Free Speech Overrides

I.     Introduction

The notion of an “absolute” First Amendment has been around for generations.1 First Amendment absolutism was championed, although not with exactly that term, by Justices Hugo Black and William O. Douglas.2 And numerous commentators, perhaps Alexander Meiklejohn most prominently,3 have joined the absolutist parade.4

Talk of an absolute First Amendment, however, is just that—talk. Even putting aside the obvious and by-now familiar point that the First Amendment’s Free Speech Clause does not even come close to covering all speech,5 the protection the Free Speech Clause offers even to communications within its scope—the communications that the First Amendment does cover—is not absolute now, has never been absolute in the past, and will not be absolute in the future. Rather, the protections of freedom of speech and freedom of the press—like the protections, prohibitions, and guarantees of other constitutional rights—are subject to being overridden by other considerations if those other considerations present themselves with sufficient weight and immediacy. In the context of equal protection, due process, and the free exercise of religion, for example, the threshold for overriding under the so-called strict scrutiny approach is typically the familiar “compelling interest” standard.6 Much the same applies in many contexts to speech covered by the First Amendment,7 and has ever since Oliver Wendell Holmes gave us the idea of “clear and present danger.”8

Recent events, with the one in my own city of Charlottesville being tragically the most notorious,9 make it important to think carefully about the kinds of dangers—harms—that can override what are undoubtedly rights under the First Amendment. At least as a matter of settled American free speech doctrine, for example, neo-Nazis,10 Klansmen,11 white supremacists,12 homophobes,13 puppy torturers,14 and endorsers of sexual violence,15 among others, have a right to publish their views and voice them in the public forum. Typically, as these examples illustrate, attempts to restrict such speakers have been met with the usually successful response that the speakers can only be restricted if the state can show that the speech would produce a harm of the greatest magnitude and immediacy, and that the harm could not be alleviated by any approach less restrictive of a speaker’s First Amendment rights.16 Importantly, governments have almost universally been unable to establish such a showing.17 Accordingly, it seems appropriate in light of recent events, especially those involving hostile audiences,18 to survey the existing doctrine and offer some guideposts as to what it would take actually to override the First Amendment in areas of its central coverage.

Yet if the rights to freedom of speech and freedom of the press can on occasion be overridden,19 then it follows that the possessors of such rights may sometimes wind up losing what the rights purport to give them. This in itself is hardly remarkable, as this conclusion flows logically from the nonabsolutism of the underlying right. But what is more noteworthy is that those whose First Amendment rights are overridden, even when properly so, wind up losing something—they lose what the First Amendment guarantees them. Yet despite having lost the opportunity to exercise their First Amendment rights, they still receive nothing to acknowledge their loss, and certainly nothing to compensate them for that loss.

Although such non-compensation or other redress for the loss of the ability to exercise a constitutional right seems so familiar as to fail to even generate concern, it does stand in contrast to how we treat those who have given up their property rights for the public good. In those instances, the so-called Takings Clause of the Fifth Amendment, applied to the states through the Fourteenth Amendment,20 provides that those whose property is taken by eminent domain, even if the taking is justified, are nevertheless entitled to “just compensation.”21 But if those whose property rights are taken for the public good are entitled to compensation for their loss, then why are not those whose First Amendment rights are similarly taken (or restricted) for the public good also entitled to compensation? That is a puzzle, and a secondary goal of this Article—although one that emerges directly from the phenomenon of the override—is to present and examine this puzzle.

II.   It All Started with Holmes

When Oliver Wendell Holmes first used the now-familiar phrase, “clear and present danger,”22 it was for him not a carefully considered choice of words at all. In Schenck v. United States,23 and then in Debs v. United States24 and Frohwerk v. United States25 only months later, Holmes treated the prosecutions as largely controlled by existing principles of criminal law. As in the criminal law, the defendant’s intent was crucial, but Holmes, having found that Charles Schenck, Eugene Debs, and Jacob Frohwerk all possessed the necessary intent to sustain their convictions,26 did not treat the First Amendment claims of all three of these defendants as worthy of serious consideration. So when Holmes mentioned that speech could be restricted when a clear and present danger existed, it was, at the time, little more than an aside.27 That Holmes wrote for the Court in upholding all three convictions underscores that he did not imagine that the idea of a clear and present danger imposed very much of an impediment to a conviction that was permissible under standard criminal law principles. Indeed, the fact that a variant of clear and present danger appears in Holmes’s subsequent change of heart in Abrams v. United States28 only in the disjunctive29 further emphasizes that at the beginning of the modern First Amendment the idea of clear and present danger did not do very much work.

