhetorical flourish, and the case behind it was about handing out political pamphlets. It is not a good case.

The Court later walked away from it. In Brandenburg v. Ohio (1969), a Ku Klux Klan leader was convicted under an Ohio law for a speech at a rally. The Court reversed the conviction. It held that the government may punish advocacy only if it is directed at inciting imminent lawless action and is likely to produce that action. Abstract advocacy of violence, even ugly advocacy, is protected. That is the rule today.

The honest caveats: Falsely shouting fire to cause a panic can still get you into legal trouble, but through the law of disorderly conduct or reckless endangerment, because you are causing a stampede, not because you are expressing an opinion. The phrase is a slogan for "there are limits." It doesn't tell you where they are.

Verdict: Mostly false as a slogan. The real test is Brandenburg, and it is far more protective than the slogan suggests. For a companion treatment, see Seven Free Speech Myths, From Hate Speech to Shouting Fire in a Theater.

Myth three: "The First Amendment applies to private companies"

The claim: If a social media company, a store or an employer removes your speech, that's a First Amendment violation.

The test: The First Amendment begins with the words "Congress shall make no law," and the Fourteenth Amendment extends that to the states. It restrains the government. A private actor is not bound by it, absent a special circumstance.

The Court made this clear in Manhattan Community Access Corp. v. Halleck (2019). The case involved a nonprofit that ran public access channels in New York City. Producers who said they were removed from the channels for their views sued, claiming a free speech violation. Justice Kavanaugh wrote the opinion, five to four, holding that a private operator of public access channels is not a state actor. Operating a forum for speech doesn't convert a private entity into the government.

There are exceptions, and they are narrow. In Marsh v. Alabama (1946), the Court held that a company town that performed all the functions of a municipality could not ban a Jehovah's Witness from distributing literature on its sidewalks. And some state constitutions go further than the federal one: in PruneYard Shopping Center v. Robins (1980), the Court let California protect leafletting in a private mall under its state constitution.

Here is a point that cuts both ways. Conservatives who complain that a platform removed their posts have a policy complaint, and it may be a good one. But if the First Amendment bound private companies, it would also bind your church, your newspaper and your small business, and it would stop them from deciding what to say. In Moody v. NetChoice (2024), the Court said that when a platform curates what it shows, it is engaged in expressive activity of its own. You may or may not like that, but the Constitution protects your right to run your own paper, too.

Verdict: False. It restrains the government, not your boss or an online platform. Whether laws should regulate platforms is a policy argument, and a legitimate one.

Myth four: "Flag burning is illegal"

The claim: Burning the American flag is a crime.

The test: In Texas v. Johnson (1989), the Court held five to four that burning a flag as political protest is protected expression. A year later, in United States v. Eichman (1990), the Court struck down the federal Flag Protection Act that Congress had passed in response, again by five to four. Attempts to amend the Constitution to allow flag protection have been proposed for decades and have not become law.

We discuss one of the votes in the majority in Antonin Scalia Sided With Flag Burners and Defendants, and That Is the Point. That justice was no fan of the act. He was a fan of the text.

The honest caveats: Flag burning can still be a crime if the prosecution is for something else that is neutral about the message, such as arson, trespass, or violating a fire ordinance. The government cannot punish the act because of the message it sends.

Verdict: False, as a matter of constitutional law. Whether it should be a matter of constitutional amendment is a question the people can debate. Many conservatives feel strongly that it should, and that is a legitimate position to hold and argue.

Myth five: "The press clause is for reporters"

The claim: "Freedom of the press" protects professional journalists and news companies, and the rest of us have only speech.

The test: The text protects "the press," and for most of the nation's history the word meant the printing press and the pamphlets, books and newspapers that came off it. In Lovell v. City of Griffin (1938), the Court held that the liberty of the press includes every kind of publication that conveys information and opinion, including a pamphlet handed out by a Jehovah's Witness. And in Citizens United v. FEC (2010), the majority rejected the idea that the government could treat media corporations differently from other speakers, in part because the line between the press and everyone else is so hard to draw.

