In my own newly famous city of Charlottesville, Virginia, as well as in Berkeley, Boston, Middlebury, and an increasing number of other locations, individuals and groups engaging in constitutionally protected acts of speaking, parading, protesting, rallying, and demonstrating have become targets for often-large groups of often-disruptive counter-protesters. And although most of the contemporary events have involved neo-Nazi, Ku Klux Klan, and other white supremacist speakers who are met with opposition from audiences on the political left, it has not always been so. Indeed, what we now identify as the problem of the hostile audience has often involved more sympathetic speakers confronted by less sympathetic crowds. Yet though the issue is hardly of recent vintage, contemporary events have highlighted the importance of reviewing the relevant constitutional doctrine and thinking again and anew about how, if at all, government and law should respond to disruptive audiences.

The value of returning to the question of the hostile audience is heightened by the fact that existing legal doctrine on the question is, at best, murky. It is now widely believed that restricting the speaker on account of the actual or predicted hostile and potentially violent reaction of the audience gets our First Amendment priorities backwards. But it is hardly clear that this belief was ever correct, and even if it were, it is even less clear that it is sufficient to deal with the constitutional and policy complexities of many of the contemporary encounters. The point of revisiting the problem of the hostile audience is therefore not so much to urge a change in existing understandings and legal doctrine as it is to emphasize how many open questions still remain, and how current events might bear on possible answers to these questions.

The Forgotten Denouement of a Remembered Event

As has been extensively documented, in 1977 a group called the National Socialist Party of America, self-described as Nazis and led by a man named Frank Collin, proposed to conduct a march in Skokie, Illinois. The neo-Nazi marchers chose Skokie as their venue because of its substantial Jewish population, a population that at the time contained an especially large number of Holocaust survivors. And in order to inflict the maximum amount of mental distress on Skokie’s residents, the marchers were planning to carry Nazi flags, display the Swastika, and wear Nazi uniforms, jackboots and all.

Despite the efforts of the Village of Skokie to prohibit the event, the Nazis’ First Amendment right to hold the march was upheld by both the Supreme Court of Illinois and the United States Court of Appeals for the Seventh Circuit. And although the United States Supreme Court’s refusal to give the case full plenary consideration was technically not a decision on the merits of the controversy, the events were so broadly publicized that the Court’s refusal to hear the case with full briefing and argument was widely understood by the public, and presumably by the members of the Court as well, as firmly announcing that the Nazis’ right to march was by then clearly established in constitutional doctrine.

Although the Nazis prevailed in several courts and thus won their right to march when, where, and how they had proposed, what is often forgotten is that the march never happened. The Nazis wrapped themselves in their legal victory but at no time actually exercised the right that they had won in court, at least in Skokie itself. They did hold similar but small events in Chicago on June 24 and July 9, 1978, both without benefit of permit, but the permit they were eventually granted to march in Skokie on June 25 went unused.

We remain unsure why. Perhaps the Nazis disbelieved Skokie’s representation that the town would make “every effort to protect the demonstrators . . . from responsive violence.” Perhaps they viewed their legal victory as more important than the march itself. Or perhaps the publicity surrounding the event made it seem more desirable to the Nazis to march in front of what they hoped would be a larger audience in Chicago. But whatever the reason, the Nazis never marched in Skokie.

Because the Skokie march never happened, we can only speculate about the answers to questions that may have arisen had it taken place. Some of those questions relate to law enforcement. Assuming that there would have been far more objectors than Nazis, and that at least some of those objectors would have been inclined to physically confront or assault the marchers, would the Skokie police department have attempted to prevent the physical encounters, and if so, how, with what size force, and with what degree of aggressiveness? And if Skokie’s police department, even assuming its best efforts, was not up to the task because of the sheer size of the hostile audience, would the Illinois State Police have stepped in? The Cook County Sheriff’s Office? The Illinois National Guard? Would Illinois state authorities have requested federal assistance to protect the exercise of a federal constitutional right? Even absent a request from the state, would federal forces have been deployed to secure compliance with a constitutional right and a judicial order, as when President Dwight Eisenhower sent troops to Little Rock, Arkansas, in 1957? These are factual and historical questions rather than normative or doctrinal ones, but they suggest a number of issues that are indeed normative or doctrinal, or both, about which the current doctrine provides few answers.

The overarching legal question that would have been presented had the Nazis marched in Skokie, and the question presented by so many of the contemporary events, is the question of the hostile audience. Assuming the existence of a group or of individuals otherwise constitutionally entitled to say what they want to say where they want to say it, how does the existence of an actually or potentially hostile audience change, if at all, the nature of the speakers’ rights or change, if at all, the obligation, if any, of official responsibility to protect the speakers and their speech?

