The Codified First Amendment

Keith Werhan. Freedom of Speech: A Reference Guide to the United States Constitution. Praeger, 2004.

[T]he diversity of communicative activity and governmental concerns is so wide as to make it implausible to apply the same tests or analytical tools to the entire range of First Amendment problems. This premise provides the impetus for making First Amendment doctrine more precise and at the same time more complex, developing tools and tests that are greater in number but consequently applicable to increasingly smaller categories of First Amendment issues.

~ Frederick Schauer

The interplay between the categorization and content distinction principles has led the Supreme Court to codify free speech doctrine. This codification, as Frederick Schauer recognized, carries both risk and reward for the development of a workable jurisprudence. On one hand, codification promotes consistency and predictability by systemizing First Amendment doctrine. On the other hand, categorization risks over-formalizing the doctrine and thus forcing outcomes in particular cases that betray rather than further free speech values. The promises and pitfalls of codification mirror each other: Codification offers to create, as it threatens to undermine, doctrinal coherence.

This chapter surveys the primary elements of the Supreme Court’s codification of free speech doctrine by speech category. According to the usual workings of the content distinction principle, a strong presumption of unconstitutionality attaches whenever the government targets a particular category of speech for any special prohibition or restraint. The best avenue of escape from the nearly certain death that awaits such a law is a demonstration that the targeted category does not merit full protection by the First Amendment. The categorization principle invites such demonstrations. The justices’ use of categorization over the years has resulted in a series of special doctrinal rules applicable to particular categories of speech. On occasion, as we have seen with respect to subversive advocacy, the Court has specially tailored the strict scrutiny rule to fit a category of fully protected speech (Brandenburg). We shall now see that in other instances, as with respect to obscenity and child pornography, the justices have developed particularized definitions of certain categories of speech in order to exclude them from First Amendment protection. And we shall also see that, as with respect to commercial advertising, the Court has formulated a middle position that provides an intermediate level of constitutional protection for still other categories of speech.

This chapter reviews the major categories of speech other than subversive advocacy for which the Court has developed special principles, with one conspicuous exception. Defamation is not considered here because it is discussed in the volume on freedom of the press in the Reference Guides to the United States Constitution series of which this book is a part. Because First Amendment defamation cases typically involve the press, that speech category is more appropriately covered in that volume.

Provocative Speech

In Brandenburg’s speech, discussed in Chapter 3, the Ku Klux Klan leader relied on racial and anti-Semitic epithets to warn of a rising tide of white supremacist feeling that might culminate in “some revengeance” being taken against those who “continue[d] to suppress the white, Caucasian race.” He also called for the removal of African Americans to Africa and of Jews to Israel. Brandenburg’s speech took place on a farm before about a dozen true believers, and his urgings produced no action on their part. But what if Brandenburg had given that same speech, surrounded by his hooded followers, on a street corner in a crowded urban center, before a large, racially diverse audience that included many people who were deeply offended by his rhetoric and who were violently opposed to his ideology?

In such a setting, Brandenburg’s speech would have presented a “provocative speech” problem, which is the mirror image of subversive advocacy. Although provocative speech, like subversive advocacy, threatens disruption of public order, it does so for a different reason. With respect to subversive advocacy, the fear is that the speech might persuade an audience to act illegally, as described by the following causal chain:

speech → persuasion → action → harm.

By contrast, the concern over provocative speech is that it might arouse such hostility among audience members that they respond violently against the speaker, or more generally into riot, or at the least, that such speech might offend the sensibilities of some listeners. The causal chain for provocative speech is thus revised:

speech → rejection → reaction → harm.

Not only is the government’s assessment of the sources of potential disorder reversed for subversive advocacy and provocative speech, but also the free speech concerns are flipped in the two problems. Advocates of free expression worry that official efforts to silence subversive advocacy invariably end up stifling legitimate communications between speakers and their audiences. When government officials control speech because of its provocative nature, the First Amendment threat is that they are enforcing a “heckler’s veto” over speakers with unpopular messages, that is, that they are allowing audience disagreement with a message to silence the messenger.

Notwithstanding these differences, the free speech problems presented by subversive advocacy and provocative speech have much in common. First, in both settings the central challenge facing reviewing courts is deciding when speakers should be liable for the actions of their audiences. Second, the speech that government officials regard as subversive or provocative typically is delivered by dissidents whose views sharply challenge the dominant sentiments of their communities. Finally, in both situations it is all too easy for government officials to conflate their legitimate interest in protecting public safety and order during truly dangerous emergencies with the impulse to protect the community from the “threat” of “dangerous ideas.” Majoritarian censorship is thus a hovering presence in both subversive speech and provocative speech problems.

Hostile Audiences

The Supreme Court first took up the provocative speech issue in the early 1940s, not long after the justices had accepted the Holmes/Brandeis brief for strongly protecting freedom of speech as a fundamental, constitutional right. Jesse Cantwell, a Jehovah’s Witness, was convicted of “inciting others to breach of the peace” as a result of his proselytizing on a public sidewalk in a “thickly populated,” predominantly Catholic neighborhood in New Haven, Connecticut (Cantwell v. Connecticut, 1940). Cantwell stopped two men, who were Roman Catholics, and received their permission to play a phonograph record for them. The record, entitled “Enemies,” attacked the Catholic Church and its pope in robust terms. According to the Court’s recitation of the facts, the two men “were incensed by the contents of the record and were tempted to strike Cantwell unless he went away.” There was no violence, however. When the two men told Cantwell to leave, he did.

The Court reversed Cantwell’s conviction. The justices began by establishing their default position on provocative speech, that speakers enjoy “a right peacefully to impart [their] views to others.” The government can trump that right, the Court held, only by proving a “clear and present danger of riot, disorder, interference with traffic upon the public streets, or other immediate threat to public safety, peace, or order.” The First Amendment protected Cantwell because he had spoken “peacefully”: He had not assaulted, threatened, or abused his audience. He was guilty only of trying “to persuade a willing listener to buy a book or to contribute money in the interest of what Cantwell, however misguided others may think him, conceived to be true religion.” It was the content of Cantwell’s speech that “naturally” had offended the two men. Cantwell had intended no offense, and the offense his speech had caused was itself insufficient to establish any “clear and present menace to public peace and order.” The Court explained:

In the realm of religious faith, and in that of political belief, sharp differences arise. In both fields the tenets of one man may seem the rankest error to his neighbor. To persuade others to his own point of view, the pleader, as we know, at times, resorts to exaggeration, to vilification of men who have been, or are, prominent in church or state, and even to false statement. But the people of this nation have ordained in the light of history, that, in spite of the probability of excesses and abuses, these liberties are, in the long view, essential to enlightened opinion and right conduct on the part of the citizens of a democracy.

The Court’s adoption of the clear and present danger test in Cantwell provided strong constitutional protection for provocative speech, essentially limiting the government to silencing speakers whose remarks trigger a hostile audience reaction only as a last resort to maintain public order.

The justices elaborated and reinforced their acceptance of provocative expression within freedom of speech later in the 1940s when they set aside a conviction of an anti-Semitic, fascist speaker for violating a state law which prohibited speech that “stirs the public to anger, invites dispute, brings about a condition of unrest, or creates a disturbance” (Terminiello v. City of Chicago, 1949). In language that is often quoted, the Court explained, ironically, that the statute described protected rather than unprotected speech. The Court in Terminiello wrote:

[A] function of free speech under our system of government is to invite dispute. It may indeed best serve its high purpose when it induces a condition of unrest, creates dissatisfaction with conditions as they are, or even stirs people to anger. Speech is often provocative and challenging. It may strike at prejudices and preconceptions and have profound unsettling effects as it presses for acceptance of an idea. That is why freedom of speech, though not absolute, is nevertheless protected against censorship or punishment, unless shown likely to produce a clear and present danger of a serious substantive evil that rises far above public inconvenience, annoyance, or unrest…. There is no room under our Constitution for a more restrictive view. For the alternative would lead to standardization of ideas either by legislatures, courts, or dominant political or community groups.

After Cantwell and Terminiello, there seemed little doubt of the Court’s commitment to protecting provocative speech under the clear and present danger rubric.

The Court’s acceptance, even celebration, of provocative speech in these original cases is all the more striking when compared with the justices’ initial, almost instinctual rejection of free speech claims in the original subversive advocacy cases that arose during World War I. But in truth, the facts of Cantwell and the state law in Terminiello made these cases easy marks for the clear and present danger test. Cantwell’s speech was delivered to an audience of two, hardly the makings of a riot. And even though Cantwell’s listeners apparently considered registering their offense over Cantwell’s choice of records violently, Cantwell defused the situation by quietly acquiescing to their demand that he depart. In Terminiello, the Court never even assessed the factual context of the speech, which was far more explosive than that of Cantwell. The justices believed that the state law was so overbroad that no speaker could be convicted constitutionally under its terms, regardless of the facts. (The “overbreadth” doctrine is discussed in the following chapter.) Thus the potential difficulty of applying the clear and present danger test to speech that many members of a community believe to be highly threatening, a difficulty that would split the Court wide open in Dennis, lay dormant in these early, easy cases.

But not for long. Just over two years after deciding Terminiello, the Court upheld the disorderly conduct conviction of a “Young Progressive” for his soapbox oratory one Friday evening on a street corner in Syracuse, where such Friday addresses had become customary (Feiner v. New York, 1951). Feiner’s speech included what the Court described as “derogatory remarks” about President Harry S. Truman (a “bum”), the Mayor of Syracuse (“a champagne-sipping bum”), and the American Legion (“a Nazi Gestapo”). This was not all. The Court reported that Feiner, who spoke in a “loud, high-pitched voice” before a racially mixed crowd of about 75 to 80 people, “gave the impression that he was endeavoring to arouse the Negro people against the whites, urging that they rise up in arms and fight for equal rights. The statements before such a mixed audience ‘stirred up a little excitement.'” At least one member of the audience threatened violence if the police officers on hand did not put an end to Feiner’s speech, and soon thereafter, the police dispersed the crowd. They arrested Feiner after he had ignored several demands to stop his speech.

The Court in Feiner followed the doctrine of Cantwell and Terminiello, but not the spirit of those two decisions. The justices accepted, almost summarily, a trial court finding that the police had silenced Feiner because “a clear danger of disorder was threatened,” and not because of an intention to suppress his point of view. Chief Justice Vinson, writing for the Court, explained, “[Feiner] was neither arrested nor convicted for the making or the content of his speech. Rather, it was the reaction which it actually engendered.” But as Justice Douglas noted in dissent, those two considerations cannot be so easily disentangled, for as in Cantwell it was the content of Feiner’s speech that had triggered the audience reaction that led to his arrest. “[I]f the police throw their weight on the side of those who would break up meetings,” Douglas argued, “the police become the new censors of speech.” Indeed, Justice Black in his dissent asserted outright that Feiner had been convicted “for his unpopular views,” and that by legitimating the conviction, the Court had sanctioned “a simple and readily available technique by which cities and states can with impunity subject all speeches … to the supervision and censorship of the local police.”

Vinson responded to Black and Douglas by arguing that Feiner had not engaged in the kind of peaceful advocacy that had characterized Cantwell. To Vinson, Feiner left his constitutional protections behind when he “passe[d] the bounds of argument or persuasion and under[took] incitement to riot.” The Court in Cantwell had hinted that a speaker intent on provoking a hostile reaction from the audience might be open to prosecution for any resulting disturbance of the peace, and Vinson suggested that Feiner’s name-calling and the pitch of his voice pointed to such an intent. Such evidence provides an unconvincing basis upon which to deny a speaker First Amendment protection, however. As Justice Douglas, the author of the Court’s opinion in Terminiello, reminded, it was “not unusual” for a speaker to resort to “exaggeration, to vilification of ideas and men, to the making of false charges.” Such was an unavoidable byproduct of free speech. Moreover, it would betray rather than follow Cantwell to infer provocative intent from the controversial nature of a speaker’s remarks or from an audience’s hostile reaction to those remarks. As Justice Black, also evoking Terminiello, observed, violent disagreement often results from public discussion on “controversial topics.” And finally, although Cantwell and Terminiello had left open the possibility of halting provocative speech when there was a clear and present danger of serious disorder, Black was surely correct in concluding, “It seems farfetched to suggest that [there had been] any imminent threat of riot or public disorder in Feiner.”

Chief Justice Vinson wrote the majority opinion in Feiner less than six months before his plurality opinion in Dennis eviscerated the clear and present danger test. The two Vinson opinions shared a structural similarity: They proclaimed freedom of speech in the abstract, but in application they accorded decisive weight to official judgments that threatening expressive activity be limited in the interest of preserving public order. In both opinions, Vinson set the clear and present danger bar so low that it imposed no real barrier to law-enforcement officials intent on safeguarding the public from the perils they ascribed to dissident speech. In both instances, the bar was sufficiently low to allow considerable governmental censorship as well. Justice Black in his Feiner dissent showed the distance between a strong clear and present danger test and Vinson’s flimsier version when he argued that even in a “critical situation,” police officers were under a First Amendment obligation to make “all reasonable efforts” to protect the speaker’s right of free speech. The First Amendment norm reflected by the clear and present danger test—that freedom of speech can be overcome only as a last resort to preserve public order—was lost in Feiner on the eve of Dennis.

The Court decided Feiner at the height of the Second Red Scare, a time in which the balance between individual freedom and state power had lurched decidedly in the direction of the latter. The pendulum swung back when the Red Scare receded and the nation took up the struggle for civil rights for African Americans. Feiner’s speech had foreshadowed that struggle, but the wave of civil rights demonstrations in the 1960s, and official reaction to it, cast Feiner in a new light. Public officials in the segregated South certainly could claim that experience had proven the potential for violent public reaction against civil rights demonstrators. If anything, the threat of violence was far more ominous in southern streets and squares of the 1960s than it had been on Feiner’s street corner. At the same time, nowhere was the censorial potential of Feiner more manifest than in the massive resistance by southern officials to civil rights demonstrations in their communities. Rarely, if ever, has the United States experienced such a formidable “heckler’s veto” of dissident speech.

