Concept
Hate Speech and R.A.V.
Regulation of unprotected expression
Hate Speech and R.A.V.
“Hate speech” is a political label, not a category of unprotected speech. Racist and bigoted expression keeps its First Amendment protection unless it independently qualifies as fighting words, a true threat, or incitement, and the incitement test is demanding: a state may forbid advocacy of force or law violation only when the advocacy is directed to inciting imminent lawless action and is likely to produce it.1 Outside those classes, the government may not ban bigoted expression merely because it offends. “Speech may not be banned on the ground that it expresses ideas that offend,” the Court has said, and it calls that rule a bedrock First Amendment principle.2 Cohen supplies the reasoning. Government cannot purge public discourse of scurrilous expression, because “one man's vulgarity is another's lyric,” and a ban on a manner of expression risks suppressing the ideas themselves.3
The controversy is old and plain. Supporters of hate-speech laws appeal to the Nation's commitment to equality; opponents appeal to its commitment to free expression and untrammeled public debate. At bottom the dispute is one more chapter in the ancient battle between liberty and equality.4 Doctrine has favored liberty for a structural reason. A rule that punishes bigoted messages selects some ideas for punishment and leaves their opposites alone, so the government ends up deciding which side of a public argument may speak. A definition of racist speech precise enough to avoid sweeping in protected speech is also hard to write. Campus speech codes illustrate both problems: the codes challenged in court were struck down, often as vague or overbroad.
Fighting words reach hateful expression only when it is directed at another person and likely to produce a violent response.5 Texas v. Johnson shows the limit. The government argued that flag burning would provoke a violent response and so amounted to fighting words. The Court refused, because the expression was not directed at a particular person, and “no reasonable onlooker would have regarded [the] generalized expression of dissatisfaction with the policies of the Federal Government as a direct personal insult or an invitation to exchange fisticuffs.”6 The same reasoning applies to a leaflet or a march that insults a group in general.
Only one decision pointed the other way. Beauharnais v. Illinois upheld, by a 5 to 4 vote, a criminal statute against group libel and affirmed a conviction for distributing a racist leaflet.7 Justice Frankfurter rested the result on three ideas: libel lies outside the First Amendment, such publications tend to produce interracial violence, and a person's dignity depends on the reputation of the group to which the person belongs. The first two have lost their footing. Beauharnais predates New York Times Co. v. Sullivan, which extended substantial protection to libelous publications, and it predates Brandenburg and Cohen, which demand a stronger showing of imminent danger than Frankfurter required.8 The Court has never expressly overruled the decision, but its result should be impossible to reach under the modern cases.9
The Skokie litigation shows the modern position at work. American Nazis planned a march through a suburb that was home to many survivors of the concentration camps. The village adopted ordinances meant to stop the march, and the Seventh Circuit struck them down, doubting that Beauharnais remained good law. Under the doctrines taught earlier in this chapter, the march was an easy case. Cohen forecloses banning a symbol because it offends, Brandenburg protects advocacy with no likelihood of imminent success, and R.A.V. bars the government from singling out expressions of racism.10
States are not left without responses, because the equality concerns behind hate-speech laws can be pursued through categories and conduct rules that do not single out a disfavored message. The next subsection takes up R.A.V., which applies the limit within a category to an ordinance aimed at bias-motivated expression. The subsection after it turns to a second route: heavier punishment for a crime committed because of the victim's protected characteristic. A third route, through the true-threats category, comes in a later section.
The R.A.V. Rule
R.A.V. v. City of St. Paul applies the limit within a category to an ordinance aimed at bias-motivated expression. Before dawn on June 21, 1990, the petitioner and several other teenagers taped broken chair legs into a crude cross and burned it in the fenced yard of a Black family who lived across the street from the house where he was staying. Other laws could have punished that conduct, but St. Paul charged him in part under its Bias-Motivated Crime Ordinance. The ordinance made it disorderly conduct to place on public or private property a symbol, object, appellation, characterization, or graffiti, “including, but not limited to, a burning cross or Nazi swastika,” that one knows or has reasonable grounds to know “arouses anger, alarm or resentment in others on the basis of race, color, creed, religion or gender.” The petitioner moved to dismiss the count as substantially overbroad and impermissibly content based. The trial court agreed, and the Minnesota Supreme Court reversed.