Given the results in Schenck, Debs, and Abrams, the idea of a clear and present danger appears as a highly permissive standard.30 In theory, it need not be so. After all, under the “rational basis” test, the test that is generally applicable to the evaluation of government restrictions on conduct not covered by the First Amendment,31 the state is permitted to take actions against dangers that are neither clear nor present. Rational basis review allows the state to speculate with respect to dangers that are not clear and to regulate for future dangers that are not present. Few would argue these days, for example, that government may not regulate the sale of electronic cigarettes or foods made with genetically manufactured organisms (GMOs), even though the alleged dangers of such products, being contested and speculative, are certainly not clear.32 Even more obviously, the government plainly may take restrictive actions to combat the dangers of climate change, even though the clear dangers of climate change are not “present” under any ordinary understanding of that word.33 As a result, and contrary to the actual results in the 1919 cases, it seems now safe to conclude, as the Supreme Court and other courts concluded in the 1960s,34 that the idea of a clear and present danger is such as to require for the regulation of speech covered by the First Amendment a showing of gravity, immediacy, and specificity substantially greater than the showing sufficient to justify the regulation of non-covered behavior.35 It is far too late in the doctrinal day to believe that speech is protected because it is harmless, and thus that any harm-producing speech loses its protection for that reason.36 Rather, even harmful speech is routinely protected, and the import of the clear and present danger idea is that the harms must be especially great and especially immediate for the protection typically available for harmful speech to be forfeited.

III.  Clear and Present Danger Revised—and Not

In Schenck, “clear and present danger” may have been little more than the relatively casual observation that the First Amendment was not absolute, but it soon became an actual test or criterion against which restrictions on covered speech were to be measured. Initially, the view that “clear and present danger” was a constitutional test rather than merely an observation emerged in a series of dissents. First was the dissenting opinion of Justice Brandeis, joined by Justice Holmes, in Schaefer v. United States,37 explicitly referring to “clear and present danger,” in objecting to the majority’s affirmation of the conviction of a wartime dissenter.38 And Brandeis relied on the then-recent article by Zechariah Chafee for the proposition that clear and present danger should properly be understood as the test for the constitutionality of a restriction on advocacy, even in times of war.39 To much the same effect, shortly thereafter was Pierce v. United States,40 where Brandeis, again joined by Holmes, once more used explicit “clear and present danger” language41 in departing from the majority’s conclusion that Pierce’s pamphlets were intended to produce military insubordination as their “proximate result”42 and that a jury could find that those pamphlets could have a “material influence”43 on such insubordination. And in the same year, Brandeis still again dissented, here in Gilbert v. Minnesota,44 continuing to use “clear and present danger” as the description of the test from which he believed that the majority had departed.45 Brandeis reiterated that position several years later in his enduring “concurring” opinion in Whitney v. California.46 In Whitney, Brandeis seemed to follow Holmes’s decision in Gitlow v. New York,47 where Holmes referred to “clear and present danger” in no uncertain terms as the “criterion” and “test” for all restrictions on advocacy,48 not only those in which a speaker was prosecuted under a general statute not restricted to speech, as in Schenck, but also those in which the legislature had made a finding of the dangers resulting from speech of a certain kind.49

The post-Schenck version of the clear and present danger standard appeared to have been discarded when a Supreme Court plurality in Dennis v. United States50 relied on the “gravity of the evil discounted by its improbability” standard employed by Judge Learned Hand in the decision below.51 However, it in fact persisted after Dennis: Something very close to a strong version of the clear and present danger idea, arguably strengthened even further, was to be found in Yates v. United States52 in 1957, and then in Scales v. United States53 and Noto v. United States,54 both decided in 1961. And although Yates, Scales, and Noto were undeniably more speech-protective than Dennis and Gitlow, the standard they embodied was still not absolute and the idea that behavior covered by the First Amendment could still be regulated, and thus that First Amendment rights could be overridden under some circumstances, still persisted.