The pivotal, post-Feiner moment occurred in 1963, when the Court decided Edwards v. South Carolina. In Edwards, the Court reversed 187 convictions for breach of the peace lodged against African American students who had participated in coordinated protests on the grounds of the South Carolina State House. The protestors marched in small groups and carried signs decrying the state’s official policy of racial segregation. Gradually, a crowd of between 200 and 300 onlookers gathered, but neither they nor the protestors threatened any violence. The police on the scene nevertheless demanded that the protestors disperse. The protesters responded by singing patriotic and religious songs, and they were promptly arrested. The Court described the peacefulness of the Edwards protest as “a far cry from the situation in Feiner.” That peremptory description has had the effect of marginalizing Feiner, restricting its reach to cases involving speakers who had actually intended to “provoke[] a given group to a hostile reaction” (Cohen v. California, 1971). The Court in Edwards, by contrast, linked the South Carolina protestors to Cantwell: They had done nothing other than exercise “basic” First Amendment rights “in their most pristine form.” Vindicating the dissenting opinions of Justices Black and Douglas in Feiner, the Court in Edwards concluded that the demonstrators had been convicted simply because “the opinion” they expressed was “sufficiently opposed to the views of the majority of the community to attract a crowd and necessitate police protection.” Freedom of speech, the justices declared, “does not permit a State to make criminal the peaceful expression of unpopular views.” The justices had revived the spirit of Cantwell.

The boldness of the Edwards declaration that speakers are immune from liability arising from hostile audience responses to the content of their message, was tested throughout the civil rights era. Significantly, the Court consistently overturned disorderly conduct convictions of civil rights demonstrators who contributed to crowd-control situations that were a “far cry” from that in Edwards. The most revealing example of the new judicial solicitousness toward provocative speakers occurred in the late 1960s, when a peaceful civil rights march in Chicago drew an expanding crowd of “unruly” onlookers. The police officers on the scene, who were accompanied by a city attorney, ordered the demonstrators to disperse in order to prevent what they believed, in apparent good faith, to be “an impending public disorder.” The protestors were arrested after they refused to comply with that order. The Court acknowledged that the actions of the police were “reasonable” and that their motives were “laudable,” and this being the case, it would appear that they had attempted to stop the demonstration because they believed there was clear and present danger of a riot. Even so, the justices overturned the conviction on the basis of the Edwards principle. It was the crowd, not the demonstrators, who had been disorderly (Gregory v. City of Chicago, 1969).

The justices brought hostile audiences doctrine up to date in Texas v. Johnson (1989), when they reversed the conviction of a demonstrator who had burned a flag of the United States as an act of public protest against the policies and actions of the Reagan administration. Johnson was convicted under a Texas law that prohibited “flag desecration,” which the law defined as “defac[ing], damag[ing], or otherwise physically mistreat[ing] … [a flag] in a way that the actor knows will seriously offend one or more persons likely to observe or discover his action.” Texas defended the law, in part, by invoking its interest in preventing breaches of the peace. But the only evidence the state could muster of any hostile reaction was the testimony of several observers who had been seriously offended by Johnson’s burning of the flag. The Court in Johnson buried Feiner deeper by holding that such a scant showing would not do. The justices refused to indulge the presumption that “an audience that takes serious offense at particular expression is necessarily likely to disturb the peace and that the expression may be prohibited on this basis.”

The Court in Johnson updated the Edwards revival of Cantwell by incorporating the Brandenburg revision of the clear and present danger test into hostile audiences doctrine. The Court wrote, “[W]e have not permitted the government to assume that every expression of a provocative idea will incite a riot, but have instead required careful consideration of the actual circumstances surrounding such expression, asking whether the expression ‘is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.'” The Court’s application of the Brandenburg test in Johnson suggests that, as with subversive advocacy, it will be very difficult for government prosecutors to convince the courts that provocative speech is “likely” to produce harm in the absence of proof that such harm has actually occurred (see NAACP v. Claiborne Hardware Co., 1982).

Fighting Words

There is one exception to the Court’s otherwise strong protection of provocative speech. The justices allow the prohibition of so-called “fighting words” (Chaplinsky v. New Hampshire, 1942). A fighting word is an insult that a speaker addresses to another, “face-to-face,” which has “a direct tendency” to cause a violent response by one who has been so insulted (Gooding v. Wilson, 1972). Such personal insults are excluded from First Amendment protection because the justices regard them as essentially devoid of speech value. Fighting words typically constitute “no essential part of any exposition of ideas” (Chaplinsky); nor are they “in any proper sense communication of information or opinion safeguarded by the Constitution” (Cantwell v. Connecticut, 1940). Given the nearly total absence of free speech value, the threat that fighting words pose to “public order” clearly outweighs their “social value as a step to truth” (Chaplinsky).

It might seem surprising, but this limited exclusion of “personal abuse” (Cantwell) from the First Amendment protection otherwise afforded provocative speech has generated considerable controversy over the years. The source of the controversy is the tension that exists between the fighting words exception and the general sweep of provocative speech precedent. This tension is highlighted by the Supreme Court’s decision in Cohen v. California (1971). Cohen is not a fighting words case. It involved the conviction of a young man for wearing a jacket with the inscription, “Fuck the Draft,” during the height of the controversy over America’s participation in the Vietnam War. Because Cohen’s statement was a comment on the draft directed generally to anyone who might read it, and was not an insult directed personally at any individual, the expression fell into the broader category of speech that provokes a hostile audience. Nevertheless, the Court’s rationale for invalidating Cohen’s conviction undermined the justification for subjecting all fighting words to governmental prohibition.

The state had argued that the inscription on Cohen’s jacket was “inherently likely to cause violent reaction.” This concern was hardly implausible. Because of the passions often aroused by antiwar advocacy during the Vietnam War, one might well anticipate that one or more of the many viewers of Cohen’s jacket would so deeply resent such a vivid statement of protest that they would react violently against Cohen. But the Court held, in line with the hostile audiences precedent, that simple fear of violent reaction could not justify the state in prohibiting speech. The justices explained that the state’s regulatory justification “amount[ed] to little more than the self-defeating proposition that to avoid physical censorship of one who has not sought to provoke such a response by a hypothetical coterie of the violent and lawless, the States may more appropriately effectuate that censorship themselves.” Cohen thus reiterated the position of Justices Black and Douglas in their dissenting opinions in Feiner, which the Court has followed in hostile audience cases since Edwards.

Cohen thus suggests the following question: If the state cannot generally presume provocative speech to be “inherently” likely to cause violent reaction, why can it make such a presumption with respect to fighting words? If anything, the governmental interest in preventing group violence and possible riot should weigh more heavily than the one-on-one confrontation threatened by fighting words. The best response to this disparity in treatment is to be found on the speech side of the balance. The First Amendment premise in a hostile audience problem is that a public speaker is exercising the core free speech right of attempting to persuade an audience to a point of view, albeit an unpopular one. That core right, the argument goes, is not implicated when a speaker personally insults another. Not only are such insults typically for private rather than for public consumption, but also they are not “essential” (Chaplinsky) to the communication of a point of view. Paradoxically, then, the government’s regulatory interest in preserving public order justifies the prohibition of fighting words not because that interest is especially strong, but because the speaker’s free speech claim is especially weak.

It is the presumed weakness of the speech value of fighting words that Cohen questions most sharply, however. In Cohen, the state claimed authority to punish use of the word “fuck” as an unnecessary and unseemly way to express a point of view. The justices reminded, however, that the First Amendment generally left rhetorical strategy to the speaker, not to the state. Moreover, the Court recognized that language often serves a “dual communicative function,” conveying speaker emotion as well as cognitive content. The statements “Stop the Draft” and “Fuck the Draft” each deliver an antidraft message. But who would claim that the impact—and meaning—of those statements are identical? Cohen’s resort to the familiar expletive conveyed the depth of his revulsion against the war effort, a communication that might have been impossible with language that was more culturally acceptable. The very crudity and offensiveness of Cohen’s use of profanity, in this sense, was essential to his message.

While personal insults undoubtedly at times are gratuitous, as are many public uses of profanity, is Chaplinsky’s assumption that fighting words are never essential to communication, in the Cohen sense, correct? The factors that caused the Court to reject Cohen’s conviction were present in Chaplinsky, the very decision that launched the fighting words exception. Chaplinsky, like Cantwell before him, was a Jehovah’s Witness whose sidewalk proselytizing stirred his audience to anger. After a disturbance erupted, a police officer dispersed the crowd and led Chaplinsky toward the station house. En route, they happened upon the city marshal. Words were exchanged, during which Chaplinsky called the marshal “a God damned racketeer” and “a damned Fascist.” The latter insult, leveled as it was near the beginning of World War II, surely must have stung, but did the insult’s harm “clearly outweigh” its value as speech? Certainly, Chaplinsky’s use of “Fascist” carried more First Amendment value than the usual insult. When leveled at the government official whom he considered responsible for violating his freedom of speech, Chaplinsky’s use of “Fascist” had the undeniable ring of political protest, in much the way that Cohen’s use of the expletive “fuck” drove home his strong opposition to the draft. Moreover, the risk of violent reaction seemed especially unlikely—or at least, untenable—in Chaplinsky. After all, Chaplinsky, while he was in police custody, insulted a government official. For the city marshal to attack Chaplinsky under such circumstances would have amounted to official misconduct. Thus, even if the Chaplinsky Court was correct that fighting words often, or even usually, do not merit First Amendment protection, that assumption would not seem to hold in Chaplinsky’s case.

These arguments for First Amendment protection were unavailing in Chaplinsky, however, because the Court did not consider the particular speech context in which Chaplinsky had insulted the marshal. Having excluded the category of fighting words from First Amendment protection, the justices simply determined, as an abstract matter, that the insult “damned Fascist,” delivered face-to-face, so qualified. They did not pause to calibrate the free speech, cost-benefit balance in Chaplinsky’s case itself. And in the years since Chaplinsky, the Court has never developed a satisfactory, principled method for separating the “fighting words” that should merit constitutional protection from those that should not. At least, the justices have not done so formally.

Although the Court continues to declare that laws which generally prohibit fighting words are consistent with the First Amendment, Chaplinsky marks the one and only occasion on which the justices have upheld a fighting words conviction. The most revealing episode occurred in 1972, when the Court managed to overturn convictions in four fighting words cases without explicitly challenging the doctrinal integrity of Chaplinsky. Each of the 1972 quartet involved a speech context in which the speaker’s resort to epithets and insults at least arguably served a communicative function similar to that recognized in Cohen, which the Court had decided the preceding year. Two of the four directly replicated Chaplinsky, involving angry citizens who had cursed an arresting officer (Gooding v. Wilson, 1972; Lewis v. New Orleans, 1972). These speakers, as had Chaplinsky and Cohen, used profanity to protest official action that they believed violated their individual rights. The two other cases tracked Cohen, but not Chaplinsky. They involved speakers who had used profanity in comments they had made at a public meeting (Brown v. Oklahoma, 1972; Rosenfeld v. New Jersey, 1972).

The best reading of the 1972 quartet might be that the Court has limited the meaning of fighting words, in practice if not in doctrine, to the utterance of face-to-face insults in speech contexts which make clear that the insult carries no First Amendment value. Such a limitation of the fighting words exception would be appropriate, but ironic. It would be appropriate, in light of Cohen, to trim Chaplinsky by limiting the fighting words exception to instances in which a speaker’s insults truly had been private, gratuitous, and therefore valueless in the First Amendment sense. If the 1972 quartet means that the category of fighting words is so limited, the ironic result is that Chaplinsky no longer holds in the factual context of that case, namely, when individuals insult government officials as criticism of their official conduct.

Hate Speech

Hate speech is an old problem that has generated First Amendment controversy and confusion in recent years. There is disagreement even over the precise contours of “hate speech” as a speech category, but a useful working definition includes speech that is insulting or threatening to members of groups that have suffered a history of discrimination in American society. It is a sobering fact of American history that there are a number of such groups, including ethnic minorities, women, gays and lesbians, Jews, and other religious minorities. Race, however, has marked the principal fault line in hate speech jurisprudence, as it has more generally in American society.

The Supreme Court took up the problem of hate speech for the first time shortly after deciding Feiner (Beauharnais v. Illinois, 1952). Beauharnais was president of a group called the White Circle League. He had organized the distribution of leaflets that called on the elected leaders of Chicago to resist the “invasion” of the white race by blacks. Not surprisingly for this genre of protest, the leaflets were laced with derogatory characterizations of African Americans. For example, they described African Americans as a group that engages in “aggressions, rapes, [and] robberies,” and that indulges in the use of marijuana. Beauharnais was prosecuted under a state law that made it illegal to publicly distribute any material that “portrays depravity, criminality, unchastity, or lack of virtue of a class of citizens of any race, color, creed or religion,” and that either “exposes the citizens of any race, color, creed or religion to contempt, derision or obloquy,” or “is productive of breach of the peace or riots.”

The Court in Beauharnais read the state statue as a criminal prohibition of “group libel” and regarded the derogatory statements in the leaflets as libelous of African Americans as a group. Because libel of an individual traditionally had not been regarded as within the freedom of speech, Justice Felix Frankfurter, writing for the majority, reasoned that libel of a group should not be protected either. He buttressed this conclusion by referencing the “tragic” history which demonstrated that “willful purveyors of falsehood concerning racial and religious groups promote strife and tend powerfully to obstruct the manifold adjustments required for free, ordered life in a metropolitan, polyglot community.” It would “deny experience,” Frankfurter wrote, to provide a First Amendment license for “extreme racial and religious propaganda” designed to produce a “powerful impact” on its audience. To do so, Frankfurter charged, would fashion the Constitution into a tool for the incitement of racial and religious violence that denied members of targeted groups “their equal right to the exercise of their liberties.”

Justice Frankfurter’s eloquent majority opinion triggered a strong response from Justice Black, one of four dissenters in Beauharnais. Black began by noting that Beauharnais’s leafleting was “a genuine effort to petition [his] elected representatives.” The freedom to so petition, he reminded, has a long history in Anglo-American constitutionalism, and is itself explicitly protected by the First Amendment. For Black, “Every expansion of the law of criminal libel … to punish discussions of matters of public concern means a corresponding invasion of the area dedicated to free expression by the First Amendment.” Beauharnais was not Chaplinsky, he argued. The leaflets, unlike personal, face-to-face insults, constituted “arguments on questions of wide public interest and importance.” In Black’s view, Beauharnais had been convicted “for publicly expressing strong views in favor of segregation.” He thus found the majority guilty of “sugar-coating … [a] system of state censorship.”

The analyses of Justices Frankfurter and Black in Beauharnais anticipated the basic positions for and against the constitutional legitimacy of hate speech regulation in contemporary free speech commentary. The five-four split on the Court in Beauharnais foreshadowed as well, the deep fissure the controversy over hate speech has opened among First Amendment scholars. But the Beauharnais decision itself, although never overruled, almost certainly has not stood the test of time. The Court itself has undermined the central position of Beauharnais—that group libel is unprotected because individual libel is unprotected—by deciding that defamatory speech is generally entitled to at least some measure of First Amendment protection (Gertz v. Robert Welch, Inc., 1974; New York Times v. Sullivan, 1964). The Court also has held outright that the First Amendment disallows defamation claims arising from statements that, like group libel, do not identify individuals (New York Times).