The Minnesota court's construction of the ordinance controlled everything that followed. It read the phrase about anger, alarm, or resentment to confine the ordinance to fighting words within the meaning of Chaplinsky, words that by their very utterance inflict injury or tend to incite an immediate breach of the peace. On that basis it rejected the overbreadth claim and upheld the ordinance as a narrowly tailored way to protect the community against bias-motivated threats.11 A federal court is bound by a state court's construction of a state or local enactment, so the Supreme Court took that reading as given and assumed the ordinance reached only proscribable fighting words. That assumption answered the first question of this chapter's two-step analysis, whether the expression falls within a category, and left the second: does the law respect the limits that remain inside the category? The Court held that it does not, because the ordinance prohibits otherwise permitted speech solely on the basis of the subjects the speech addresses.12
The judgment was unanimous, but the reasoning split five to four. Justice Scalia wrote for five Justices, who held the ordinance facially invalid because it discriminated on the basis of speech content and was underinclusive. By singling out only fighting words based on race, color, creed, religion, or gender, it was selective on the basis of content or viewpoint.13 Justice White, joined by Justices Blackmun and O'Connor, and in part by Justice Stevens, wrote for four Justices who agreed that the judgment should be reversed.14 They would have decided the case within established First Amendment law, holding the ordinance fatally overbroad because it criminalizes protected expression along with unprotected expression.15 Only the majority's reasoning is the Court's holding, so it supplies the rule. The concurrences still matter, because they show where the majority's rule is most exposed, and the overbreadth view they press is the drafting limit taught in the previous section.
The majority began by reading the older statements that these categories receive no First Amendment protection as shorthand. Obscenity, defamation, and fighting words may be regulated because of their constitutionally proscribable content, but that permission does not make them categories “entirely invisible to the Constitution.” Government may proscribe libel; it may not proscribe only libel critical of the government.16 The reason lies in why fighting words are unprotected at all. They fall outside the Amendment not because their content communicates any particular idea but because it embodies a particularly intolerable and socially unnecessary mode of expressing whatever idea the speaker has in mind.17 A law that sorts fighting words by the idea they carry abandons the only reason the category exists and begins to punish messages.
The majority called the resulting rule a limit on content discrimination and rejected the concurrences' description of it as an underinclusiveness rule. A state may, for example, prohibit obscenity only in certain media or markets. That line leaves some obscenity untouched, but it does not sort speech by message.18
Applied to St. Paul's ordinance, the rule gave a short answer. Read with the Minnesota court's construction, the ordinance reached only fighting words that insult or provoke violence “on the basis of race, color, creed, religion or gender.” Abusive invective, however vicious, stayed permissible unless it addressed one of those topics. A speaker who used fighting words to express hostility based on political affiliation, union membership, or homosexuality fell outside the ordinance.19 In operation the ordinance also discriminated by viewpoint. A sign calling all anti-Catholic bigots misbegotten would pass, while a sign calling all papists misbegotten would not, because only the second provokes violence on the basis of religion. Fighting words that do not invoke race or religion, such as aspersions on a person's mother, would be open to those who favor tolerance and closed to their opponents. The city had no authority to let one side of a debate fight freestyle while binding the other to rules.20
The city's own defense confirmed the suspicion. St. Paul argued that a general fighting-words law would not meet its needs, because only a content-specific measure could tell minority groups that the majority does not condone the group-hatred aspect of such speech. A city that justifies selectivity by the message it wants to send has conceded that it is choosing among ideas. The Court answered that majority preferences must be expressed in some fashion other than silencing speech because of its content.21 The county attorney had said the same thing more bluntly in juvenile court: the burning of a cross expresses a message, and the content of that message is what the ordinance legislates.22
Because a rule against all selectivity would be unworkable, the majority said its limit is not absolute. The general ban on content discrimination exists to keep government from driving ideas or viewpoints out of the marketplace, and discrimination among instances of a proscribable class often poses no such danger.23 The majority identified three situations in which selectivity within a class is permitted:
- The distinction rests entirely on the very reason the whole class is proscribable. A reason neutral enough to exclude the class from protection is neutral enough to sort within it. A state may prohibit only the most patently offensive obscenity, measured by prurience, and Congress may criminalize only threats against the President. Neither may single out obscenity that includes offensive political messages, or only threats that mention the President's policy on aid to inner cities.24