In a narrow sense, Schenck is no longer good law. The specific context in which the clear and present danger standard first arose in Schenck—the advocacy of unlawful conduct—was and remains superseded by the test that emerged from Brandenburg v. Ohio.55 The Brandenburg standard, arguably incorporating some version of the requirement of explicit incitement first introduced by Judge Hand a half-century earlier in Masses Publishing Co. v. Patten56 and still retaining (and strengthening) the evidentiary (“clear”) and temporal (“present”) dimensions of the clear and present danger idea,57 superseded Schenck and remains the applicable rule today.58 Indeed, to the extent that lower courts have tended to apply Brandenburg to civil cases involving negligent causation of unlawful acts,59 the case has emerged as an even broader and stronger protection of speech bearing a relationship to subsequent acts of illegality. Even so, however, the test is not absolute, and it remains possible, at least in theory, for even Brandenburg to permit the First Amendment to be overridden in cases of intentional, explicit, advocacy of immediate substantial illegality when such illegality is likely to occur.60

IV.  The Persistence of Clear and Present Danger

Although the test set forth in Brandenburg has superseded clear and present danger as the standard to be applied to putative restrictions on the advocacy of unlawful conduct, it would be a mistake to assume that Brandenburg represents the complete demise of the clear and present danger test as an actual standard to be applied today to actual restrictions. In some number of domains, the clear and present danger standard persists, largely because the basic idea of requiring reasons of special strength to override the First Amendment remains central, even though not all reasons of special strength fit the Brandenburg formula, designed as it is to deal with the specific problem of advocacy of unlawful conduct.

Consider, for example, the line of cases dealing with speech that has the potential of interfering with the judicial process. It is now wisely as well as widely accepted that newspaper and other public comments about trials and judges, even during the pendency of the trial, are protected by the First Amendment.61 In reaching this conclusion, the Supreme Court has explicitly established that clear and present danger to the administration of justice is the relevant standard.62 And although the cases so holding predate Brandenburg, it seems plain that the Brandenburg formula would fit poorly with a situation in which the potential danger is to the impartiality of judges and jurors, and is not that some reader or listener will engage in unlawful acts against those judges or jurors (or litigants). When the Court in Cox v. Louisiana63 suggested that physical parading and picketing outside a courtroom or a courthouse might be governed by different standards,64 it appeared implicitly to reaffirm that clear and present danger would be the standard applied to so-called pure speech about pending trials.

Although the Cox majority treated the physical aspect of parading and picketing as grounds for its ambivalence about the applicability of the clear and present danger standard, that ambivalence seems a bit surprising. Twenty-five years earlier, in Thornhill v. Alabama,65 the Court did indeed discuss clear and present danger as the standard appropriate to a situation in which the petitioners’ labor-related picketing was held to be protected.66 Cox thus appears as a slight anomaly, and a fair conclusion to be drawn from the cases just discussed—none of which have been overruled or questioned—is that clear and present danger still has its place even after Brandenburg, and that the Brandenburg formulation—for all of its enduring importance—still might be understood as an exception to a more pervasive and persistent clear and present danger approach.67

Much more importantly, however, clear and present danger retains continuing—indeed, increasing—vitality in the context of what has come to be understood as the problem of the hostile audience.68 The paradigm application of Brandenburg, a paradigm going back to Schenck, is to a speaker (or writer) addressing an actually or potentially sympathetic audience and urging that audience to action. Charles Schenck, Eugene Debs, Jacob Frohwerk, and Jacob Abrams, for example, each tried to persuade those who were already inclined to share their socialist or anarchist or anti-war proclivities to put those proclivities into action by resisting the draft or in other ways interfering with the war effort.69 And Clarence Brandenburg, speaking to his fellow Klansmen (and maybe some cows) on a field in southern Ohio, was prosecuted for, again, encouraging predisposed followers to unlawful action.70

What makes this characterization of the line of cases from Schenck to Brandenburg interesting here is precisely the fact that not all danger-producing speakers produce that danger by encouraging, urging, or inciting their sympathetic followers to take socially detrimental and typically unlawful actions. Even putting aside the cases typically applying Brandenburg to civil actions seeking to hold speakers (or, typically, publishers) liable for negligently inspiring or facilitating unlawful action,71 there are many instances in which violence is the genuinely unintended (by the speaker) and truly undesired (by the speaker) byproduct of an otherwise lawful speech. Typically this occurs when an audience reacts violently to what a speaker non-violently has said, and this, in a nutshell, is the problem of the hostile audience.