With the erosion of Beauharnais‘s grounding on defamation law, the exclusion of hate speech from freedom of speech must be justified by its own categorical balancing analysis. It certainly is possible, citing the history to which Justice Frankfurter referred in Beauharnais, to make the Chaplinsky case that the harm associated with hate speech clearly outweighs its communicative value. The alignment with Chaplinsky is especially powerful when hate speech, like a fighting word, is directed personally and face-to-face to the member of a group that historically has suffered such verbal abuse. But when hate speech is employed by a public speaker in an effort to persuade an audience to a point of view, the more fitting analogy is the hostile audiences precedent. That doctrine, as we have seen, grants strong First Amendment protection to speakers who provoke or offend their audience by the content of their ideas or the language of their expression.

The difficulty of outlawing hate speech outside the fighting words context was highlighted in the best-known judicial encounter with this phenomenon. In 1970, the National Socialist Party of America, better known as the American Nazi Party, announced plans for a demonstration in Skokie, Illinois, a small, suburban village just north of Chicago. Skokie has a significant Jewish community, and many of its residents at the time were Holocaust survivors. The Skokie demonstration was to be one of a series that the Party had planned for various Jewish communities.

All of the makings for an explosive confrontation were present in Skokie. Party members planned to wear uniforms that mimicked those of the Nazi storm troopers of the Third Reich. Many residents of Skokie objected strongly to their city being selected as a site for a Nazi demonstration, and plans for a potent counter-demonstration quickly took shape. Meanwhile, at Skokie’s request, a state court issued an injunction that prohibited anyone, while within the village, from, among other things, “parading” in Nazi uniforms and “displaying the swastika.” The Illinois Supreme Court, however, quickly overturned the injunction as a violation of the First Amendment (Village of Skokie v. National Socialist Party of America, 1978). The deeply offensive nature of these Nazi symbols for many Skokie residents, the court ruled, could not justify their banishment. Such symbols “spoke,” and even though what they expressed was anathema to almost everyone in and out of Skokie, the First Amendment protected this expression to the same degree it protects those who speak out against Fascism and anti-Semitism. (“Symbolic expression” is discussed in the final section of this chapter.)

While the injunction suit was pending, Skokie enacted several ordinances in a fallback effort to block the Nazi march. One of those ordinances, naturally enough, tracked the Illinois group libel law that the Supreme Court had upheld in Beauharnais. It outlawed the dissemination of material that intentionally “promotes and incites hatred against persons by reason of their race, national origin, or religion.” The leader of the American Nazi Party successfully challenged the new Skokie ordinances in federal court (Collin v. Smith, 1978). Consistently with both subversive advocacy and hostile audiences precedent, the federal court of appeals ruled that the mere tendency of a speech act to produce violence was not sufficient to justify governmental suppression. Neither the unacceptability of Nazi ideology nor the torment their demonstration surely would cause many Skokie residents, particularly Holocaust survivors, justified the village’s efforts to block the demonstration. In the process of upholding the First Amendment right of the Nazis to march in Skokie, the court of appeals made clear its belief that Beauharnais had long since been overwhelmed by contemporary free speech jurisprudence. The Supreme Court, for its part, remained mum by declining to review the case (Smith v. Collin, 1978). In an ironic (some would say, perverse) dénouement of this controversy, the American Nazi Party, having won the right to march in Skokie to the anguish of many, abruptly cancelled the demonstration. About twenty-five Party members instead held a rally in a Chicago park, without serious incident.

The Skokie litigation has become a symbol of First Amendment jurisprudence, for those who celebrate as well as for those who question the strength of the American commitment to free speech. Indeed, one constitutional scholar has based a general theory of justification for freedom of speech on the Skokie experience. The importance of the Skokie litigation for the problem of hate speech, lies in what did not happen: neither the state court nor the federal court revised conventional free speech jurisprudence in order to allow Skokie to outlaw the expressive activity of a hate group. Instead, the courts held true to the central principle that all ideas have equal status under the First Amendment, and that the most hateful messages are to be protected with the same vigor as the most enlightened discourse. As Justice William J. Brennan, Jr., writing for the Court in the equally controversial decision that protected flag-burning as a First Amendment right (Texas v. Johnson, 1989), proclaimed:

If there is a bedrock principle underlying the First Amendment, it is that the government may not prohibit the expression of an idea simply because society finds the idea itself offensive or disagreeable…. The First Amendment does not guarantee that … concepts virtually sacred to our Nation as a whole—such as the principle that discrimination on the basis of race is odious and destructive—will go unquestioned in the marketplace of ideas.

Skokie was the dog that did not bark.

The Skokie courts’ treatment of hate speech as simply an example of provocative speech contrasts sharply with the more recent decision of the Supreme Court in R.A.V. v. City of St. Paul (1992), which seemed to especially protect hate speech from governmental restriction. R.A.V. involved another symbol that spoke—a cross-burning on the residential property of an African American family. The perpetrators were convicted of violating a city ordinance that prohibited the use of symbols that one should know “arouses anger, alarm, or resentment in others on the basis of race, color, creed religion or gender.” The ordinance explicitly mentioned “a burning cross” as an example of such a prohibited symbol. The state supreme court proved itself willing to do what the Skokie courts had not done: It bent free speech doctrine in order to accommodate the ordinance and to validate the conviction. The state high court ruled that the ordinance was limited to hate speech that also qualified as fighting words, which as we have seen, the government is free to prohibit. That ruling was a stretch, though, because the ordinance, by its own terms, was not limited to speech that was delivered as a face-to-face insult.

The Court was unanimous in its decision that St. Pauls’s hate speech ordinance violated the First Amendment, but they split five-four on the rationale. Four of the justices tracked the approach of the Skokie courts by applying conventional provocative speech doctrine. To these four, the state supreme court had stretched too far. They found the ordinance to be invalid precisely because it was not limited to hate speech uttered as fighting words. The five-justice majority, however, accepted the state court’s interpretation, but held that the ordinance, even as limited to fighting words, violated the First Amendment. This was because the ordinance, according to the majority, targeted hate speech for special restriction because city officials disapproved of the message of hatred that it conveyed.

Although they reached opposite conclusions, the state supreme court’s opinion justifying the ordinance and Justice Antonin Scalia’s majority opinion invalidating the ordinance mirrored each other. While the state supreme court stretched the fighting words category in order to uphold the hate speech ordinance, the Supreme Court majority undermined the logic of categorization in an apparent effort to inhibit governments from targeting hate speech for any special restriction. As we have seen, the Court developed the categorization principle in order to allow the state to prohibit certain types of speech, such as fighting words, on the ground that their harm overwhelmed their First Amendment value. The justices had never before invalidated a governmental prohibition of expression that it had found to be within an otherwise unprotected category of speech. The R.A.V. majority decided, however, that the selective regulation of hate speech within the otherwise unprotected category of fighting words constituted a message-based restriction that was inherently censorial, and for that reason, unconstitutional.

The Court’s conclusion in R.A.V. that a hate speech prohibition which is limited to fighting words constitutes government censorship is highly doubtful, however. Justice Scalia acknowledged in his majority opinion that there was “no significant danger” of official censorship when a selective, content-based restriction of speech that otherwise is unprotected was based “entirely [on] the very reason the entire class of speech … is proscribable.” As the four justices who disassociated themselves from the majority opinion in R.A.V. emphasized, Scalia’s acknowledgment fit a hate speech restriction that was limited to fighting words. Surely, a government might conclude that personal insults based on the race, religion, gender, or sexual orientation of the recipient are more harmful and are more likely to trigger violence, and thus are more appropriate for official prohibition, than are other types of insults that a recipient is more likely to brush off.

In recent decisions, the justices have retreated from the implication in R.A.V. that hate speech enjoys some special immunity from selective governmental restrictions within otherwise unprotected expression or activity. Just one year after deciding R.A.V., in fact, the Court upheld a typical hate crime statute against a First Amendment challenge (Wisconsin v. Mitchell, 1993). A Wisconsin law authorized increasing the prison terms of perpetrators who had selected their victims “because of [their] race, religion, color, disability, sexual orientation, national origin or ancestry.” A broad reading of R.A.V. would have made such sentence enhancements vulnerable to the claim that they turned on the viewpoint or ideology of the perpetrator, rather than on the severity of the criminal action per se. If a racial fighting word was not more harmful than a nonracial fighting word, as R.A.V. seemed to have held, could it be said that a racially motivated attack on a victim was more harmful than one that was not so motivated? The answer in Mitchell, surprisingly, was, “yes.” Even more surprising was the Court’s explanation, which adopted the reasoning of the four justices who had disagreed with the R.A.V. majority opinion. The Court wrote in Mitchell, “[T]he Wisconsin statute singles out for enhancement bias-inspired conduct because this conduct is thought to inflict greater individual and societal harm. For example,… bias-motivated crimes are more likely to provoke retaliatory crimes, inflict distinct emotional harms on their victims, and incite community unrest…. The State’s desire to redress these perceived harms provides an adequate explanation for its penalty-enhancement provision over and above mere disagreement with offenders’ beliefs or biases.”

If this explanation was adequate to justify the hate crime law in Mitchell, why was it insufficient to support the hate speech law in R.A.V.? The Court answered this awkward question by invoking the speech-conduct distinction. R.A.V., the Mitchell Court explained, applied only to governmental restrictions on speech, and not to “conduct unprotected by the First Amendment” (emphasis added). A sentence-enhancement provision keyed to the racial motivation of the perpetrator was not different, the Court reasoned, from the many federal and state laws that prohibit harmful actions, such as the denial of employment or housing, when they are taken with a discriminatory purpose. Indeed, a perpetrator’s motivation for committing a criminal offence, the Court observed, is a common factor in sentencing.

Although the justices in Mitchell provided a convincing account of why sentence enhancement for hate crimes did not violate the First Amendment, they were less successful in distinguishing R.A.V. The speech-conduct distinction is of limited utility when the speech (fighting words), like the underlying conduct in Mitchell, is generally subject to prohibition without violating the First Amendment. It therefore is no surprise that the facile R.A.V./Mitchell distinction did not hold when the justices returned to the problem of cross burning in Virginia v. Black (2003). On the surface, Black might seem consistent with R.A.V. The justices in Black invalidated a Virginia law prohibiting cross-burning undertaken with “intent to intimidate a person or group of persons.” But the Court’s justification for its ruling undercut the R.A.V. rationale.

The Court in Black began by confirming what had been suggested in a couple of earlier cases (see, e.g., Watts v. United States, 1969), that “true threats,” like fighting words, constitute a category of expression that states may outlaw without violating the First Amendment. The Court thus announced itself willing to uphold a law that was limited to prohibiting cross-burning as a true threat, that is, with the intent “to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals.” The problem with the Virginia law was that it was not so limited because it permitted a jury to find the requisite intent from the mere fact of cross-burning. In order to qualify as a true threat, the Court held, the state had to prove, rather than presume, that the cross-burner intended to communicate a threat to a particular person or to a distinct group of persons.

This was precisely the line of reasoning advanced by the four justices who had disassociated themselves from the majority opinion in R.A.V. Those justices had voted to invalidate the hate speech ordinance not because it targeted for prohibition only those fighting words that constituted hate speech (as had the majority), but because the prohibition was not limited to speech that fell within the unprotected category of fighting words. They had insisted as well that a law limited to prohibiting hate speech within an otherwise unprotected category of speech should be constitutional. That position became the majority view in Black. The Court in Black explained, “The First Amendment permits [the government] to outlaw cross burnings done with the intent to intimidate because burning a cross is a particularly virulent form of intimidation. Instead of prohibiting all intimidating messages, [the government] may choose to regulate this subset of intimidating messages in light of cross burning’s long and pernicious history of impending violence.”

Although the Court has not had the opportunity to reconsider whether governments may outlaw hate speech more generally within otherwise unprotected speech categories (such as fighting words), it is quite possible that the majority opinion in R.A.V. has not withstood the Court’s second and third looks at the hate speech problem in Mitchell and in Black. In a brief, separate opinion in R.A.V., Justice Harry A. Blackmun had suggested the possibility that the R.A.V majority opinion would not “significantly alter First Amendment jurisprudence but, instead, [would] be regarded as an aberration—a case where the Court manipulated doctrine to strike down an ordinance whose premise it opposed, namely, that racial threats and verbal assaults are of greater harm than other fighting words.” It was also possible, and “even more regrettable,” Justice Blackmun believed, that “the Court ha[d] been distracted from its proper mission by the temptation to decide the issue over ‘politically correct speech’ and ‘cultural diversity.'” If Blackmun’s surmise was correct, it might be that the facts of Mitchell and Black had convinced enough of the R.A.V. majority that the individual and social harms caused by hate speech and hate crimes are real, and that governmental efforts to target such activity are not necessarily censorship, at least when those efforts are confined to speech and conduct that is not generally afforded First Amendment protection.

On this reading, hate speech remains simply a type of provocative speech. It is thus subject to neither more nor less First Amendment protection, or governmental restriction, than are other forms of provocative speech.

Sexual Expression

Although the Supreme Court’s struggle with the problem of subversive advocacy for most of the twentieth century is the central narrative of the protection of freedom of speech in the United States, it is oddly revealing of American culture that efforts at government control over sexual expression run a respectable second place. But perhaps the American preoccupation with sexual expression is not so strange. As Justice William J. Brennan observed, sex is “a great and mysterious motive force in human life” that has “indisputably been a subject of absorbing interest to mankind through the ages” (Roth v. United States, 1957). Justice Brennan’s observation provides not only the rationale for First Amendment protection of much sexual expression, but also, paradoxically, the motivation that has driven government officials over the years to target sexual expression for special restriction. Sexual expression, no less than political speech or other forms of social commentary, at times provides a powerful means for individuals to explore, as well as to challenge, the central cultural commitments and values of their society. And sexual expression, no less than other forms of speech that we value highly, can be deeply subversive of those values and thus offer a tempting target for official censorship.

Although government restrictions on sexual expression have taken a variety of forms, anti-obscenity laws have the longest pedigree. Today, the states generally, together with the United States, outlaw the production, distribution, and exhibition of “obscenity,” which often is described as “hard core pornography.” In recent years, the regulatory focus on sexual expression has broadened to include laws against child pornography, special zoning restrictions on the venues for sexual expression, and laws that shield children from exposure to sexual expression. Regardless of the nature of the restriction, the Court has found itself in the difficult position of balancing its general commitment to free speech, including sexual expression, with its belief that states should possess greater power to restrict sexual expression than other types of speech. The jurisprudence that has resulted from the Court’s balancing act has been at least as controversial as any other area of free speech.