The Doctrinal Foundation—Holes and All

One motivation for this paper, even apart from the way in which current events have highlighted the problem of the hostile audience, is that there is less settled law on the questions raised at the close of the previous section than we might have supposed. But what law there is dates to 1940 and the case of Cantwell v. Connecticut. Jesse Cantwell, like so many of the individuals who helped forge the modern First Amendment tradition, was a Jehovah’s Witness. He, along with his brother and father, travelled to a street in New Haven, where he attempted to sell books and pamphlets in a neighborhood that was approximately 90 percent Catholic, in the process playing a phonograph record vehemently attacking the Catholic Church. The content of the record angered several onlookers, some of whom testified that they “felt like hitting” Cantwell if he did not leave immediately. In fact he did leave when confronted, but Cantwell, his brother, and his father were nevertheless charged with “breach of the peace” under Connecticut law. The convictions of the brother and father were reversed by the Supreme Court of Connecticut, but Jesse’s was upheld. On appeal to the U.S. Supreme Court, however, Jesse’s conviction was found to be a violation of the First Amendment, with Justice Owen Roberts writing the unanimous opinion. The Court held out the possibility that a conviction might be sustained were a speaker to use profane words, indecent language, epithets, or personal abuse directed specifically at particular listeners or were a statute to have been aimed precisely at conduct such as Cantwell’s and to have incorporated a legislative finding that such conduct constituted a clear and present danger. But because Cantwell was prosecuted only under the general common law notion of breach of the peace, and because there had been no profane or indecent direct personal abuse, the Court concluded that Cantwell’s conviction could not stand.

Even though the Cantwell Court suggested that the reactions of listeners could justify punishing a speaker under certain circumstances, the opinion still says next to nothing about what a legislature would need to do in order to craft a constitutionally permissible statute. And because the audience in Cantwell consisted of only a few people to whom Cantwell had generally addressed his words and his recording, the decision in its entirety is at best a precursor to the hostile audience problem and appears to be more of an angered listener case than a hostile audience one. The first true hostile audience case was yet to come, and it arrived nine years after Cantwell in Terminiello v. Chicago, a decision worthy of more extended discussion.

Arthur Terminiello, a suspended Catholic priest, was invited to Chicago from his Birmingham, Alabama, base to deliver a well-publicized and virulent anti-Communist and anti-Semitic speech, in which he referred to “atheistic, communistic . . . Zionistic Jews” as “slimy scum,” accused former Secretary of the Treasury Henry Morgenthau of plotting to cut off the limbs of German babies, charged Franklin and Eleanor Roosevelt and the Supreme Court with being part of the same Jewish Communist conspiracy, and announced that the New Deal itself was among the prominent manifestations of this conspiracy. Terminiello’s audience in a rented hall consisted of somewhere between 800 and 1,000 people, most of them sympathetic but some hostile. Outside there were gathered another 200 to 300 more people who had been turned away for reasons of space, an additional several hundred members of a picket line, and more than 1,000 other protesters, all of whom were being monitored by 70 police officers. The police presence turned out to be nowhere near sufficient to prevent the pushing, milling, fighting, and rioting that broke out both inside and outside the hall, and the disorder was marked by the throwing of stones, sticks, bricks, bottles, and other makeshift weapons. As a result of this tumult, Terminiello was arrested, tried, convicted for breach of the peace, and fined one hundred dollars.

On appeal to the Supreme Court, Justice William Douglas wrote for a 5-4 majority. Relying heavily on Stromberg v. California, the Court reversed Terminiello’s conviction, largely because the ordinance under which Terminiello had been convicted made it an “offense merely to invite dispute or to bring about a condition of unrest.” Chief Justice Fred Vinson and Justice Felix Frankfurter dissented, but it was Justice Robert Jackson’s dissent, joined by Justice Harold Burton, that focused most closely both on what Terminiello had said and on the disorder that had ensued. On his reading of the events, Justice Jackson concluded that the existence of a “riot and . . . a speech that provoked a hostile mob and incited a friendly one” justified Terminiello’s conviction. Justice Jackson did not, however, draw much of a distinction between the hostile and the friendly mobs, and so did not distinguish Terminiello’s responsibility for the actions of the supporters he incited from his responsibility for the reactions of the opponents he angered. And, thus, although the question of the hostile audience was clearly on the table in Terminiello, the case still produced little guidance about the extent to which a speaker might be responsible not for what he urged others to do but instead for what his words might prompt others to do to him and his supporters.