- The subclass is tied to secondary effects of the speech, so that the regulation is justified without reference to content. A state could, for example, permit all obscene live performances except those involving minors.25
- The discrimination is of such a nature that there is no realistic possibility that official suppression of ideas is afoot. The Court's example was a ban on only those obscene motion pictures with blue-eyed actresses.26
The opinion adds a separate point about conduct. Words can violate laws directed at conduct and not at speech. A law against treason is violated by telling the enemy the Nation's defense secrets, so a content-based subcategory of a proscribable class can be swept up incidentally by a statute aimed at conduct. The Court's example was sexually derogatory “fighting words” that may violate Title VII's general prohibition on sex discrimination in employment.27
St. Paul's ordinance fit none of the three exceptions. On the first, the question is what makes fighting words punishable, which is their intolerable mode of delivery. The city had not selected, for example, only those fighting words that convey ideas in a threatening, as opposed to a merely obnoxious, manner. It had proscribed fighting words of whatever manner that carry messages of racial, gender, or religious intolerance.28 On the second, any harm the ordinance addresses comes from the emotive impact of the speech on its audience, and that is not a secondary effect of the kind the exception covers. On the third, the city's own statements in the case gave ample basis for suspecting that it meant to handicap particular ideas.29
A content-based law outside the exceptions survives only under strict scrutiny, which presumes the law invalid and requires the government to show necessary means to a compelling interest.30 St. Paul said the ordinance protects the basic human rights of members of groups that have historically faced discrimination, including the right to live in peace where they wish. The Court did not doubt that these interests are compelling or that the ordinance promotes them.31 The dispositive question was necessity. An ordinance not limited to the favored topics would have precisely the same beneficial effect, and the only interest distinctively served by the content limitation is the council's display of special hostility toward the biases it singled out.32 Students often assume that the narrower, more carefully targeted law is the safer one. Inside a category, R.A.V. reverses that assumption, because the broader ban serves the same interest without picking sides, and its availability makes the topic-based ban unnecessary. The Court closed by condemning cross burning in a front yard while observing that St. Paul had sufficient means to prevent it without adding the First Amendment to the fire.33
The Court has itself shown which side of the line a cross-burning law can fall on. In Virginia v. Black it upheld the power to ban cross burning done with intent to intimidate, reasoning that R.A.V. never held that the First Amendment prohibits every form of content discrimination within a proscribable area.34 Virginia's statute did not single out speech directed toward one of the disfavored topics. It made no difference whether the burner meant to intimidate because of the victim's race, gender, or religion or because of the victim's political affiliation, union membership, or homosexuality, and the Court observed that one burner had set a cross in the yard of the lawyer who was prosecuting him.35 The same flaming cross therefore produced two results. St. Paul's line followed the idea expressed. Virginia's line followed intimidation, which is the very reason threats are proscribable. A later section returns to Black and to the separate flaw in the prima facie provision that still brought the Virginia statute down.
The slur is identical in all three towns, so the vocabulary cannot decide the answer. The basis of the line decides it. A line drawn by topic fails, a line drawn by general danger passes, and a line drawn by a danger-related feature of delivery passes under the exception for distinctions that track the reason the class is proscribable.
Justice White found the case easy and the majority's route unnecessary. The fighting-words cases hold that generalized reactions of anger, offense, or hurt feelings do not strip expression of protection, so an ordinance built on those reactions criminalizes protected speech along with fighting words.36 He and Justice Blackmun also objected to the majority's method. By holding that a state cannot regulate speech that causes great harm unless it also regulates speech that does not, they argued, the Court set law and logic on their heads. The Court seemed to abandon the categorical approach and to relax scrutiny of content-based laws, and that weakens the traditional protections of speech, since protection will be scant if all expressive activity must receive the same protection.37 White added that the majority's exceptions looked ad hoc. The reasoning behind the exception for threats against the President would equally support a city council's judgment that threats based on race, religion, or gender cause more severe harm than other threats.38
Justice Stevens argued that the ordinance is evenhanded and not viewpoint based: it would bar a Muslim's sign attacking Catholics just as it would bar a Catholic's sign attacking Muslims.39 The majority answered that the distinct injury Stevens identified is produced by a distinctive idea conveyed by a distinctive message, and it called the contrary description wordplay.40 That exchange marks the argument most likely to tempt a student: a legislature has singled out the worst subset of a category. The subset line is safe only when the feature that makes the subset worse is tied to the reason the whole category is punishable, such as a threatening manner of delivery. It fails when the feature is the offensiveness of an idea.