Although obscenity has been prevalent since the dawn of western civilization—obscenity was commonplace, for example, in ancient Greece—it was not a great source of official concern in antebellum America. Only four states had anti-obscenity laws at the beginning of the Civil War. Although a number of states at the time regarded obscenity as an offense at common law, there were no serious governmental efforts to suppress obscenity. As recounted in our consideration of the history of free speech in the United States in Chapter 1, that official complaisance changed abruptly with the emergence of Anthony Comstock’s crusade against the “monstrous evil” of obscenity in the last quarter of the nineteenth century. Because the Comstock campaign had pretty much played out by the beginning of World War I, it was the subversive advocacy cases that arose from that war, rather than obscenity convictions, that ushered in modern free speech jurisprudence. But while the sense of urgency faded, public concern over obscenity never disappeared. Ever since the Comstock era, the issue of obscenity has ebbed and flowed in the national consciousness.

It is surprising, in the light of this history, that the Supreme Court did not resolve the question whether obscenity was within the freedom of speech protected by the First Amendment until 1957 (Roth v. United States). In the prior century, the justices had brushed off a First Amendment challenge to the Comstock Act, but as we have seen, this was before the Court had activated freedom of speech as a fundamental right (Ex Parte Jackson, 1878). The modern Court in Chaplinsky v. New Hampshire (1942) also assumed that obscenity was among the speech categories that “have never been thought to raise any Constitutional problem,” but that assumption was not tested until the Court was petitioned to set aside an obscenity conviction arising from the publication of Edmund Wilson’s Memoirs of Hecate County (Doubleday & Co. v. New York, 1948). The Court affirmed the conviction, but only by default because the justices divided equally on the question.

The Court in Roth finally made official the exclusion of obscenity from First Amendment protection. The anti-obscenity statute at issue placed the justices in a position that had become familiar from their experience in subversive advocacy and provocative speech cases. Congress and state legislatures outlawed obscenity on the same general reasoning that had prompted statutes prohibiting subversive advocacy and provocative speech: Legislators believed that such expression produced a social harm that was their responsibility to prevent. The difference lay in the elusiveness of the claim of harm for obscenity. The challengers in Roth focused their attack on that vulnerability, arguing that anti-obscenity laws were incompatible with the First Amendment because the government had failed to prove that obscene materials created a “clear and present danger of antisocial conduct.” There is no indication in Justice Brennan’s majority opinion in Roth that the government had shouldered that burden. Brennan finessed that shortcoming by categorizing obscenity out of First Amendment protection. Because the First Amendment does not protect obscenity, Brennan reasoned, the government had a free hand in outlawing expression that fell within that category, even without proof of harm. This stance seemed inconsistent with the categorization method the Court had outlined in Chaplinsky, which provided for the exclusion of a particular speech category from First Amendment protection only if the justices were satisfied that the harms associated with that speech greatly exceeded its communicate value. Instead of undertaking such a cost-benefit balancing, Justice Brennan justified the exclusion of obscenity from First Amendment protection essentially by invoking “the universal judgment that obscenity should be restrained.” That judgment was manifested to the Court’s satisfaction by historical assumptions and by contemporary anti-obscenity laws enacted by Congress and by all of the states. To the justices, the “universal” outlawing of obscenity in the United States registered an authoritative cultural judgment that obscenity was “utterly without redeeming social importance,” and for that reason alone, was outside the conventional American understanding of the freedom of speech.

Unlike the Court’s initial encounter with subversive advocacy, the justices in Roth were mindful of the constitutional risks of excluding obscenity from First Amendment protection. They tried to minimize those risks by defining obscenity in a way which would ensure that sexual expression generally would remain protected, subject only to a narrow exception for depictions of sex that went beyond the pale. To the Roth Court, the distinctive characteristic of obscenity was its appeal to the “prurient interest” of its audience, which, incorporating a statement from the American Law Institute, the justices defined as “a shameful or morbid interest” in sexuality. Sexual expression untainted by the dominance of such an appeal, the Court emphasized, remained protected. “[S]ex and obscenity,” Justice Brennan advised in his majority opinion, “are not synonymous,” and the mere “portrayal of sex … is not itself sufficient reason to deny material the constitutional protection of freedom of speech and press.” In Roth, then, the Court attempted to define a First Amendment position that would allow an “unfettered interchange of ideas” exploring human sexuality, regardless how “unorthodox” or “controversial,” while preventing what, in the Roth view, amounted to an abuse of that freedom, namely, sexual expression intended simply to “excite lustful thoughts.”

The Court’s decision in Roth proved to be less a settlement of the obscenity issue than the beginning of a long, contentious argument. The Court’s commitment to free sexual expression was tested almost immediately after the Roth decision (Kingsley Int’l Pictures Corp. v. Regents of the University of the State of New York, 1959). In Kingsley, the justices overturned a state’s refusal to license the showing of a film version of Lady Chatterley’s Lover. Although the film was not obscene under the Roth definition, the state nevertheless had banned it because “the whole theme” of the film was “immoral.” According to the state, the film depicted “adultery as a desirable, acceptable and proper pattern of behavior.” The Court ruled that the state’s conception of “thematic obscenity” was no substitute for the Roth definition of obscenity, and accordingly, that it did not provide a basis for prohibiting sexual expression. By shifting emphasis from “the manner of [the] portrayal” of a sexual relationship to the idea communicated by the portrayal, the Court in Kingsley held, the state had struck “at the very heart” of the freedom of speech. Thematic obscenity constituted official censorship, pure and simple. It was no answer that the idea advocated by the film defended activity—adultery—that itself violated state law, for as the Court reminded, subversive advocacy doctrine protected the mere advocacy of unlawful action absent a clear and present danger that the film would immediately incite the action. Kingsley thus delivered on Roth‘s promise that obscenity doctrine would function as a relatively narrow exception to an otherwise robust freedom of individuals to express “[a]ll ideas” on sexual subjects that have “even the slightest redeeming social importance,” regardless how “unorthodox” or “controversial.”

Ironically, even as the justices united to prevent states from using such dodges as thematic obscenity to bypass the restrictions Roth had placed on the scope of obscenity regulation, their allegiance to the Roth definition itself unraveled. This should have come as no great surprise. Only five justices—the slimmest of majorities—had signed onto the Roth definition in the first place. When, nearly a decade later, Justice Brennan tweaked that definition, only two other justices joined him (Memoirs v. Massachusetts, 1966). The justices went their separate ways, agreeing to disagree on the obscenity issue. The Brennan group used the Memoirs variation of the Roth definition, while others invented their own definitions of obscenity. Another group of justices was disdainful of the entire definitional enterprise because they opposed the constitutionality of all anti-obscenity laws. The sense of exasperation that arose from the post-Roth doctrinal disarray was forever captured by Justice Potter Stewart when he wrote, “[P]erhaps I could never succeed in intelligibly [defining obscenity]. But I know it when I see it.” (Jacobellis v. Ohio, 1964).

In 1967, one year after Justice Brennan’s failed attempt in Memoirs to reconstruct a consensus on the Court by reassembling a constitutional definition of obscenity, the justices institutionalized their disagreement on the definitional question by launching the process of “redrupping” obscenity convictions (Redrup v. New York, 1967). In this process, each justice reviewed the expression at issue and determined, according to his own definitional understanding, whether or not it constituted obscenity. If five or more justices found the expression to be obscene, the Court affirmed the conviction without opinion. If five or more justices found the material not to be obscene, the Court reversed the conviction, again without opinion. As the Court later confessed, “The Redrup procedure … cast us in the role of an unreviewable board of censorship for the 50 states, subjectively judging each piece of material brought before us” (Miller v. California, 1973). Redrupping was the antithesis of principled constitutional adjudication, and it powerfully symbolized the breakdown of obscenity doctrine in the wake of Roth.

During the redrupping era, the justices, naturally enough, experimented with an alternative approach to categorization in their continuing effort to solve the obscenity problem. Labeled “variable obscenity,” this approach drew on Chief Justice Earl Warren’s concurring opinion in Roth, in which he insisted that the “conduct of the defendant” should be “the central issue” in obscenity cases, not the content of the defendant’s expression. “It is not the book that is on trial,” Warren observed, “it is the person.” Eschewing the strict categorization approach of Roth, Warren would have had the Court examine the defendant’s expression “in context.” It therefore would be appropriate, he argued, for the Court to reach different judgments regarding the same speech content in “different setting[s].”

The Court developed Chief Justice Warren’s concept of variable obscenity in the years following Roth as a contextual counterpoint to the otherwise dominant approach of categorization. In this line of cases, two variables seemed especially important to defining the context for evaluating allegedly obscene materials. The first variable was whether the defendant could be said to have engaged in “pandering.” The concern with pandering came directly from Warren’s concurring opinion in Roth, where the Chief Justice had suggested that obscenity prosecutions were best targeted against those who had engaged in “conduct” that amounted to “the commercial exploitation of the morbid and shameful craving for materials with prurient effect.” In Ginzburg v. United States (1966), the Court followed Warren’s suggestion, holding that “the commercial exploitation of erotica solely for the sake of their prurient appeal” could support an obscenity conviction even if the content of the expression, standing alone, did not satisfy the Roth definition. Ginzburg supplemented, but did not supplant, Roth categorization. The pandering variable broadened anti-obscenity laws to reach defendants who were engaged in otherwise protected sexual expression, but it did not limit the application of those statutes to panderers.

The other prominent variable that affected the Court’s review of obscenity convictions during the redrupping era was the presence of children in the audience. In Butler v. Michigan (1957), which the Court decided the same year as Roth, the justices invalidated a statute that prohibited the general distribution of sexual expression that, among other things, “manifestly tend[ed] to the corruption of the morals of youth.” The Court cautioned legislators that they could not “reduce the adult population … to reading only what is fit for children.” Butler, like Kingsley, reinforced the definitional approach of Roth by preventing states from expanding the field of illegal sexual expression beyond the bounds of the newly created category of obscenity. But just one year after Redrup, in Ginsberg v. New York (1968), the Court allowed states to outlaw the sale of non-obscene, sexual expression to minors. As with pandering, the child-audience variable allowed the states to expand the reach of anti-obscenity statutes to sexual expression that did not fit the constitutional definition of obscenity.

The subversive potential of variable obscenity for the Roth regime became apparent soon after the Ginzburg and Ginsberg decisions. In Stanley v. Georgia (1969), the Court for the first time drew on the contextual orientation of variable obscenity to reduce, rather than to expand, the constitutional range of anti-obscenity laws. Stanley had been convicted for possessing obscenity in his home. The “mere categorization” of material as obscene, the Court ruled, was “insufficient justification” for such an “unwarranted governmental intrusions into one’s privacy.” Context was all. Following Chief Justice Warren’s alternative method of evaluating obscenity regulation, the Court distinguished Stanley from Roth because of the conduct of the defendants, not the content of their expression. While Roth had engaged in the “public distribution” of obscenity, Stanley had simply possessed such materials at home.

The subversiveness of Stanley was not limited to its protection of the private possession of unprotected expression. The Court’s First Amendment rationale in Stanley undercut the constitutional grounding of Roth, and indeed, the legitimacy of the entire anti-obscenity project. Writing for the Court, Justice Thurgood Marshall directly challenged the Roth premise that obscenity could be pushed outside the freedom of speech because it lacked social value. He wrote, the “right to receive information and ideas, regardless of their social worth,… is fundamental to our free society” (emphasis added). Marshall rejected as well the paternalistic assumptions behind anti-obscenity laws in general as irreconcilable with free speech values. Marshall again: “If the First Amendment means anything, it means that a State has no business telling a man, sitting alone in his own house, what books he may read or what films he may watch. Our whole constitutional heritage rebels at the thought of giving government the power to control men’s minds.” Finally, Justice Marshall’s opinion for the Court in Stanley dismissed Georgia’s effort to justify its prohibition of the possession of obscenity as a hedge against “deviant sexual behavior or crimes of sexual violence.” Marshall noted the insufficiency of empirical support for any such showing, observing, “Given the present state of knowledge, the State may no more prohibit mere possession of obscene matter on the ground that it may lead to antisocial conduct than it may prohibit possession of chemistry books on the ground that they may lead to the manufacture of homemade spirits.” Marshall reminded that the First Amendment generally limits states to punishing individuals for what they do, not for what they read or view.

Although the Court in Stanley was careful to limit its holding to the constitutional protection of the “mere private possession of obscene material,” Justice Marshall’s withering rejection of the conventional justifications for anti-obscenity laws in general provided a platform for a reconsideration, and possible overruling, of Roth itself. Surely state efforts at “controlling a person’s private thoughts” were just as chilling outside the home. Similarly, the empirical support for the legislatively presumed linkage between obscenity and sexual misconduct was no more compelling when the reader or viewer of the material was outside the home. The Court nevertheless quickly drew back from the broadest implications of its opinion in Stanley, insisting that it “neither overruled nor disturbed the holding in Roth” (United States v. Reidel, 1971). But having answered the Stanley challenge, the Court once again found itself in an awkward doctrinal position, having held, essentially, that states may criminalize the sale of materials that individuals have a constitutional right to possess.

The doctrinal uncertainty introduced by variable obscenity in general, and by Stanley in particular, coupled with the erosion of judicial integrity that attended the Court’s redrupping of obscenity cases, undermined the coherency of the Roth regime. The status quo was unsustainable, and in 1973, the justices announced a new settlement of the obscenity issue (Paris Adult Theatre I v. Slaton (1973); Miller v. California (1973)). Paris and Miller were companion cases that interrelated tan-talizingly to hint that the Court might abandon Roth categorization entirely for the more contextual variable obscenity method. Paris concerned the application of an anti-obscenity law to a so-called “adult” movie theater that exhibited motion pictures to consenting adults only, while Miller involved the mailing of unsolicited advertisements containing obscene materials to unsuspecting recipients. One might distinguish the two cases on variable obscenity grounds as mirror images of each other. The method of distribution in Miller created a significant danger of offending the sensibilities of unwilling recipients and of exposure to children, two factors that the Court had recognized as significant variables when reviewing obscenity convictions. By contrast, the exhibitor in Paris avoided such judicial sensitivities by taking precautions to avoid exposure of obscene films to either unwilling adults or to children. And indeed, several justices, led by Justice Brennan, the author of Roth, used Paris to urge that states no longer should be permitted to prohibit obscene expression entirely, but rather should be restricted to regulating such expression to protect children or unwilling adults from exposure. A bare five-justice majority of the Court, however, pointedly ignored the factual differences between Paris and Miller and used the pair of cases to reaffirm their commitment to Roth.