The Terminiello scenario was repeated, albeit on a smaller scale, several years later in Feiner v. New York, noteworthy in part for giving rise to the first actual judicial use of the term hostile audience. But again the facts made the conclusions to be drawn from the case less than crystal clear. Irving Feiner had given a speech in Syracuse criticizing President Harry Truman, the mayor of Syracuse, the American Legion, and others, but the chief import of his open-air address to a racially mixed crowd of 75 to 80 was “to arouse the Negro people against the whites.” Believing the crowd to be on the verge of erupting into a number of racially charged fights, the police first requested and then demanded that Feiner stop talking, but he refused to do so and was accordingly arrested for disorderly conduct. His conviction was affirmed by the New York Court of Appeals and then by the U.S. Supreme Court, the latter, in an opinion written by Chief Justice Vinson, observing that “[i]t is one thing to say that the police cannot be used as an instrument for the suppression of unpopular views, and another to say that, when as here the speaker passes the bounds of argument or persuasion and undertakes incitement to riot, they are powerless to prevent a breach of the peace.” Justice Hugo Black dissented, as did Justice Douglas, joined by Justice Sherman Minton, all of them relying heavily on Cantwell and on their belief that the real problem was the presence of police who saw their job more as siding with the hostile portion of the audience than as protecting the speaker.

Feiner is commonly taken to represent the extreme case of allowing the hostile audience to exercise the so-called heckler’s veto, but a close reading of the opinion makes that conclusion not nearly so obvious. Was Feiner charged with provoking a hostile audience or with inciting a sympathetic one? Was his wrong in giving his speech or in disobeying the plausible orders of the police? Could he, as a lone speaker, have been physically removed rather than arrested? Should he have been?

Although Feiner is hazy in just these ways, the decisions that mark Feiner’s burial contain fewer uncertainties. In three different 1960s cases dealing with civil rights demonstrators confronted by unambiguously hostile audiences, the Supreme Court appears to have eviscerated Feiner of whatever authority it may have had. In Edwards v. South Carolina, an 8-1 Supreme Court reversed the breach of the peace conviction of 187 Black civil rights demonstrators who were marching and singing in Columbia, South Carolina. The police, fearing impending violence by the hostile audience, threatened the demonstrators with arrest if they did not disperse, and the arrest, trial, and conviction of the demonstrators followed their refusal of the police demand. For Justice Potter Stewart, this case was “a far cry from the situation in Feiner and seemed to him to involve “no more than that the opinions which [the demonstrators] were peaceably expressing were sufficiently opposed to the views of the majority of the community to attract a crowd and necessitate police protection.” Quoting Terminiello, Justice Stewart concluded that criminal convictions of the speakers would be permitted only if there was a “clear and present danger . . . that rises far above inconvenience, annoyance, or unrest.”

Facts similar to those in Edwards were presented in Cox v. Louisiana, where again the leader of a civil rights demonstration, the Reverend Elton Cox, was arrested for, among other things, disturbing the peace in Baton Rouge, Louisiana. As a result of his arrest, Cox was sentenced to a cumulative one year and nine months in jail for leading a group of about 2,000 singing, cheering, and clapping demonstrators. In response to the city’s claim that action was taken against the demonstration because “violence was about to erupt,” Justice Arthur Goldberg again distinguished Feiner, pointing out that no actual violence had ensued, that the presence of local and state police appeared sufficient to prevent any violence, and that “constitutional rights may not be denied simply because of hostility to their assertion or exercise.” And thus by the time similar facts were once again presented in 1969 by a case involving a Chicago demonstration led by the comedian and civil rights activist Dick Gregory, Chief Justice Earl Warren was able to begin his majority opinion, an opinion that followed Edwards and Cox in all material respects, with the observation: “This is a simple case.”

Although some might suspect that the results in Edwards, Cox, and Gregory were heavily influenced by the Warren Court’s sympathy with the civil rights demonstrators and the causes they were espousing, that conclusion is belied by subsequent developments, most directly the Supreme Court’s 1992 decision in Forsyth County, Georgia v. Nationalist Movement. Here the demonstrators were members of a white supremacist organization who proposed to conduct a demonstration in opposition to the federal holiday commemorating the birth of Martin Luther King Jr. But because the county sought to impose a fee under a county ordinance allowing the amount of the fee to vary with the expected costs of law enforcement for the particular demonstration, the demonstrators refused to pay the fee and instead challenged the constitutionality of the ordinance.