The hostile audience problem has been around and generating Supreme Court opinions for almost a century. Early on, Feiner v. New York72 held it permissible to restrict the speaker in order to prevent violence brought about by an audience angry at the speaker (and thus not incited or encouraged by the speaker).73 But a series of cases in the 1960s involving civil rights demonstrators and marchers effectively overruled Feiner, and required that restrictive actions in cases of hostile and potentially (or actually) violent reactions to speakers be directed not against the speaker, but against those who engaged in or threatened to engage in the reactive violence.74

As recent events have made clear, the problem of the hostile audience is not only still with us, but increasing at a rapid rate.75 And thus the question persists—in an age of burgeoning listener violence—as to when speakers might be restricted in order to deal with audience violence, or, more commonly, when an entire event might be shut down, thus restricting the speakers as well as the audience.

Here again, it turns out that the clear and present danger standard may still be with us. In what is perhaps the first hostile audience case, Cantwell v. Connecticut,76 the Supreme Court explicitly used clear and present danger as the standard to be applied when violence is threatened by those who react negatively to a speaker’s speech.77 And not only did Terminiello v. City of Chicago78 nine years later employ the same standard,79 but in the same year so also did Feiner v. New York, even if the subsequent cases of the 1960s have made clear that the Feiner Court’s toothless application of that standard could not satisfy the requirements of the First Amendment.

The fact that neither Cox, nor Edwards, nor Gregory employed clear and present danger language in casting grave doubts on Feiner suggests that the best conclusion, in light of Cantwell and Terminiello, is simply that the question remains open. And that conclusion is supported by the way in which lower courts have wrestled with the issue, with some of those courts concluding that clear and present danger remains the test for when a speaker or an event can be closed down because of the reactions of a hostile audience,80 while other courts and judges do not mention clear and present danger in the process of protecting speakers from restrictions arising out of the reactions of a hostile audience.81

Although the law remains unfortunately unclear on the issue, it is hard to imagine that speakers (or the events at which they are speaking) can never be restricted because of the actual or potential reactions of a hostile audience. As a result, perhaps the best we can imagine as a workable standard is some version of a clear and present danger test combined with a least restrictive alternative approach. Consider, for example, a clear and present danger of violence that comes from the reaction of a hostile audience to a speaker who did not intentionally provoke that audience. Such a scenario, increasingly common, might justify not the immediate arrest of the speaker, but instead a dispersal order by law enforcement, the disobedience of which might then, and only then, justify actions against a speaker who disobeyed that order.82 Or, similarly, the existence of that clear and present danger might be grounds—subject to judicial review—for ordering speakers to change locations or times in the least restrictive way possible while still avoiding the danger, with, again, further restrictions on speakers (including prosecution) being permissible only if those orders to change times and/or places are disobeyed. And whether the exact language of “clear and present danger” is employed or not, a fair conclusion seems to be that at least some version of that idea must persist. When the just-described less restrictive alternatives cannot prevent audience violence, and when existing law enforcement resources are unable to do the same, then it would be hard to imagine that the ability of speakers to speak when and where they choose, even in the face of violence that reasonable law enforcement efforts cannot contain, is required by the First Amendment. And whether it is clear and present danger or some variant thereof that represents the standard, it is equally hard to imagine that Brandenburg, designed for a very different kind of problem, would be the starting or ending point of the analysis.

V.    The Question of Redress

There is much more that could be said about the problem of the hostile audience, and in light of recent events much of that is likely to be said in the near future by both courts and commentators. But rather than engage in further speculation, I want to examine a particular consequence of understanding First Amendment rights as overridable, and thus of understanding the holder of First Amendment rights as vulnerable to losing the ability to exercise those rights because of overriding circumstances. More particularly, I want to expose an anomaly in how we treat overridden rights, an anomaly especially apparent in hostile audience situations.