The Court used Paris, the more difficult of the two cases, to reaffirm Roth‘s exclusion of obscenity from First Amendment protection. Given all that had transpired since Roth, it is not surprising that the majority’s justification extended beyond the earlier decision’s simple reliance on the history and uniformity of anti-obscenity laws in the United States. The Court now emphasized “the interest of the public in the quality of life and the total community environment” as a legitimate and sufficient basis upon which states might outlaw obscenity. The Paris majority analogized the spread of obscenity outside the privacy of the home to the emission of pollutants into the atmosphere. The result of this moral toxicity, according to the Court, was the corrosion of community standards. The justices recognized that the existence of moral pollution, unlike environmental pollution, could neither be proved nor measured. But they nevertheless insisted that local communities possessed a right “to maintain a decent society,” and that communities could “reasonably determine” that anti-obscenity laws served that purpose. The Paris Court’s identification of a community harm caused by obscenity responded to Stanley‘s challenge to the conventional justifications of obscenity regulation, even as it supported the limitation of Stanley to protecting the private possession of obscene materials.

In Miller, the Court drew on the community interest in preserving a moral environment identified in Paris to revise, or rather to augment, the Roth definition of obscenity. In their refinement of Roth, the justices attempted to make anti-obscenity laws usable by communities, while limiting the reach of those laws to “hard core pornography.” The Miller definition of obscenity incorporated the Roth limitation to material that is dominated by an appeal to the “prurient interest,” and added two additional requirements. To be obscene, the material also must depict or describe “specifically defined sexual conduct” in “a patently offensive way,” as well as lack “serious literary, artistic, political, or scientific value.” The Miller Court regarded its limitation of obscenity to depictions or descriptions of “hard core” sexual conduct “specifically defined” by statute as an important tightening of the Roth definition, in the interest of furthering freedom of speech. A loosening, however, complemented that constriction. While the Court in Roth expressed its commitment to protect “[a]ll ideas having even the slightest redeeming social importance,” the Miller definition upped the ante for sexual expression that otherwise qualified as obscenity by requiring that it possess “serious” value in order to escape that designation. The Court believed this loosening of the definition to be a significant improvement of Roth as well, in the interest of enabling state prosecutors to enforce anti-obscenity laws.

Perhaps the most important innovation of Miller was the stipulation that jurors, acting as a collective “average person” applying the standards of their community, would determine each element of the new definition of obscenity, with the exception of the serious value criterion (see Pope v. Illinois, 1987). The justices thought it “an exercise in futility” to devise a national community standard for obscenity, explaining, “our Nation is simply too big and too diverse for this Court to reasonably expect that such standards [as “prurient interest” and “patently offensive”] could be articulated for all 50 States in a single formulation.” The Court continued, “It is neither realistic nor constitutionally sound to read the First Amendment as requiring that the people of Maine or Mississippi accept public depictions of conduct found tolerable in Las Vegas, or New York City.” Yet, given the demands of national marketing, the effect of allowing local communities to apply the Miller requirements through the lens of their own standards risked denying the people of Las Vegas and New York City access to sexual expression that is unacceptable in Maine or Mississippi. In resolving this dilemma, the Court had a choice to make, and the justices in Miller, as they had in Paris, chose to err on the side of empowering local communities to block the entry of obscene sexual expression.

The Court has made clear that each jury supplies the content of the Miller standards on an ad hoc basis, drawing on its “own understanding of the tolerance of the average person in their communit[ies]” (Smith v. United States, 1977). The depth of Miller‘s commitment to the primacy of jury determinations in obscenity cases was strikingly apparent when the Court prohibited state legislatures from defining the content of the community standards to be applied by the juries of their respective states (Smith v. United States, 1977). In deciding obscenity cases, jurors uniquely represent, because they re-present, the community.

The community-standards overlay of the Miller criteria was as important institutionally as it was substantively. Miller operates as an inverse Redrup: it prescribes that juries, rather than the justices, shoulder the primary responsibility of distinguishing protected sexual expression from unprotected obscenity. It is no accident that the Court’s docket of obscenity cases has reduced markedly since Miller. Indeed, shortly after deciding Paris/Miller, the Court ruled, again five-four, that even prosecutions for violation of national anti-obscenity laws would be determined by juries applying the local standards of the communities from which they were drawn (Hamling v. United States, 1974).

The jury-administered, community-standards orientation of the Miller definition of obscenity fits the Paris community-morality justification for anti-obscenity laws. If such laws protect the level of decency and moral tone of life in a community, as Paris would have it, there is a symmetry in allowing randomly selected representatives of each community to determine for themselves which material crosses the line separating legitimate expression and unprotected obscenity. But because Miller‘s reliance on jurors as community filters of sexual expression enables majoritarian control over freedom of speech, such a system is in severe tension with conventional First Amendment practice. A core function of the First Amendment, after all, is to protect expression that challenges majoritarian sentiment. Indeed, one might say that modern free speech jurisprudence began with Holmes’s recognition that the impulse to censor is a natural, human reaction to expression that challenges deeply held community values. The typical First Amendment concern over the possibility of majoritarian censorship, moreover, is especially acute with respect to obscenity because criteria such as “patently offensive” and “prurient interest” are freighted with subjectivity. Miller, thus, threatened to “solve” the redrupping problem by substituting local juries for the Supreme Court as boards of censorship.

Yet, the experience of the last quarter century has shown that juries generally have handled their Miller responsibilities remarkably well. There has been no upsurge in obscenity convictions in the thirty years following the Paris/Miller resettlement. Perhaps this development has not been so remarkable after all. Jurors, no less than judges and other Americans, have internalized the cultural commitment to freedom of expression, and they have oriented their decision making accordingly. Their administration of Paris/Miller thus reflects a notable convergence of legal and social norms governing sexual expression. In addition, obscenity doctrine creates a web of constraints on juries that provides judges the authority to rectify the inevitable instances of majoritarian overreaching. The justices emphasized that juries would not enjoy carte blanche in obscenity cases when, just one year after Paris/Miller, they overturned a verdict by a Georgia jury finding the mainstream film Carnal Knowledge to be obscene (Jenkins v. Georgia, 1974). Just as the Court had emphasized in Roth that sex could not be equated with obscenity, the justices in Jenkins reemphasized that nudity alone, without “patently offensive” depictions of “hard core sexual conduct for its own sake,” could not rob sexual expression of First Amendment protection. Jenkins thus provided a reassuring, doctrinal counterpoint analogous to that of Kingsley‘s protection of so-called thematic obscenity in the wake of Roth, namely, that the Court, in the language of Miller, would invoke its “ultimate power” to safeguard the Constitution by not allowing juries to roam beyond the “carefully limited” boundaries of obscenity in order to censor sexual expression more generally.

The Court’s opinions in Paris/Miller brought a measure of peace to the longstanding constitutional controversy over obscenity. At the least, the justices, by their reliance on juror applications of community standards to decide obscenity cases, managed largely to extricate themselves from these conflicts. In a larger sense, however, the promise of a more general settlement with respect to sexual expression that Roth/Paris/Miller had promised has not materialized. The original Roth settlement, which held that the First Amendment generally protects sexual expression, with the exception of material that falls within a narrowly confined category of obscenity, has been challenged by two developments. The most fundamental challenges have come from attempts to add additional categories of unprotected sexual expression to the traditional exception for obscenity. The most prominent example of this challenge has been the Court’s addition of child pornography to the unprotected list. A second, subtler challenge has come from the efforts of government officials to impose restrictions on sexual expression short of outright prohibitions. These developments are briefly discussed, in turn.

Child Pornography and the Border Security of Protected Sexual Expression

Just as the Court has described “hard core pornography” as the “obscenity” that is unprotected by the First Amendment, it may be said that obscenity itself is the core of unprotected sexual expression. Sexual expression need not be obscene, however, in order to be denied First Amendment protection. This became apparent when the Court accepted the constitutionality of child pornography laws in New York v. Ferber (1982). Ferber had been charged with violating New York’s laws prohibiting obscenity and child pornography for his role in selling films depicting young boys masturbating. The jury acquitted Ferber on the obscenity count, but convicted him of promoting child pornography. Based on the definitional approach to obscenity begun in Roth and cemented in Paris/Miller, one might have thought that because the films were not obscene, the First Amendment disallowed outright prohibition of them because of their sexual content. The Ferber jury, by operation of Miller, must have found that the films were not dominated by an appeal to the prurient interest, or that the films did not depict sexual conduct in a patently offensive manner, or that the films possessed serious value. At least one such finding had saved the films from the unprotected category of obscenity.

The justices, however, used Ferber to revise, once again, their settlement over sexual expression. While Roth/Paris/Miller had found obscenity to be an unprotected subcategory of the larger category of sexual expression, the Court now held, those decisions did not foreclose the creation of additional unprotected subcategories. In Ferber, the Court held that child pornography lacked First Amendment protection as well. Moreover, the Ferber Court made clear that states would have “greater leeway” in restricting child pornography than Roth/Paris/Miller allows them with respect to obscenity.

The Court justified the broader prohibitive power of the government to address child pornography by invoking Chaplinsky’s cost/benefit assessment. The societal costs of child pornography, which center on the harm suffered by the child subjects of such depictions, are indeed profound. As the Court in Ferber explained, “[t]he prevention of sexual exploitation and abuse of children constitutes a governmental objective of surpassing importance.” By contrast, the moral-environmental harms the Paris Court associated with obscenity seem attenuated and speculative. The Court’s conclusion in Ferber was irresistible: the governmental interest underlying child pornography laws is both more “particular” and more “compelling” than that which justifies anti-obscenity laws.

The focus of child pornography statutes on the protection of child subjects not only accounted for the heightened social interest at stake, but also for a diminished judicial concern over the free speech values potentially threatened by such laws. Although the Court in Ferber acknowledged that child pornography laws, like anti-obscenity laws, run “the risk of suppressing protected expression,” those risks seemed less pressing. The Court explained:

The value of permitting live performances and photographic reproductions of children engaged in lewd conduct is exceedingly modest, if not de minimis. We consider it unlikely that visual depictions of children performing sexual acts or lewdly exhibiting their genitals would often constitute an important and necessary part of a literary performance or scientific or educational work…. The First Amendment interest is limited to that of rendering the portrayal somewhat more “realistic” by utilizing or photographing children.

Free expression of ideas or practices concerning juvenile sexuality can be accommodated readily by the use of older performers or models able to pass for children, subject of course, to anti-obscenity constraints. Accordingly, the Court noted, child pornography laws do not “censor[] a particular literary theme or portrayal of sexual activity.” They simply prohibit the use of child subjects.

To the justices, Ferber‘s cost/benefit analysis revealed the inadequacy of the Miller test in addressing child pornography. While obscenity and child pornography are each subcategories of sexual expression, their defining characteristics fundamentally differ. The key signifier of obscenity is a work’s appeal to the prurient interest of the audience. By contrast, the distinctive quality of the prohibition of child pornography is its focus on the harm experienced by the child subjects. According to the Ferber Court, the presence or absence of an appeal to the prurient interest of the audience “bears no connection to the issue of whether a child has been physically or psychologically harmed in the production of the work.” That observation applies to another Miller criterion—the offensiveness of the depiction. Moreover, community standards, the framework within which juries evaluate “prurient interest” and “patent offensiveness” for obscenity determinations, likewise are out of place in child pornography cases.

As to the serious value of the work, the Ferber Court suggested, somewhat startlingly to those accustomed to the presence of this First Amendment safety valve of obscenity jurisprudence, that “a work which, taken on the whole, contains serious literary, artistic, political, or scientific value may nevertheless embody the hardest core of child pornography.” But the Court hedged that seeming disavowal of the value element of Miller with regard to child pornography. When the Ferber Court described how the Miller test would be “adjusted” to fit child pornography, the justices explicitly deleted only Miller‘s “prurient interest” element and its requirement that depictions of sexual conduct be “patently offensive.” While the justices also decided that “the material at issue need not be considered as a whole,” they did not address the relevancy of the value of the depiction itself.

The Court in Ferber ultimately finessed the issue of whether socially valuable material that otherwise fits the definition of child pornography might retain First Amendment protection. The justices had in mind two possibilities—photographs appearing in a medical text or in National Geographic. They believed, however, that such cases would be rare and decided to save them for another day. A child pornography law, the Ferber Court concluded, is “the paradigmatic case of a state statute whose legitimate reach dwarfs its arguably impermissible applications.” While the Court has not yet definitively resolved whether the value of material that otherwise fits the definition of child pornography would redeem the expression sufficiently to bring it within the protective embrace of the First Amendment, the strong suggestion throughout the cases is that it does not.

The Court’s reluctance in Ferber to endorse explicitly even the serious value criterion of Miller highlights the expansive potential of the child pornography exception to First Amendment protection. As in Roth, though, the justices in Ferber acknowledged their responsibility to limit the category of child pornography in order to accommodate free speech values. Early in its opinion, the Ferber Court described child pornography laws as legislation “directed at or limited to depictions of sexual activity involving children.” The Court’s essential statement on the First Amendment limits of such laws required that “the state offense” be tied to “the nature of the harm,” and thus be “limited to works that visually depict sexual conduct by children below a specified age.” (Emphasis in original.) The Ferber Court emphasized that non-obscene material which does not involve “live performance or photographic or other visual reproduction of live performances” of children remained protected by the First Amendment. As to the appropriate statutory definition of “children,” the Court in Ferber appeared to embrace state laws that included those under the age of eighteen.

The Ferber opinion reflects clearly that the Court saw, and approached, child pornography as a different and far more ominous social evil than obscenity. While the justices have been consistent in their exclusion of obscenity from First Amendment protection, they have always understood anti-obscenity laws as a troubling constitutional project. In Ferber, the justices revealed no trace of the ambivalence so prevalent in their obscenity decisions. Paradoxically, the convincing constitutional support for child pornography laws may pose more of a threat to the cultural and constitutional values of free expression than that presented by anti-obscenity laws. It is revealing that the Ferber Court, certain of the cultural and constitutional grounding of governmental efforts to eradicate child pornography, emphasized the reach, not the limits, of that regulatory project.

The wholehearted endorsement of child pornography laws in Ferber laid the groundwork for extending governmental prohibitions significantly beyond that approved in the Ferber decision itself, not to mention that envisioned in Roth/Paris/Miller. Such an outward push of the doctrine legitimating state restrictions on child pornography stood in marked contrast to the post-Roth and post-Paris/Miller experiences. Recall that the Court quickly followed Roth with a decision disallowing states from prohibiting so-called thematic obscenity, ensuring that the expression of ideas challenging conventional sexual mores would continue without the threat of government prosecution. Recall as well that soon after Paris/Miller, the Court set aside a jury verdict finding the film Carnal Knowledge obscene, ensuring that overly sensitive juries would not expand obscenity beyond “hard core pornography” to include simple nudity. In child pornography jurisprudence, by contrast, the Court’s early moves after Ferber projected in the opposite direction.