The flip side is that the press clause does not give reporters special legal privileges that others don't have. In Branzburg v. Hayes (1972), the Court held that reporters have no First Amendment right to refuse to testify before a grand jury about confidential sources. Congress and many states have passed "shield laws" in response, but that is by statute, not by the Constitution.

Verdict: Mostly false. The clause covers anyone who publishes, and it doesn't give journalists special immunity. For the conservative press tradition, see From a Small Magazine in 1955 to a Movement Press That Reshaped the Debate.

Myth six: "The First Amendment protects everything"

The claim: Nothing you say can be punished, because of the First Amendment.

The test: This is the mirror image of myth one, and it's just as wrong. The Supreme Court has recognized categories of speech that the government can restrict because of the harm they cause, which are defined narrowly.

• Incitement. Brandenburg again: speech directed at producing imminent lawless action and likely to do so.
• True threats. In Counterman v. Colorado (2023), the Court held that the government must prove that the speaker acted at least recklessly, meaning he consciously disregarded a substantial risk that his words would be viewed as threatening violence.
• Defamation. False statements of fact that damage reputation. In New York Times v. Sullivan (1964), the Court required public officials to prove "actual malice," that the defendant knew the statement was false or recklessly disregarded whether it was. Whether that standard should change is a live conservative argument, and we lay out both sides in Should New York Times v. Sullivan Be Revisited? The Case For and Against.
• Obscenity. Under Miller v. California (1973), material that meets a three-part test is not protected.
• Child sexual abuse material. Unprotected under New York v. Ferber (1982).
• Fighting words. Face-to-face insults likely to provoke an immediate fight, from Chaplinsky v. New Hampshire (1942), a category the Court has narrowed substantially.
• Fraud and speech integral to crime. A bank robber's demand note, a price-fixing agreement and a false advertisement can all be punished.

Verdict: False. Those exceptions are real. But each is a narrow, defined category, and the government has to fit the case inside it. The default is protection.

Myth seven: "Lying is not protected"

The claim: A false statement has no First Amendment value, so the government can punish lies.

The test: In United States v. Alvarez (2012), Xavier Alvarez lied about having received the Congressional Medal of Honor, and was prosecuted under the Stolen Valor Act. A divided Court struck down the law. A plurality of four justices, led by Justice Kennedy, said that the First Amendment does not allow a general ban on false statements, and two more justices agreed on narrower grounds. Three justices dissented.

The Court did not say lies are valuable. It said that the government cannot punish a lie solely because it is a lie. It can punish lies that cause a legally recognized harm: fraud, perjury, defamation, false statements to federal investigators and impersonating a government official. Narrow laws can pass. A general power to decide what is true is a power no free country should hand to its government.

If you've ever worried about a future administration that you don't trust holding the pen to define what counts as "disinformation," you already understand why conservatives should defend the rule.

Verdict: Mostly false. The law punishes lies that do harm, not lies as such.

How to use this list

When someone tells you the First Amendment does or does not protect something, ask four questions. Is the speaker the government, or a private party? What exactly is being punished, the message or the conduct? Does the speech fall into one of the defined exceptions? And what does the cited case actually say? Most of the time, a few minutes with the opinion will settle it. If you want the cases in one place, see Eight Supreme Court Cases That Explain What the First Amendment Protects, and for how these rules play out on campus, see Seven Myths About Free Speech on Campus, Tested Against the Cases.

The bottom line

• The First Amendment is a limit on the government. It does not bind your employer or a social media company, and a debate about regulating platforms is a policy argument, not a constitutional right.
• There is no general exception for hateful, offensive or false speech. There are narrow exceptions for incitement, true threats, defamation, obscenity and a few others.
• The rule that protects the speech you hate protects the speech you love. Conservatives, who are often on the unpopular side of the culture, have the most to gain from keeping it strong.
• Read the cases. R.A.V., Brandenburg, Halleck, Johnson and Alvarez are free on the Supreme Court's website and each is readable.
• Next time someone says "that's not protected," ask them which case says so.