Much of Justice Harry Blackmun’s majority opinion dealt with the question whether a facial challenge was appropriate, but the most important aspect of Forsyth County for present purposes is the Court’s conclusion that imposing a fee based on the expected costs of police protection and other security constituted impermissible content regulation, given that the amount of security required would plainly vary with the content of the speech. And although Chief Justice William Rehnquist dissented, joined by Justices Byron White, Antonin Scalia, and Clarence Thomas, the thrust of the dissent was that no such content-based fee had been imposed in this case, or even in previous demonstrations under the ordinance. And so although we do not know how the dissenters would have voted had such a fee been imposed here, it is at least plausible to speculate that in 1992 all nine members of the Supreme Court believed it a violation of the First Amendment to impose upon demonstrators the cost of police protection and related security to control a hostile audience whose hostility was in response to the content of an otherwise constitutionally protected speech, protest, march, demonstration, parade, or rally.

On the surface, it appears that we now have a considerable amount of law on the problem of the hostile audience, a body of doctrine going back at least to the 1930s and continuing through a collection of modern lower-court cases applying what seems to be Supreme Court doctrine through Forsyth County. But appearances can be deceiving. In important respects, the doctrine leaves substantial and increasingly salient questions unanswered. For example, almost all of the Supreme Court decisions described above rely heavily or at times exclusively on the vagueness of the relevant governing standards and thus on the consequent worry that such vagueness will allow excess official discretion, a discretion that would create an undue risk of impermissible content-based (and especially viewpoint-based) discrimination among different types of demonstrators. But a vagueness determination is often the path of least resistance, allowing courts to skirt the direct question of which kinds of non-vague restrictions are permissible and which are not. As a consequence, we know less than we might—here and elsewhere—about just what kinds of specific (non-vague) standards would be permissible or just how specific the standard must be to avoid invalidation on vagueness grounds.

Other “holes” in the existing doctrine are equally apparent. We know that law enforcement must take all reasonable steps before shutting down a speaker whose words have created a dangerous situation, but we know little about what counts as reasonable; what degree of deployment a police department is required to use; and whether a local police department is required to call upon county law enforcement, state law enforcement, or state military (i.e., the National Guard) forces before taking action against the speaker. Relatedly, if law enforcement fails to protect the speaker, what state and federal remedies, if any, are available to the speaker against the relevant officials and agencies? What follows is an attempt to explore these and other questions now arising in the context of a growing number of increasingly hostile confrontations between speakers and those who would protest and demonstrate against them.

How Much Law Enforcement Is Enough?

As recent events make clear, the modern problem arising from demonstrations, rallies, and even individual inflammatory speakers is rarely that speakers will urge their listeners to take unlawful actions against third parties, as was the case in Brandenburg v. Ohio. Nor is it that listeners may be offended, disgusted, or shocked, as in Cohen v. California, although that is surely often the case. Rather, the problem, and especially the First Amendment problem, is that offended, shocked, disgusted, angry, or strongly disagreeing listeners (or viewers) will threaten or take violent action against the initial speakers and disorder will ensue, as occurred in the cases described above, and as probably would have occurred had the Nazis actually marched in Skokie in 1977. Often the accounts of such events are unclear and conflicting, so it is probably wise to include those events in which audience reactions eventually lead to violence and disorder, even as it is sometimes unclear exactly who, literally or figuratively, has thrown the first punch. Still, the basic idea is that a speaker’s words or images are sometimes viewed as highly controversial, offensive, or harmful by some portion of an audience, and genuine physical violence or the imminent threat of it ensues.

One thing that is now settled, if anything is settled, and that the post-Feiner cases make reasonably clear, is that law enforcement may not initially or prematurely arrest the speaker. It also appears settled that the state may not prosecute on grounds of incitement the speakers whose speech has prompted the reactions. To do so would be to create what we now call the heckler’s veto, allowing those who would protest against a speech the right, in effect, to stop the speaker from speaking. And it is probably settled as well that even if getting the speaker to stop speaking would be genuinely necessary to prevent violence, punitive actions against the speaker may not take place until after the speaker has been given the opportunity to leave the venue on his or her own and, perhaps, absent the speaker’s willingness to depart, until after there has been an attempt at a non-punitive removal of the speaker.

But again, although this much seems more or less clear, what is importantly not clear is just how much the authorities must do before taking some sort of action against the speaker or before bringing the entire event to a halt. In some past cases, as in the civil rights demonstration cases, the authorities have done little, often because of greater police sympathy with the objectors than with the initial speaker. But even assuming good faith and the best intentions on the part of a municipality, how much must that municipality do? Must it deploy all or most of its police force, even if doing so would create other dangers to safety and security? Must it call on non-local authorities, such as county police, state police, or even the National Guard?