The consequences for hate-speech laws follow from the majority's reasoning. A general ban on fighting words could still be enforced against hate speech, so long as enforcement is not selective. A hate-speech law not limited to fighting words fares worse, because it would still reach only certain forms of hate speech, and the majority would likely find no justification for that underinclusiveness.41 R.A.V. also adds a question to the category analysis. Even when speech falls in a proscribable class, the form of the regulation must satisfy First Amendment standards, because the class is not invisible to the Constitution.42 Finally, R.A.V. is one of three ways the Court has overturned fighting-words convictions. It has confined the doctrine to speech directed at another person that is likely to produce a violent response.43 It has struck laws as vague or overbroad. And it has treated laws that punish only some fighting words, such as expression of hate based on race or gender, as impermissible content-based restrictions.44
Penalty Enhancement
R.A.V. left states a second way to answer bias-motivated crime: leave the speech alone and punish more severely a crime that was already an offense when the defendant committed it because of the victim's race or another protected characteristic. Hate speech may involve conduct that is already subject to sanction, such as destruction of property or disturbing the peace, and enhancing the penalty when racial animosity motivates the offense makes obvious sense.49 R.A.V. itself says that where government does not target conduct on the basis of its expressive content, acts are not shielded from regulation merely because they express a discriminatory idea or philosophy.50 The Court approved the route one term later in Wisconsin v. Mitchell, which marks where the content-discrimination limit stops.
Mitchell and a group of young men and boys beat a white boy severely after Mitchell picked him out as a target. Before the attack the group had discussed a scene from a film in which a white man beats a Black boy, and Mitchell had asked whether they felt ready to “move on some white people.” A jury convicted him of aggravated battery, which ordinarily carried a maximum of two years. Because the jury also found that he intentionally selected his victim because of the boy's race, a Wisconsin provision raised the maximum to seven years. The Court unanimously rejected his First Amendment challenge to the increase.
Mitchell argued that the higher penalty rested on his discriminatory motive in picking the victim, so the statute punished his thoughts and beliefs. The Court answered that the motive for a crime is an established consideration in sentencing. Motive plays the same role under the Wisconsin enhancement that it plays under antidiscrimination laws aimed at employers, which regulate unprotected conduct and leave protected activity alone. The employer comparison shows the mechanism. An employer who fires a worker for a discriminatory reason turns an otherwise lawful act into a violation, and an employer who assaults a worker for that reason makes the attack more culpable.51 In both settings the law attaches consequences to an act, and the actor's reason for choosing the target determines how serious the act is.
The Court separated this reasoning from R.A.V. The ordinance struck down there was explicitly directed at expression, meaning speech or messages, while the Wisconsin statute is aimed at conduct unprotected by the First Amendment.52 St. Paul's ordinance reached a symbol because of what it communicated, so the thing punished was the expression, and the line it drew sorted messages by topic. Wisconsin's statute starts from a battery and adds to the punishment when the defendant picked the victim for a prohibited reason, so nothing in the enhancement depends on whether the defendant said anything. A physical assault lies wholly outside the First Amendment, whereas fighting words, as R.A.V. insisted, remain visible to the Constitution.53 Mitchell thus keeps R.A.V.'s limit for laws aimed at expression. Reading R.A.V. to forbid every law that treats bias differently skips the first question: what does the law punish? An enhancement needs an underlying offense, so it adds to the penalty for a crime such as destroying property and creates no new crime of expressing bigotry.54
The state's reason for the higher penalty also differs from the one St. Paul gave. Wisconsin could reasonably believe that bias-inspired crimes are more likely to provoke retaliatory crimes, inflict distinct emotional harms on their victims, and incite community unrest. Each harm follows from the act and from the choice of a victim for group membership, so the justification does not depend on the state's disagreement with the offender's ideas. The Court also saw no First Amendment barrier to using a defendant's words as evidence of motive, intent, or the elements of a crime. It treated the fear that protected speech might someday be offered to enhance a sentence as too speculative to support an overbreadth or chilling-effect claim. Mitchell's own remarks before the attack show how this works: they explained why he singled out that boy.
The permission has a limit, and Dawson v. Delaware marks it. A murder defendant was sentenced to death after the prosecution introduced evidence that he belonged to a white-racist prison gang, although the gang's beliefs had no bearing on the crime. The Court held that the First Amendment barred the evidence of abstract beliefs and membership. In Mitchell the defendant's bias explained why he committed this particular battery, so the enhancement tracked something about the crime itself. In Dawson the beliefs explained nothing about the offense, so the added punishment could fall only on what the defendant thought and with whom he associated. The connection between the motive and the offense keeps an enhancement on the permitted side.
The burden of proof follows from what the enhancement does to the sentence. In Mitchell the finding of racial selection raised the maximum from two years to seven, so the biased motive operates as the fact that exposes the defendant to the heavier penalty. Apprendi v. New Jersey held that placing a hate-crime enhancement in the sentencing part of the criminal code does not excuse the state from proving the biased purpose, which operates as an element of the offense, to a jury beyond a reasonable doubt. The motive must therefore be proved beyond a reasonable doubt wherever it raises the maximum penalty. In Mitchell itself the jury made the finding that the defendant intentionally selected his victim because of race.