For example, the justices exhibited no hesitation in affirming an expansive reading of Ferber that allowed states to prohibit not only depictions of children engaged in sexual conduct, but also depictions of nude children, at least where there is a “lewd exhibition” of or a “graphic focus” on genitalia (Osborne v. Ohio, 1990). The movement from sexual conduct to nudity, of course, is precisely what the Court had rejected in obscenity regulation after Paris/Miller. In addition, because of “the importance of the State’s interest in protecting the victims of child pornography,” the Court allowed prosecutions for the simple possession of child pornography in the privacy of the home, another extension the justices have prohibited regarding obscenity (Osborne v. Ohio, 1990).

The Court finally identified a firm boundary limiting future extensions of Ferber in Ashcroft v. The Free Speech Coalition (2002). Aschcroft involved the longest, most suggestive legislative stretch of Ferber. In the Child Pornography Prevention Act of 1996, Congress outlawed so-called “virtual child pornography,” that is, visual images that appeared to be depictions of children engaging in sexual conduct. Such apparent depictions included computer generations, which were the primary target of the law, as well as images of adult models portraying minors. The prohibition of virtual child pornography reached beyond the explicit limits of Ferber, which because of its focus on the harm suffered by the child subjects exploited in the production of child pornography, had limited the unprotected category of child pornography to depictions of actual, not simulated, children. In order to uphold the new Act, the Ashcroft Court believed, it would have been “necessary” to make “virtual child pornography … an additional category of unprotected speech,” and this the justices declined to do. The very concept of virtual child pornography troubled the justices. Such a category, unlike Ferber child pornography, would have been “much more” than a “supplement” to the obscenity exception of free speech protection for sexual expression. Rather, in the Court’s assessment, the ban of virtual child pornography was an example of the kind of thematic obscenity that the Court protected almost fifty years earlier in Kingsley. The Court in Ashcroft wrote, “The statute proscribes the visual depiction of an idea—that of teenagers engaging in sexual activity—that is a fact of modern society and has been a theme in art and literature throughout the ages.”

In resisting the congressional effort to ban virtual child pornography, the Court in Ashcroft suggested that Ferber might be the only exception to the “general rule” that sexual expression “can be banned only if obscene” within the meaning of Miller. This suggestion helps to explain the judicial rejection of anti-pornography laws put forward by some (but by no means all) feminists as an alternative to Miller obscenity. In this alternative, pornography is considered harmful not because it appeals to the prurient interest of its audience, but because it discriminates against women. This is so, supporters of such laws claim, because pornography engenders in men a distorted view of women as subordinate sexual objects rather than as equal and fully human, a view that encourages discrimination and violence toward women. As described by Andrea Dworkin, one of the pioneers of this alternative understanding of the harmful potential of sexual expression, “Pornography, unlike obscenity, is a discrete, identifiable system of sexual exploitation that hurts women as a class by creating inequality and abuse.” Following that diagnosis, the essence of unlawful pornography is expression that depicts “the graphic sexually explicit subordination of women.” The social, literary, or artistic value of any depiction that qualifies as pornography, in this approach, is trumped by the harm it caused, and thus, as is probably true of Ferber child pornography, any such value cannot save the depiction from governmental prohibition.

This alternative approach to Miller obscenity was halted in its tracks when the Supreme Court summarily affirmed (that is, upheld without opinion) a decision by a federal court of appeals which invalided a city ordinance that embodied the anti-pornography approach to sexual expression (American Booksellers Association v. Hudnut, 1985). The court of appeals in Booksellers found the definition of pornography in this approach to be fatally infected by viewpoint discrimination. As the court explained, “Speech treating women in the approved way—in sexual encounters ‘premised on equality’ …—is lawful no matter how sexually explicit. Speech treating women in the disapproved way—as submissive in matters sexual or as enjoying humiliation—is unlawful no matter how significant the literary, artistic, or political qualities of the work taken as a whole. The state may not ordain preferred viewpoints in this way.”

The court of appeals’s rejection of anti-pornography laws in Booksellers anticipated the Ashcroft rationale for invalidating virtual child pornography. In both instances, legislatures had stepped beyond the boundaries marked by Miller and Ferber for the prohibition of sexual expression. While Miller had allowed communities to enforce their standards of moral decency concerning the explicitness of sexual expression and Ferber had allowed communities to protect children from the abuse caused by their use as subjects in the portrayal of sexual activity, upholding the anti-pornography and virtual child pornography laws would have given legislators a free hand to outlaw ideas and themes of sexuality with which they disapproved. This amounted to censorship in its purest form. From the perspective of what might have been, obscenity, as supplemented by child pornography, represent limited exceptions to the general right under the First Amendment to engage in sexual expression.

With that said, any careful reading of the court of appeals’s dismissal of anti-pornography laws in Booksellers (as well as Ashcroft‘s protection of virtual child pornography) calls into question the constitutional case for permitting anti-obscenity laws as well. Just as the two kinds of anti-pornography laws are constitutionally suspect because they seek to enforce a government-preferred view of sexual ideas and themes, anti-obscenity laws should be equally suspect for their effort to enforce a majoritarian view of sexual morality. Each legislative effort is paternalistic, and in the stark expression of the court of appeals in Booksellers, each on some level constitutes “thought control.” Each law is grounded on the legislative worry that sexual expression of a particular kind will generate a mental predisposition in the audience to act in an antisocial manner. Indeed, if anything, the basis for legislative concern with respect to antisocial conduct is probably more supportable with respect to virtual child pornography and pornography than it is with respect to obscenity. The Court flashed recognition of this fundamental conflict between anti-obscenity laws and free speech principles in Stanley, but as we have seen, the justices almost immediately backed away from the implications of that insight. It is only the prohibition of Ferber child pornography that “does not rely on a paternalistic interest in regulating [the recipient’s] mind” (Osborne v. Ohio, 1990), and thus it is only the exclusion of that category from protected sexual expression that squares with the First Amendment.

Ultimately, the First Amendment exception for obscenity might best be explained not as a reasoned elaboration of free speech principles, but rather as a strategic judgment by the Court reflecting the difficulty of challenging anti-obscenity laws that many local communities believe are necessary to preserve their moral integrity. Perhaps that is the meaning of the Roth Court’s otherwise tautological emphasis of the “universal” outlawing of obscenity by the states as a justification for denying such expression First Amendment protection.

Channeling Non-obscene Sexual Expression

After Ashcroft, legislatures will be hard-pressed to justify banning sexual expression that does not fall within the Miller definition of obscenity or the Ferber definition of child pornography. But the Court has allowed government officials considerably more leeway in restricting sexual expression, short of outright prohibition, than it has with respect to other forms of protected speech. The justices have not offered a satisfactory account of why this is so. The constitutional clearance for greater regulatory control over sexual expression might be explained by a judicial reluctance to accord such expression the same degree of First Amendment protection provided other types of speech. Or perhaps the justices believe that governments have especially powerful regulatory justifications for restricting the times and places of sexual expression. Most likely, the answer lies in some combination of those two explanations.

The initial position of the Court following Paris/Miller suggested that the First Amendment prevented government officials from targeting non-obscene sexual expression for special restrictions. Just a couple of years after upholding anti-obscenity laws in Paris and in Miller, the Court invalidated a local ordinance that prohibited drive-in theaters from showing films that included nudity on screens that were visible to the public outside the theater (Erznoznik v. City of Jacksonville, 1975). The ordinance reflected a legislative judgment that nude scenes would be “especially offensive to passersby.” But nudity alone, the Court had held just the prior year, does not render a film obscene (Jenkins v. Georgia, 1974). The Court in Erznoznik thus put the Jacksonville ordinance through the constitutional wringer of strict scrutiny reserved for “censorship of otherwise protected speech on the basis of its content.” The ordinance was doomed. As the Court explained, “when the government, acting as censor, undertakes selectively to shield the public from some kinds of speech on the ground that they are more offensive than others, the First Amendment strictly limits its power.” The Court in Erznoznik thus reinforced the understanding, traceable to Roth and consistent with Ferber, that sexual expression, which was neither obscene nor otherwise proscribable remained fully protected. The First Amendment denied government officials carte blanche to inhibit sexual expression because they or some community members found it offensive.

Chief Justice Warren E. Burger, the author of the majority opinions in Miller and Paris, dissented in Erznoznik. He was joined by the future Chief Justice, William H. Rehnquist. Burger accused the majority of engaging in a “rigidly simplistic” application of the categorization and content distinction principles. He discounted the free speech interests affected by the Jacksonville ordinance because, as he put it, the nude scenes possessed only a “slight to nonexistent” communicative value to passersby. Burger ignored, however, the expressive value of such scenes for those who viewed the film in the theater. Burger, on the other hand, assessed the community’s regulatory interest as quite high, owing to the uniquely “eye-catching” nature of drive-in movie screens. Because his cost/benefit balance decisively favored the nudity restriction, the Chief Justice criticized the majority for slavishly following conventional free speech doctrine to invalidate the Jacksonville ordinance.

The very next year, the justices began to experiment along the lines advocated by Chief Justice Burger in his Erznoznik dissent (Young v. American Mini Theatres, Inc., 1976). Young involved a city zoning ordinance that placed special restrictions on the location of movie theaters that specialized in sexually explicit, but non-obscene films, restrictions that did not apply to theaters that limited themselves to more mainstream fare. By targeting so-called “adult” movie theaters, the Young ordinance would seem to fit the Erznoznik description of “censorship of otherwise protected speech on the basis of its content,” and thus, to face the same fate of the Jacksonville ordinance. But a bare majority of the justices voted to uphold the Young zoning restriction, although they were unable to agree on a justification for doing so. All five justices who endorsed the constitutional legitimacy of the Young ordinance agreed on one proposition, however, that “there is surely a less vital interest in the uninhibited exhibition of material that is on the borderline between pornography and artistic expression than in the free dissemination of ideas of social and political significance.”

In an opinion for the Court that in the critical section only gathered a four-justice plurality, Justice John Paul Stevens built on that premise to suggest a modification of the categorization principle to accommodate government regulation of sexual expression on the “borderline” of First Amendment protection. Following the lead of Chief Justice Burger’s dissenting opinion in Erznoznik, Justice Stevens claimed that the erotic films targeted by the zoning law in Young, while not obscene, nevertheless did not merit full First Amendment protection. Rejecting the all-or-nothing, protected-or-unprotected dichotomy of categorization analysis, Stevens envisioned as an alternative, a sliding scale allowing the justices to calibrate a distinct level of First Amendment protection for particular categories of protected speech, with each adjusted according to its level of service to free speech values. Stevens, writing for the Young plurality, argued for a low level of constitutional protection for “erotic materials” because their First Amendment value was “of a wholly different, and lesser magnitude than the interest in untrammeled political debate that inspired Voltaire’s immortal comment.” The Voltaire quotation to which Justice Stevens referred was, “I disapprove of what you say, but I will defend to the death your right to say it.” In Young, Stevens famously added, “few of us would march our sons and daughters off to war to preserve the citizen’s right to see [sexually explicit films] exhibited in the theaters of our choice.” Having reduced the level of protection for erotic materials, and perhaps for sexual expression generally, Stevens was careful not to eliminate First Amendment protection entirely. The Young plurality made clear that it would not condone “the total suppression” of such fare. But the plurality justices were willing to assess regulations of erotic materials short of a total ban permissively, against the background understanding that the speech at stake resided at the periphery, rather than at the center, of constitutional concern.

Justice Stevens’s devaluing of sexually explicit films marked a departure from Erznoznik, which had treated films containing nudity no differently than other types of protected speech. Indeed, that was precisely Chief Justice Burger’s objection to the Erznoznik decision. There was another departure from Erznoznik in the Young plurality opinion. Whereas the Erznoznik Court had labeled the targeting of a disfavored subject matter (nudity) as censorship, the Young plurality was unfazed by the government’s use of the sexual content of a film as the basis for special restrictions. The Stevens plurality limited its censorship concern to laws directed at the “social, political, or philosophical message” communicated by a film.

Young‘s rejection of strict scrutiny for laws imposing special restrictions on protected expression based on its sexual content hardly went unchallenged. Justice Potter Stewart, writing for the four dissenting justices in Young, criticized the plurality’s revisionism in strong terms, charging the plurality with “rid[ing] roughshod over cardinal principles of First Amendment law.” To the dissenters, Stevens had “[stood] ‘Voltaire’s immortal comment’ on its head.” Stewart explained, “For if the guarantees of the First Amendment were reserved for expression that more than a ‘few of us’ would take up arms to defend, then the right of free expression would be defined and circumscribed by current popular opinion. The guarantees of the Bill of Rights were designed to protect against precisely such majoritarian limitations upon individual liberty.”

There is considerable power in Justice Stewart’s objection to Stevens’ challenge to conventional free speech doctrine. But there is an undeniable commonsense appeal to Justice Stevens’s sliding scale of First Amendment protection as well. It seems obvious that the myriad varieties of protected speech do not equally serve free speech values. Yet categorization divides all expression into two basic groups, protected and unprotected. Such an approach provides a blunt instrument indeed, at least for some kinds of speech. A sliding scale of First Amendment protection, as opposed to the bipolar model of protected-or-unprotected speech, is alluring because it would allow courts to calibrate more carefully the nature and level of protection appropriate for the various types of expression that merit at least some First Amendment protection.

There are nevertheless hazards associated with Stevens’s proposed undertaking. Especially for those who are attracted by negative free speech theory—that is, those who are skeptical, even cynical, concerning the competence and integrity of government officials to regulate expression—the prospect of fine-tuning free speech doctrine to provide for less-than-fully-protected speech would be an ominous development. As the experience with obscenity and child pornography shows, it can be perilously difficult for judges to define a satisfactory boundary between the broad categories of protected and unprotected speech. The difficulty is compounded by proliferating subcategories of lesser-protected speech. Along with the increased difficulty, there would be greater danger of official censorship was well. The sliding scale approach provides government officials, including judges, broader discretion to register both their own and their sense of the community’s approval or disapproval of particular kinds of expression. It is easier, and for that reason tempting, to reduce First Amendment protection for disfavored forms of speech than it is to justify denying such protection altogether. The Stevens plurality in Young, as the Stewart-led dissenters charged, may have been guilty of this very failing when they removed erotic materials to the borderline of First Amendment protection.