These problems come to us now with real numbers and real dollars attached to them. The University of California at Berkeley estimates a cost in the vicinity of $2 million to provide law enforcement and related services for the actual or planned speeches by only three individuals—Ann Coulter, Ben Shapiro, and Milo Yiannopoulos. The city of Charlottesville spent $70,000 on the widely publicized Unite the Right rally on August 12, 2017, as well as $30,000 for providing extra police and other services for a July 8 march by the Ku Klux Klan. It might once have been said that claims about the expense of providing protection for speakers likely to cause a violent confrontation with a hostile audience were prone to exaggeration by state and local officials eager to prevent the speech in the first place, but today such suspicions rest on a flimsier basis. Now, with the benefit of actual events and hard numbers, we can understand that officials who are worried about monetary costs, as well as about physical danger and psychological distress, will be concerned that every dollar spent on protecting unpopular or even hateful speakers will be a dollar unavailable to seemingly—at least to them—more worthy purposes.

The question of “How much must the authorities do?” in such situations involves a complex intersection of financial, logistical, personnel, and jurisdictional considerations. In the typical hostile audience scenario, the size of the hostile audience is substantially greater than the size of the group of “primary” speakers. And often the size of the hostile audience dwarfs the size of the local police force. In such cases, local law enforcement appears to have a series of choices, and it will be useful to explore several alternatives.

One possibility, when faced with the prospect of a large and potentially violent counter-demonstration, is simply for local law enforcement to do the best it can with its existing resources and personnel. If violence breaks out during the event or is about to break out imminently, the police could then, but not earlier, use whatever local mechanisms are available—in Charlottesville it was the declaration of an “unlawful assembly” —to bring the event to a halt and to restrict or apprehend those who do not obey lawful orders to disperse. Assuming that the police do not selectively single out the initial speaker (or speaking group) for apprehension or restriction, and assuming that any anticipatory halting of the event requires genuine immediate prospects of violence, there appears to be no First Amendment problem with such an approach. The police do, after all, break up fights all the time, and no one has even suggested that this creates a First Amendment problem, even if, as is so often the case, the fight was ignited by an exchange of what would be, in other contexts, constitutionally protected speech.

For admirable reasons, however, law enforcement officials would generally prefer preventing fights to breaking them up, and here matters become more difficult. Assuming, realistically, that officials anticipate disorder prior to some event, and assuming, more and more realistically these days, that local law enforcement believes that it is unlikely, with existing personnel and resources, to be able to control the violence once it erupts, what should the officials do? What must they do? As we know from the cases described above, the government cannot, under existing First Amendment doctrine, simply sanction the speaker or refuse to grant a permit or otherwise refuse to allow the event to take place. And as we know from actual events, local law enforcement will typically request support from regional and state law enforcement, such requests being commonly granted if reasonable. But the costs of this extra-jurisdictional or cross-jurisdictional support are considerable, and at some point a city, county, or state will not have the available personnel or funds (however “available” is defined). When this point is reached, can a city then deny a permit or otherwise prohibit the event? And assuming that such decisions will eventually come before the courts, can a city, county, or state argue that providing more than some amount of law enforcement would take funds away from school lunches? From emergency medical services? From public housing? From low-income welfare assistance? From providing services that themselves have a constitutional aspect, such as providing a certain quality of unpaid legal assistance for defendants in criminal cases or, in earlier times, providing police protection for Black children seeking to attend previously segregated schools? Now that we know the actual costs involved in places such as Berkeley and Charlottesville, we cannot avoid confronting the fact that protecting certain speakers exercising their First Amendment rights will come at a cost to other worthy municipal goals, including, at times, protecting or enforcing other constitutional rights.

These problems are exacerbated by the fact that more extreme speakers are likely to require more protection from hostile audiences. And thus, perhaps perversely, the more extreme a group’s views are, the greater will be its ability to call upon public resources for its protection, and the greater the amount of the resources it will require for its protection. And thus, even more ironically, it may turn out that difficult decisions of allocation of state and local resources are being made not by legislatures, and not by courts, but by those who are least representative of the public whose resources are being expended.

Alternatively, law enforcement might at some point simply say to prospective demonstrators that it can provide only so much protection and that any injuries caused by disruption beyond the ability of the police to control the audience will simply have to be borne by those who have suffered them. But although local police may on occasion stand back and let a small number of combatants in a bar fight simply fight it out, such an approach seems morally, politically, and perhaps constitutionally unacceptable both for bar fights and for larger events. And as long as that is so, then there is no avoiding the problem occasioned by the fact that providing security for the exercise of First Amendment rights will draw resources away from other uses.