The hazards associated with the kind of wholesale sliding scale approach that Justice Stevens advocated in Young were exacerbated by his relaxation of the content distinction principle. Although the Young plurality opinion encouraged judges to vigorously review restrictions placed on sexual expression because of any particular message it conveys, it permitted them to lower their guard when evaluating restrictions on the basis of the sexual content of the expression. While viewpoint discrimination is the paradigmatic instance of censorship, and message-based restrictions are therefore highly suspect, it hardly follows that subject-matter discrimination is constitutionally trouble-free. Although a governmental focus on subject matter rather than on viewpoint makes it plausible that a restriction may have been grounded on legitimate regulatory concerns and not simply on official or majoritarian disapproval of the restricted speech, any such assumption is treacherous when the subject matter is sexual expression. The Court ordinarily invokes strict scrutiny to evaluate subject-matter distinctions because it is prudent to presume the worst and to obligate judges to review regulatory justifications closely and skeptically to ensure that they are not pretexts for censorship. The lesser scrutiny of Young, by contrast, lulled the Court into accepting at face value the city’s proffer of conventional zoning justifications for its targeting of adult theaters for special restrictions, while a closer, more skeptical look may well have uncovered a censorial motive.

Justice Stevens was never quite able to garner majority support for his revised approach to sexual expression. While the Court has retained its permissiveness toward zoning laws that impose special restrictions on adult entertainment venues, the justices have subtly shifted their rationale for doing so. In an important later decision that, like Young, upheld a city zoning restriction that targeted adult movie theaters, the Court virtually ignored the centerpieces of the Young plurality’s analysis—the supposed lesser-protected status of erotic materials and the assumed benign quality of subject-matter restrictions—and relied instead on the content distinction principle (City of Renton v. Playtime Theatres, Inc., 1986). But just as the Young plurality found it necessary to adjust conventional free speech doctrine in order to accommodate such a zoning restriction, the Court in Renton had to tweak the content distinction principle to achieve that result. The justices evaded strict scrutiny in Renton by finding the zoning law to be content neutral, even though it, like the law in Young, had explicitly imposed special zoning restrictions on theaters that specialized in erotic films. The Court reached this conclusion in the spirit of Young, by looking beyond the content-specific text of the law to the legislative purpose behind limiting the location of adult movie theaters. According to the Court, strict scrutiny of the restriction would have been appropriate had the “predominate concerns” of the legislature concerned the “primary effect” of the adult films, that is, the effect of the films on its viewers. The Court found, however, that the legislators had focused on the “secondary effects” associated with adult movie theaters, such as increases in crime and decreases in property values in neighboring areas. Because such secondary effects were not caused by the audience response to the content of the films those theaters exhibited, the Court considered them to be content-neutral justifications for the zoning restriction. In the Renton revision of the content distinction principle, then, the government’s use of speech content as a basis for restriction is not itself a sufficient warning of official censorship to warrant strict judicial scrutiny. Unless it also can be proven that the law was “enacted for the purpose of restraining speech on the basis of its content,” the more forgiving, but still heightened, level of intermediate scrutiny suffices.

This revised understanding of the content distinction principle allowed the Court to distinguish the zoning laws that it upheld in Young and Renton from the regulation of drive-in theaters that it invalidated in Erznoznik, even though in each case the community had restricted the exhibition of certain films explicitly on the basis of the content of those films. The key difference was that the zoning laws involved in Young and Renton, unlike the law in Erznoznik, were not justified by reference to the primary effect of the films on those who viewed them. The restriction invalidated in Erznoznik reflected a legislative desire to protect passersby from the offense they might experience upon viewing nudity on a drive-in movie screen. But the restrictions in Young and Renton, the Court found, were predominately motivated not by a legislative design to suppress the erotic content of adult films, but by the more benign agenda of controlling the secondary effects of adult movie theaters on their neighborhoods. In the Renton framework, while a community cannot ban completely sexually explicit, non-obscene expression, it has considerable leeway in channeling such expression to times and places that minimize their secondary effects (City of Los Angeles v. Alameda Books, Inc., 2002).

The effort in Renton to settle the controversy created by the doctrinal innovations of the Young plurality was itself controversial. The Court’s acceptance of a content-neutral justification to lessen the scrutiny of a content-based law created a concern that lawmakers would seize on such justifications as a pretext to cover up censorship. This concern was exacerbated by the Renton Court’s disinclination to require that legislators be motivated exclusively by concerns over the secondary effects of regulated speech before considering a content-based law to be neutral. It is highly questionable whether a judicial finding that legitimate regulatory concerns predominated over an illegitimate censorial agenda in the legislature should justify lessening the usual strict scrutiny that attends content-based speech restrictions. The legislature’s use of a content distinction principle in a speech restriction itself signals a strong possibility of a censorial agenda and should not be set aside lightly. It is often difficult, to say the least, for courts to determine with any sense of confidence whether one legislative intent predominated over another in the enactment of a statute. Moreover, a statute that regulates the right of expression on the basis of speech content has the same censorial effect regardless of whether a censorial or a noncensorial intent predominated in the legislature. The effect of the restrictions in Young and Renton was that theaters that exhibited erotic films, and the audiences for those films, were denied venues that were open to theaters that exhibited films that were not sexually explicit.

Nevertheless, the Renton focus on governmental purpose as keying the content distinction fit conventional free speech jurisprudence better than had the innovations Justice Stevens had suggested in Young. And perhaps for that reason, the effect of the Renton revision of the content distinction principle has been far more limited than that which would have followed the Court’s adoption of the Young plurality position. The cases involving the selective zoning of adult businesses remain the only instances in which the justices have allowed textually content-based statutes to escape strict scrutiny through a governmental showing of a secondary effects motivation. Indeed, Justice Anthony M. Kennedy, while endorsing the outcomes in Young and Renton, recently labeled the Renton rationale a “fiction” (see City of Los Angeles v. Alameda Books, Inc., 2002, concurring in the judgment).

Moreover, the Court’s refusal in Renton to endorse the Young plurality’s finding of a lesser-protected position for sexual expression importantly suggested that the justices were prepared to presume the unconstitutionality of governmental restrictions enacted for the purpose of addressing the “primary effects” of such expression, most commonly, the offense experienced by some unintended audience members. And indeed, the Court has since made clear that Young and Renton have “no application to content-based regulations targeting the primary effects of protected speech” (see United States v. Playboy Entertainment Group, Inc., 2000). Such laws, like the law invalidated in Erznoznik, receive strict scrutiny if they target protected, sexually explicit expression because it offends “the sensibilities of listeners [or viewers].”

But to hold such laws to strict scrutiny does not mean that they are necessarily invalid. The Court has continued to enforce the line of variable obscenity precedent allowing prohibitions on the distribution of non-obscene sexual expression to minors. The justices in Playboy, for example, recognized the authority of the government to restrict the free flow of sexual programming, via broadcast media, cable television, or the Internet, “unwanted into homes where children might see or hear it against parental wishes or consent.” Because of strict scrutiny, however, the Court accepts this justification only if the regulation is the least restrictive option for prohibiting access by children. In the years since Renton, the justices have tended to inquire whether available technology allows parents to block the access of their children to the sexual expression from which they wish to shield them. If so, the Court has not allowed governmental restrictions (see, e.g., Ashcroft v. American Civil Liberties Union, 2004; Playboy; Sable Communications of California, Inc., v. FCC, 1989). If not, the Court has accepted restrictions of protected sexual expression that are appropriately tailored to the overriding and long-recognized societal interest in preventing the harmful effects of exposing children to erotic materials (see Ashcroft v. American Civil Liberties Union, 2002).

Commercial Speech

In the formative years of free speech jurisprudence, the Supreme Court attempted the same kind of categorical exclusion of commercial speech that it later would try with obscenity. The justices believed that government restrictions on commercial speech, which they defined as “purely commercial advertising,” did not implicate the First Amendment because such restrictions did not threaten “the freedom of communicating information and disseminating opinion” (Valentine v. Chrestensen, 1942). Valentine provides an amusing illustration of this early treatment. Chrestensen had attempted to distribute leaflets advertising the availability of submarine tours for a stated fee. Local law prohibited the distribution of commercial leaflets, and the authorities advised Chrestensen to desist. Chrestensen then revised his leaflet, adding a second side that protested the locality’s refusal to provide certain services for his submarine. The Court held that the newly added protest did not transform Chrestensen’s commercial advertising into protected advocacy, for “[i]f that evasion were successful, every merchant who desires to broadcast advertising leaflets in the streets need only append a civic appeal, or a moral platitude, to achieve immunity from the law’s command.”

There is commonsense appeal to the Court’s refusal in Valentine to protect commercial speech. At first it seems readily apparent that “purely commercial advertising” does not merit inclusion in the freedom of speech. To paraphrase Justice Stevens, few of us would march our children off to war to defend the right of business firms to hawk their products. But Justice Stevens’s rhetorical flourish provides a self-defeating approach to defining First Amendment boundaries because it subordinates free speech to majoritarian tastes. On inspection, what is it about commercial speech that justifies its exclusion from First Amendment protection? Does not commercial advertising typically involve the communication of “information” and “opinion,” which the Valentine Court described as defining characteristics of protected speech? The traditional answer conceded the point, but distinguished the business nature of the information and opinion communicated by commercial advertising from the social and political commentary that lay at the heart of the First Amendment. Commercial speech was conceptualized as a business practice rather than as individual expression, and government restrictions on advertising were regarded as unproblematic instances of economic and business regulation, not censorship.

The categorical exclusion of commercial speech, like the similar exclusion of obscenity, relied heavily on the ability of judges to distinguish commercial speech from noncommercial speech. And as with obscenity, this enterprise proved to be exceedingly difficult. What was it precisely that marked an advertisement as “purely commercial,” and thus outside the freedom of speech? It was not simply that the speech took the form of an advertisement. In New York Times Co. v. Sullivan (1964), the Supreme Court strongly protected the newspaper against a defamation claim for running an advertisement soliciting contributions for the civil rights movement in the South. Even though the Times is a business and was compensated for running the ad, the Court rejected the argument that the advertisement was unprotected commercial speech within the meaning of Valentine. The Court distinguished Times from Valentine based on the content of the advertisements in the two cases. While Chrestensen’s ad promoted tours on his submarine, the ad in the Times “communicated information, expressed opinion, recited grievances, protested claimed abuses, and sought financial support on behalf of a movement whose existence and objectives are matters of the highest public interest and concern.” The Court described the Times ad as an “editorial advertisement,” which was protected under the First Amendment because it provided “an important outlet for the promulgation of information and ideas by persons who do not themselves have access to publishing facilities.” Simply put, the Times ad promoted a cause, not a product.

The Times distinction between commercial and editorial advertising followed Valentine‘s separation of commercial speech from protected speech. The Court’s refusal to allow a commercial motive (that is, the desire to make money) on the part of the speaker to determine its categorization of the nature of the speech was shrewd. Had the justices not so held, they would have pulled the constitutional rug out from under professional authors and their publishers. But the Times focus on advertising content was hardly foolproof. The Court’s decision in Bigelow v. Virginia (1975) is a striking illustration of the difficulty inherent to the commercial/noncommercial distinction. Bigelow had been convicted of violating a Virginia statute that made it unlawful to distribute a publication that encouraged or facilitated abortions. His newspaper, which was distributed in Virginia at a time when abortions were illegal there, carried an advertisement offering counseling and placement services for lawful abortions in New York. (The ad ran before the Court had decided Roe v. Wade (1973), which recognized a constitutional right for women to have an abortion.) Was Bigelow’s advertisement commercial speech? Like the ad in Valentine, his ad promoted a service-for-fee arrangement, but like the ad in Times, it also promoted a cause, abortion access. Bigelow’s advertisement, in other words, shared the qualities of both commercial and editorial advertising, without fully fitting in either category. Yet, however one described Bigelow’s ad, one thing seemed clear: it was not “purely commercial advertising,” which was how the Court in Valentine had defined the unprotected category of commercial speech. Bigelow’s advertisement included, the Court found, “factual material of clear ‘public interest.'” Moreover, the Virginia law banning publications that encouraged or facilitated abortions followed the blueprint for official censorship rather than economic regulation. It reflected a transparent legislative effort to stifle advocacy of, and to deny information concerning, abortion access, which, then as now, was a highly charged political and social issue.

The Court in Bigelow could have used the free-speech soundings of the abortion-services advertisement to conclude, as it had in Times, that the ad was not “purely commercial advertising” and thus was not excludable from the First Amendment under Valentine. The justices instead used Bigelow far more provocatively to challenge Valentine head on, declaring for the first time, “commercial advertising enjoys a degree of First Amendment protection.” The Court declared, “The relationship of speech to the marketplace of products or of services does not make it valueless in the marketplace of ideas.” The suggestion in Bigelow that commercial speech would be afforded “a degree of First Amendment protection,” however, suggested less than a full-fledged, judicial embrace of such speech within the freedom of speech. And while the Court was prepared to provide some constitutional protection, it was not ready to define the nature and extent of that protection. The justices settled for the observations that the abortion-services advertisement was neither “deceptive nor fraudulent,” that the advertised activity (abortion) was lawful in the state in which it would take place (New York), and that Virginia lacked any substantial interest in shielding its residents from information concerning the availability of lawful abortion services. The Court overturned Bigelow’s conviction as a violation of the First Amendment.

The Court expanded the Bigelow opening for the constitutional protection of commercial speech the following year in the breakthrough decision, Virginia State Board of Pharmacy v. Virginia Citizens Consumer Council (1976). At issue was another Virginia law. This one prohibited licensed pharmacists from advertising the prices of prescription drugs. Because Virginia permitted only licensed pharmacists to dispense such medications, the effect of the law was to impose a complete ban on price advertising of prescription drugs in the state. Unlike the ads of Times and Bigelow, the advertising Virginia targeted in Virginia Pharmacy was purely, indeed paradigmatically, commercial in nature. The Court observed, “Our pharmacist does not wish to editorialize on any subject, cultural, philosophical, or political. He does not wish to report any particularly newsworthy fact, or to make generalized observations even about commercial matters. The ‘idea’ he wishes to communicate is simply this: ‘I will sell you the X prescription drug at Y price.'” The justices seemed to relish the opportunity to return to Valentine.

If Times and Bigelow had taught how elusive the distinction between commercial and traditionally protected speech can be, Virginia Pharmacy revealed to the justices a public interest cognizable by the First Amendment even in what the Valentine Court had written off as “purely commercial advertising.” Notably, the First Amendment challenge to the Virginia statute had been brought not by pharmacists who wished to promote their products, but by a group that represented consumers who required price and product information in order to purchase necessary medications at affordable prices. The trial record in Virginia Pharmacy demonstrated that the prices charged by pharmacies for prescription drugs differed markedly within the same locality, but that the Virginia ban on price advertising kept consumers in the dark about those price differences. Reflecting on the effect of Virginia’s ban, the justices became convinced that an individual’s interest in “the free flow of commercial information … may be as keen, if not keener by far, than his interest in the day’s most urgent political debate.” They also recognized that commercial advertising served a crucial societal function as well. The Court explained:

Advertising, however tasteless and excessive it sometimes may seem, is nonetheless dissemination of information as to who is producing and selling what product, for what reason, and at what price. So long as we preserve a predominantly free enterprise economy, the allocation of our resources in large measure will be made through numerous private economic decisions. It is a matter of public interest that those decisions, in the aggregate, be intelligent and well informed. To this end, the free flow of commercial information is indispensable.

The indispensable role of commercial advertising in pumping price and product information through the market system, the Court decided in Virginia Pharmacy, justified constitutional protection. In addition, the justices expanded Bigelow‘s anti-paternalism theme, suggesting that the First Amendment was offended whenever states manipulate consumer choice by denying the information consumers need to make their own purchasing decisions.

Notwithstanding the boldness of its rhetoric, the Court in Virginia Pharmacy retained the purposeful ambiguity of Bigelow on the place of commercial speech within the freedom of speech. As in Bigelow, the Court carefully circumscribed the holding in Virginia Pharmacy, asserting just that commercial speech would not be denied all First Amendment protection. Because of what the justices regarded as “commonsense differences” between commercial speech and other types of protected speech, they explicitly stopped short of providing commercial advertising the same protection they afforded to speech on political or social issues. The Court made clear the limits of Virginia Pharmacy at the conclusion of its opinion, when the justices cleared government officials to continue restricting commercial advertising that is “false or misleading.” Outside the commercial speech area, the Court steadfastly has refused to make truth a prerequisite for First Amendment protection (see New York Times Co. v. Sullivan, 1964). The Court in Virginia Pharmacy also stipulated that advertising which promoted unlawful activity would be denied protection as well, even though as we have seen, Brandenberg generally protects advocacy of law violation except in cases of incitement. In the end, the Court in Virginia Pharmacy, as it had in Bigelow, left commercial speech in the ambiguous middle position of being neither fully protected nor completely neglected by the First Amendment. Virginia Pharmacy produced the muddled effect of freeing commercial advertising from the otherwise pervasive power of governments to regulate economic activity, while subjecting such advertising to forms of government control that otherwise were anathema in conventional free speech jurisprudence.

Within a couple of years the Court made explicit what Bigelow and Virginia Pharmacy had implied: commercial speech does not warrant full constitutional protection (Ohralik v. Ohio State Bar Ass’n, 1978). This consignment of commercial speech to a lesser-protected status was roughly contemporaneous with Justice Stevens’s effort to reduce the protection of erotic but non-obscene films to a similar middling position (Young v. American Mini Theatres, Inc., 1976; see also FCC v. Pacifica Foundation, 1978). The rationale for less-than-full First Amendment protection was similar in both instances. Listening to the Court in Ohralik one hears Stevens in Young: “[W]e … have afforded commercial speech a limited measure of protection, commensurate with its subordinate position in the scale of First Amendment values, while allowing modes of regulation that might be impermissible in the realm of noncommercial expression.”

Not long after the Ohralik pronouncement, the justices institutionalized the “subordinate position” of commercial speech by adopting a multi-factored test for evaluating government restrictions on commercial advertising (Central Hudson Gas & Electric Corp. v. Public Service Comm’n of New York, 1980). At the outset, the so-called Central Hudson test, following Bigelow and Virginia Pharmacy, completely denied constitutional protection to commercial advertising that promotes an unlawful activity or that is false or misleading. For advertising that crosses that threshold, the Central Hudson test is essentially an adaptation of the intermediate level of judicial scrutiny that courts use when reviewing content-neutral restrictions of speech: The government must justify a restriction of protected commercial speech as being narrowly tailored to serve a substantial public interest. Notwithstanding the relaxation of constitutional protection for commercial speech, the Court served warning in Central Hudson that, as in Bigelow and Virginia Pharmacy, governments would be hard-pressed to justify a total ban of protected advertising.

The intermediate scrutiny prescribed by Central Hudson no doubt appealed to the Court because it finessed the hard choice between strict scrutiny and no scrutiny of content-based commercial speech restrictions. Those all-or-nothing alternatives apparently struck the justices as inappropriate. Moreover, intermediate scrutiny enabled the Court to evaluate commercial speech restrictions on an ad hoc basis and within a flexible, moderate framework. But that very flexibility produced an indeterminacy that has plagued commercial speech jurisprudence. The Central Hudson test, for example, provides no guidance for determining how “substantial” a government interest must be in order to justify an advertising restriction. Nor does it specify how neatly the restriction must fit that interest in order to be “narrowly tailored.” To make matters worse, the Court’s application of the Central Hudson test has exacerbated rather than ameliorated its inherent indeterminacy. In some cases, the Court has applied the test permissively to allow substantial restrictions on advertising content (e.g., Posadas de P.R. Assocs. v. Tourism Co., 1986). In others, the Court has strictly applied the same criteria to invalidate similar restrictions (e.g., Rubin v. Coors Brewing Co., 1995). Having compromised the decision whether or not to strongly protect commercial speech, the justices were left without a normative anchor to secure commercial speech jurisprudence.

In recent years, however, the Court seems to have addressed the problem of indeterminacy by embracing the norm of anti-paternalism to guide its commercial speech jurisprudence. That norm has encouraged the justices to review closely and skeptically commercial speech restrictions that “seek to keep people in the dark for what the government perceives to be their own good” (44 Liquormart, Inc. v. Rhode Island, 1996, Justice Stevens, prevailing opinion). Such paternalism often drives content-based restrictions on truthful advertising. Indeed, government paternalism accounts as well for many content-based restrictions outside the commercial speech area and in general is a strong indicator of official censorship. The anti-paternalism norm thus has firm grounding in the free speech tradition. It also was a prominent theme both in Bigelow and in Virginia Pharmacy.

The anti-paternalism norm blurs the difference in First Amendment status between fully protected speech and commercial advertising. If the First Amendment generally disfavors speech restrictions that are infected by government paternalism, the nature of the speech restricted in a particular instance seems less important. It is the nature of the speech restriction, rather than the nature of the speech restricted, that is of paramount constitutional concern. It is not surprising, then, that the Court in recent years has applied the Central Hudson test with considerable vigor and has consistently invalidated paternalistic restrictions on commercial advertising (e.g., 44 Liquormart; Greater New Orleans Broadcasting Ass’n v. United States, 1999). Indeed, the Court’s application of the Central Hudson test has been so vigorous that its commercial speech decisions of late are more in tune with the strict scrutiny visited upon content-based restrictions of fully protected speech than with the middling judicial concern normally associated with intermediate scrutiny. While the Court has shown no inclination to open up First Amendment protection for commercial advertising that is false, misleading, or a promotion of illegal activity, the justices have closely reviewed restrictions on the content of commercial speech that do not fall within one of those unprotected categories. Notwithstanding its “subordinate position” in the First Amendment pantheon, commercial speech receives considerable constitutional protection by the contemporary Court.

Symbolic Expression

The final category of speech we shall consider differs from the others. The speech categories we have surveyed thus far are defined by the subject-matter content of the expression. Symbolic expression, by contrast, is marked by its nonverbal mode of expression. The symbolic “speaker” communicates a message by conduct, rather than by the spoken or written word. We saw such speech-by-conduct in the Skokie litigation, where the silent march by American Nazi Party members, clad in uniforms that evoked those of the storm troopers of the Third Reich, would have spoken volumes had it occurred as planned. Symbolic expression also raises a different categorization question than we have seen. In fact, the doctrine of symbolic expression is a mirror image of the categorization principle. Instead of examining whether a particular category of speech forfeits First Amendment protection because its social costs overwhelm its communicative value, symbolic expression cases require judges to determine whether a particular action merits constitutional protection because of its expressive quality. Thus, while the categorization principle excludes some types of literal speech from the freedom of speech, the doctrine of symbolic expression opens the First Amendment to at least some forms of non-literal speech.

Although it might seem linguistically counterintuitive to treat some forms of conduct as speech, one suspects that this reading of the First Amendment would have neither surprised nor disappointed the founding generation. Surely they appreciated that the Boston Tea Party of December 16, 1773 communicated the emerging revolutionary resistance to British rule in America at least as forcefully as the best colonial pamphlets, and it did so more succinctly. We know from our everyday experience that individuals sometimes intend for their actions to communicate a message, as when one nods in agreement to a point made, winks or blows a kiss to a loved one, or waves goodbye to a companion. We also know, as the saying goes, that “actions speak louder than words,” so that on occasion, symbolic conduct may provide a far more powerful medium of communication than oral or written speech. To speak against the Vietnam War was one thing; for a veteran of that war to return his or her medals in protest to the government was quite another. For free speech jurisprudence to deny First Amendment protection outright to such symbolic expression would be to deny the reality and richness of human communication. Yet, to bring all conduct that speaks within the protective umbrella of the First Amendment would be equally unrealistic. What of terrorists who destroy lives and property in order to send a message of intimidation? Or the assassin who advocates regime change by killing a political leader? Those actions also speak clearly, but few would protect them as free speech.

Symbolic expression creates a constitutional dilemma. It intertwines an element of speech, which as a default matter is protected by the First Amendment, with an element of conduct, which typically is subject to government control without First Amendment constraints. Symbolic expression thus collapses the traditional distinction between speech and conduct, which has had a formative influence on free speech jurisprudence. The interplay between speech and conduct, with the resultant cross-pulls toward protection and non-protection, were evident in United States v. O’Brien (1968), the Court’s initial effort to solve the symbolic expression puzzle. O’Brien had been convicted of burning his draft card on the steps of the South Boston Courthouse. He testified that his purpose was to persuade those who witnessed the event to reevaluate their position on the Vietnam War. The justices had no difficulty recognizing the expressive quality of O’Brien’s act of protest, and they assumed, without discussion, “the alleged communicative element in O’Brien’s conduct [was] sufficient to bring into play the First Amendment.” But they coupled that assumption with the advisory, “We cannot accept the view that an apparently limitless variety of conduct can be labeled ‘speech’ whenever the person engaging in the conduct intends thereby to express an idea.” The Court’s advisory became a warning when the justices ultimately denied O’Brien’s free speech claim because of what they regarded as an overriding governmental interest “in assuring the continued availability of issued Selective Service certificates.”

The Court’s initial ambivalence over protecting symbolic expression gradually evolved into a two-part analysis that attempts to steer O’Brien‘s middle course, protecting some, but not all, conduct with an “alleged communicative element.” In this approach, courts first decide whether the conduct at issue qualifies as “speech.” To so qualify, the conduct, in context, must be sufficiently communicative to warrant First Amendment protection. More specifically, courts decide whether the individual engaged in the conduct at issue with the “intent to convey a particularized message,” which “in the surrounding circumstances,” was likely to be understood by those who saw it (Spence v. Washington, 1974). If so, the individual’s conduct communicated a message to an audience, and therefore, it was the functional equivalent of speech.

The Court has not been particularly demanding when enforcing the requirement that a symbolic message be “particularized” in order to qualify for First Amendment protection. O’Brien’s burning of his draft card, delivered during the heat of controversy over Vietnam, unmistakingly conveyed a message of opposition to the draft, as well as to the war, that would have satisfied any definition of “particularized.” But the justices have not pressed this requirement when assessing conduct that spoke with less particularity. The Court has protected flag-burning, for example, even though the act often expresses only the most general message of anti-Americanism (Texas v. Johnson, 1989; United States v. Eichman, 1990). In a further stretch, the justices also assumed that demonstrators, by sleeping at a protest site in a downtown park, symbolically expressed the plight of the homeless (Clark v. Community for Creative Non-Violence, 1984). The requirement that an actor’s conduct convey a “particularized” message has not imposed much of a barrier to the First Amendment’s protection of symbolic expression.

The important limitation has arisen from the second part of the Court’s analytic framework. When courts find that an individual’s conduct was sufficiently communicative to qualify as speech, they then apply the content distinction principle to the governmental restriction at issue. This move from the nature of the individual’s expressive act to the nature of the government’s restriction finds its genesis in the O’Brien decision itself. The Court signaled in O’Brien that the symbolic speaker’s First Amendment claim would be far stronger in cases “where the alleged government interest in regulating conduct arises in some measure because the communication allegedly integral to the conduct is itself thought to be harmful.” But when the government’s regulatory interest is “limited to the noncommunicative aspect” of a symbolic speaker’s conduct, as the Court held was the case in O’Brien, the justices were prepared to accept more readily restrictions on the individual’s conduct. The content distinction principle provides the contemporary method of enforcing the O’Brien distinction between the restrictions of symbolic expression that are constitutionally suspect and those that are not.

The content distinction principle addresses the symbolic expression dilemma posed by the fusion of speech and conduct by disentangling those elements. If an act of symbolic expression violates a law that targeted the “expressive element” of a symbolic speaker’s conduct, courts treat the law as a content-based restriction of speech (Texas v. Johnson, 1989). Because such a law is activated by what the symbolic conduct “said,” it is no different from any other attempt by the government to silence a speaker because officials believe that the content of the speech is itself harmful to the public interest. So long as the speech content does not fall within a category of unprotected speech, such a law is presumptively censorial and will survive only if it withstands strict judicial scrutiny, which is a rare event indeed. If on the other hand, the law at issue targets the conduct element of symbolic expression, courts reduce their scrutiny to the intermediate level appropriate for content-neutral speech restrictions (see Clark v. Community for Creative Non-Violence, 1984). This relaxation of the standard of review is justified because the conduct itself is harmful, regardless whether it happens to be expressive, and the First Amendment typically does not limit the state in prohibiting harmful conduct. Perhaps for this reason, the Supreme Court has suggested that governments have more leeway in addressing the “noncommunicative impact” (O’Brien) of expressive conduct than they possess when imposing content-neutral restrictions on pure speech (Texas v. Johnson, 1989).

The content distinction principle thus provides that the intertwining of speech and conduct in symbolic expression neither authorizes government censorship of valuable speech nor immunizes harmful conduct from governmental control. While the justices have never upheld a restriction on symbolic expression that they have understood to be a regulation of speech content, they have accepted almost uncritically every restriction that they have understood to be concerned solely with harmful conduct by a symbolic speaker.