Faced with this inevitability, the entity responsible for the additional costs might attempt to pass them on to others. The University of California or the University of Virginia could raise tuition, presumably following the lead of the car rental companies and the airlines with their fuel price fees, by calling the tuition increase a “First Amendment security fee” or something like that. But universities operate in competitive and price-elastic markets and, thus, attempting to pass costs on to students—or on to faculty by cutting salaries, for example, or on to a university’s athletics fan base by raising ticket prices or reducing the number of football and basketball scholarships—will involve other sacrifices. So too with cities and towns, especially ones that seek to attract business by offering lower taxes or special tax breaks or rebates. Here again the incentives are almost all in the direction of municipalities attempting to reduce the costs of what the courts have said they must do in the name of the First Amendment, and as long as it remains generally constitutionally impermissible for federal courts to order state and local governments to raise taxes, the problem appears ever more intractable.

These problems of allocation and prioritization are beyond easy solution, but they have arisen in analogous contexts. Most relevantly, the problem of constitutionally mandated and potentially unpopular allocation and reallocation of scarce resources has been widely discussed in the context of judicially enforced social welfare rights. Such rights—to housing, education, welfare, and health care, for example—are not much part of the American constitutional landscape, but they have a larger presence elsewhere. And thus when the Constitutional Court of South Africa mandated greater governmental expenditures on housing, it forced the South African Parliament to allocate scare resources among, say, housing, education, food, and health in a way other than that which might have been chosen by a judicially unconstrained Parliament. And when the same court required the government to provide anti-retrovirals to those suffering from AIDS, it inevitably drew resources away from those suffering from other illnesses. This is not the place to replay the many debates already existing on the subject, but it is a reminder that the constitutionalization of an interest will produce resource reallocations, and when constitutionalization is accompanied by judicial enforcement, the consequence, sometimes wise and sometimes not, is judicial determination of social resource allocations. Of course existing First Amendment doctrine itself involves judicially mandated reallocation of resources in the direction of speech protection, as compared to how most legislatures would allocate resources absent First Amendment constraints. But even if we take some considerable amount of American-style free speech protection as the baseline, it is worth noting the incremental resource allocation decisions that courts are making when they require substantial protections of speakers in the face of increasingly hostile audiences.

More precisely, therefore, the basic normative tension produced by the hostile audience scenario arises from the way in which requiring protection for those who would exercise their First Amendment rights requires a government to spend more than it would have spent had it possessed the ability simply to prohibit the speech—and requires the government either to secure additional resources for such protection or to reallocate resources toward protecting highly provocative speakers and away from other potentially worthy goals. Moreover, as noted above, the problem is not that this reallocation is at the direction of a less popularly responsive court. Rather, it is that under existing doctrine, such First-Amendment-driven reallocations of governmental resources turn out to be substantially in the control of the speaking groups themselves, and so the greater the provocation, the greater the reallocation. Provocative speakers have the ability to capitalize on this phenomenon by forcing the government to spend money for their protection, money that might otherwise be spent in ways the provocative speaker disfavors.

Part of the dilemma of the hostile audience, and part of the dilemma of the resource reallocation that dealing with a hostile audience occasions, is created by the reluctance of officials to impose very much in the way of serious penalties when actual violence breaks out at a rally or demonstration. The possible causes for this reluctance are multiple. Some might stem from a general unwillingness of law enforcement to employ arrest and prosecution as a remedy for street disturbances, regardless of their nature. Some might be attributable to law enforcement sympathy with counter-protesters and hostility to the speakers whose words and symbols have caused outrage in others. And some might arise from the difficulty in identifying actual perpetrators under circumstances in which angry words and then angry gestures and then physical contact appear to come from both sides. But whatever the cause, it is impossible to completely discount the possibility that the actual or impending violence that creates the need for an expensive law enforcement presence is related to the fact that for many members of a hostile audience the initiation of physical contact, often by throwing various projectiles or wielding sticks and poles, seems to them to be a relatively risk-free act, legally if not physically. And thus, perhaps again ironically, a hostile audience may at times draw resources away from governmental goals of which members of the hostile audience might very well approve.

On the Responsibilities—Legal and Otherwise—of the Hostile Audience

Although actual physical violence has, tragically, ensued from rallies and demonstrations in Charlottesville and elsewhere, far more often a hostile audience manifests its hostility in non-physically-violent ways. Sometimes such tactics include impeding speakers’ access to the designated location for the demonstration, as happened in Charlottesville when a group of religious leaders locked arms to block the white supremacists from reaching the park where the rally was to occur.

This type of obstruction raises few conceptual difficulties. Such blocking or obstructing is typically unlawful, and the reluctance to sanction the obstructers is often a combination of law enforcement sympathy for their cause and law enforcement fear of embarrassing publicity if the obstructers must be carried away or otherwise physically removed, a fear heightened in an age when everything is photographed by someone with a smartphone. But there is little doubt that blocking can be subject to sanctions of some variety, and we would scarcely hesitate before concluding that the police and the legal system may intervene to punish or restrict those who would keep people from getting to their home, their place of business, or the bus stop. There should be even less doubt about the impermissibility of the police responding to the blocking by sanctioning or moving the speakers who are being blocked. If speakers are prevented by others from getting to where they have a legal and constitutional right to be, the remedy cannot, at least in theory, be to tell the speakers that they must go elsewhere.

There remains an interesting question, however, as to what private remedies are available against those who would obstruct others in the exercise of their constitutional rights. If the obstruction is physical, presumably an injunction would be available, but the typical demonstration scenario involves little notice in advance and takes place in time periods shorter than the ability of formal legal processes to respond. There could still, however, be after-the-fact common law or statutory remedies that might, depending on the circumstances, undergird an action for false imprisonment, assault, creating a public nuisance, or possibly interference with advantageous relations, even though some or all of these, in most circumstances, might be somewhat of a stretch as a matter of tort law. Initially plausible might be some variety of civil rights action, given that blockers are preventing the exercise of a constitutional right. And we might think that the most likely civil rights action would be based on, ironically, Section 1985(3) of Title 42 of the United States Code, which creates a private cause of action against those who would conspire to deprive others of their civil rights. But Section 1985(3) does not on its face apply to those who would deprive others of all or any of their civil rights, only to those who would conspire to deprive others of the “equal protection of the laws” or of their “equal privileges and immunities under the laws.” Although the latter phrase might be argued to apply, especially if we conveniently ignore the word “equal,” the far more likely textual conclusion, and one supported by the cases, is that neither Section 1985(3) nor any other federal statute makes it a crime, or creates a private cause of action, when private persons interfere with the free speech rights of others.

The unavailability of a federal civil rights action does not mean that a state could not create such an action, or that a community could not enact an ordinance prohibiting such interference, or that a police officer could not issue a lawful order to counter-protesters to stop interfering, the violation of which would constitute the independent offense of disobedience of a lawful order of a police officer. But although some number of universities have or are now creating rules forbidding students from interfering with speakers at public university events, there do not yet seem to exist parallel state or local laws explicitly prohibiting obstructing the exercise of free speech rights. At the present time, therefore, those who would wish for sanctions against interference must rely on controversial and expansive interpretations of existing criminal law or tort law. None of this is to say that enactment or use of such laws would necessarily be wise as a matter of policy. The risks of such laws may well outweigh their potential benefits, and their misuse might well overwhelm their apt uses. But even if such laws—at least on the state or national level, as opposed to, say, university regulations regulating members of the university community—are in the final analysis unwise, it is worth thinking carefully and critically about the consequences of not providing speakers with remedies, apart from those available under general criminal and tort law, against those who would interfere with the exercise of their constitutional rights.

An even greater problem, however, even assuming the existence of some sort of rule prohibiting interference or disruption, concerns how we would define the interfering or disrupting behavior. Until now we have been considering only physical interference, where the application of various existing criminal, civil, and regulatory laws might be plausible, but what of the counter-protesters who bang on drums, engage in loud yelling, or create deafening noises with air horns? The use of such aural obstruction is these days as common as it is effective, but again a question arises about whether there is (or should be) a remedy for this kind of obstruction. Presumably a venue might impose a content-neutral noise regulation, along the lines of the regulation upheld for the public streets in Kovacs v. Cooper, but such a regulation would be most plainly constitutional only if it restricted the original speakers as well as those who are trying to keep them from being heard. The more difficult question is whether a content-neutral restriction on those who would, with noise or signs or otherwise, interfere with the lawful speech of a speaker can be sustained against the claim that the objectors have no fewer free speech rights than does the original speaker. We worry about the so-called heckler’s veto, but the First Amendment, after all, presumably protects the hecklers as well as the hecklees.

Although it is plain that hecklers have free speech rights, specifying just what those rights are presents more difficult issues. If a protester has a free speech right to post signs, does a counter-protester have the free speech right to remove the signs or paint over them? If a protester has a right to speak, does a counter-protester have a free speech right to drown out the original speaker with horns or drums? In the clear cases of sign removal or plain drowning out, the argument that hecklers or counter-protesters have free speech rights to prevent a speaker’s (or writer’s) message from being heard or seen seems strained. As long as reasonable and content-neutral time, place, and manner restrictions are permissible, then some sort of first-come, first-served regulations would seem constitutionally permissible as well. But that which is constitutionally permissible may remain politically or morally problematic, and we might wonder whether privileging those who get there first, or apply for a permit first, will reward those who have the resources to be, figuratively and occasionally literally, first in line. As a constitutional matter, however, such objections seem less powerful than the argument for at least the reasonableness of regulations aimed at making the right to free speech meaningful and designed to restrict those who would make its exercise ineffective.

Short of actual drowning out by sound or by obstruction or destruction of a written message, however, the question of how to deal with other forms of interference by hecklers or counter-protesters is more problematic. If there is a public concert on July 4 featuring the community band playing John Philip Sousa marches, may audience members bring their own instruments (or CD players) to play peace songs by Peter, Paul, and Mary at the same time and at the same volume? If the community orchestra is playing an overture from a Wagner opera, may members of an audience offended by Wagner’s anti-Semitic associations play annoying popular music—“The Little Drummer Boy,” for example—at the same time, even if at lower volume? And if we suppose that restricting such interference is permissible, can we distinguish these examples from someone playing “God Bless America” while a speaker is delivering an anti-war speech, or from someone who holds up the graphic homoerotic photographs of Robert Mapplethorpe during a public lecture about pretty much anything? In such cases, restriction of the counter-speaker seems often inappropriate and sometimes unconstitutional, but non-restriction may amount to the heckler’s veto in different clothing.

In dealing with such issues, much appears to depend on the nature of the forum, in the non-technical sense of that term. And thus restrictions on non-drowning-out interference that might be impermissible for public speeches might be permissible for public concerts, and sign restrictions that would be permissible for lectures in a lecture hall might not be permissible in a public park. But even when restrictions on disruption of or interference with speakers lawfully entitled to speak in a given venue are permissible, questions about what forms of behavior actually amount to interference will persist, although they may be questions of cognitive psychology more than questions of law. Consider, for example, the experiences of many of us when the person next to us at the theater, at a concert, or at the movies is loudly chewing gum. And what if, instead, those people were holding blocks of smelly Limburger cheese? Or pieces of dog poop? Or nibbling on the carcass of an uncooked dead bat? These increasingly disgusting examples are designed to suggest that interference need not come only from greater decibels produced by the audience, or from someone who literally obliterates someone else’s sign, but from a large number of other behaviors that some would consider only mildly annoying and easily ignored, but others would find as obstructing of their ability actually to hear and appreciate the speaker as the distractions caused by drums and loud horns. And the same is true of visual distractions: It is hard to imagine a pro-choice speaker not feeling interfered with were a large number of audience members to hold up signs portraying aborted fetuses.

The question of how much interference with free speech should be permitted is thus not only a question of physical interference, although such interferences appear to be the ones most amenable to legal remedies. It is also a question of which forms of non-physical interference are likely to be more or less disruptive, which should be sanctioned in some way, which should be encouraged, and which should be discouraged. But to answer those questions we would need to confront directly the largest issues of just what it is that a free speech regime—legal and otherwise—is designed to accomplish, a task that obviously transcends the hostile audience issues on which this paper has focused.

Conclusion: The Limits of the Law

There is more that can and should be said about the law on interference with or disruption of demonstrations, such as it is, but as is so often the case, the behavior that takes place in the world is often more or less than what the law permits or requires. Although this paper began with the hard law of the First Amendment as that law has in the past applied and does now apply to the problem of the hostile audience, the analysis has exposed some larger issues about speaker and listener behavior beyond the law. Under all but the self-expression justifications for freedom of speech, and thus under the arguments from truth-finding and from democratic deliberation, among others, free speech is about a certain kind of environment in which we learn from each other, deliberate with each other, and engage in various forms of collective communicative activity. In important respects, the actively hostile audience challenges not only the speaker but also the particular free speech-inspired legal and social rules according to which the speaker is protected in the first place. But whether that challenge, in the age of the neo-Nazis, in the age of the resurgence of the Klan, and in the age of the rise of other white supremacist speakers, is fundamentally sound or fundamentally misguided is a question that can hardly be answered in the context of this more limited inquiry.

© 2017, Frederick Schauer.


Cite as: Frederick Shauer, The Hostile Audience Revisited, 17-02 Knight First Amend. Inst. (Nov. 2, 2017), [].