Whether it be by use of the clear and present danger test or with some other test yet to be developed, it seems plain that there are at least some instances in which speeches, parades, demonstrations, rallies, and the like can be ordered to close down or to move because of the reactions of a hostile audience. As a matter of state law, such responses by state and local law enforcement authorities are typically effectuated by means of a declaration of an unlawful assembly,83 but the exact details are not important here. What is important is that there are, and have been instances in which some of the consequences of actions by a hostile audience are such that speakers who would otherwise have First Amendment rights to say what they are saying will have those First Amendment rights restricted in some way because of the actual or potential reactions of their unsympathetic listeners.

Under such circumstances, we might then ask what is owed to those, including many whose moral profile is vastly superior to the “Unite the Right” demonstrators in Charlottesville, whose First Amendment rights have been curtailed through no fault of their own.84 If by virtue of what is now commonly labeled the “heckler’s veto”85 a group of speakers is justifiably restricted in the exercise of what would otherwise be within their First Amendment rights, then what rights of redress or compensation do the restricted speakers have?

As should be apparent, the answer to this question, as a matter of existing law and existing political practice, is “nothing.” If the reactions of a hostile audience rise to the level of a genuine clear and present danger, and thus if law enforcement is constitutionally justified in restricting the speakers by, for example, declaring an unlawful assembly and bringing the event to a close, the prevailing practice is that the restricted speakers are entitled to no compensation or other redress. Law enforcement having, by hypothesis, done the right thing, the state’s obligations come to an end.

But compare this scenario to the taking of land by eminent domain. If the state takes (or even, sometimes, restricts the use of86 ) someone’s land by eminent domain, then the land-owner who has been deprived of her land (and therefore her property rights) is entitled to “just compensation” by order of the Fifth Amendment, and that is so even if the taking was entirely justified.

The anomaly should now be apparent: the land-owner whose property rights are overridden for the public good is compensated, but the speaker or demonstrator whose First Amendment rights are overridden or restricted is entitled to nothing.

This anomaly might be explained in some number of ways. Perhaps the anomaly is a function of the longstanding belief that property is tangible and valuable in ways that rights are not.87 Perhaps it is a function of the ability to place a monetary value on the property taken in ways that would be far less possible for the deprivation of free speech rights.88 Or perhaps it is simply a matter of historical path-dependence or the power of the we’ve-never-done-it-before-so-we shouldn’t-do-it-now argument.

If none of these explanations are persuasive (and I offer them as explanations and not as justifications), then perhaps the anomaly between how we treat rightful property deprivations and how we treat rightful free speech deprivations could be “cured” in some way. Assuming that the Constitution prohibits limiting the right to compensation for takings of land, then the only other way to lessen the gap between what we do for property takings and what we do for speech takings is to at least think about compensating those who in some way have had their free speech rights diminished for the public good. And although we rarely think about this possibility—the possibility of compensating those whose free speech rights are overridden—it is a possibility that finds support from two other areas of thought.

One of these areas of thought is in private law, where the questions about Vincent v. Lake Erie Transportation Co.89 are about whether one who justifiably injures another’s property is required to compensate the owner of the damaged property despite the justifiability of the action. If, as in Vincent, someone who justifiably damages another’s dock in order to keep from foundering in a storm must nevertheless compensate the dock-owner for the damages caused, nevertheless, does the state analogously owe damages to those whose rights are justifiably overridden for the public good?

Once the question is posed this way, it becomes clear that there is also a relevant domain of philosophical thinking. Many of the philosophers who have thought about nonabsolute (and thus overridable) rights—Judith Thomson,90 Frances Kamm,91 and Walter Sinnott-Armstrong,92 for example—have argued that when rights are rightfully overridden there is a moral residue,93 such that the infringer still owes something by way of compensation or other redress to the right-holder whose rights have been overridden. If these and other philosophers94 are right, then is there a constitutional residue when constitutional rights are overridden, such that the overrider—the state—similarly owes compensation even though the state has done the right thing?

In the context of this article and this symposium I do not propose to answer these questions here. But if we apply those questions specifically to free speech rights under the First Amendment, it turns out that the questions raised by Vincent v. Lake Erie Transportation Co. and the aforementioned philosophers are very real, especially in the context of the problem of the hostile audience.

One qualification is worth noting. In many instances of speeches or demonstrations that are justifiably restricted, it is the restricted speaker who has triggered the restriction, and it might seem odd to think that such a speaker is entitled to redress. If a modern-day Clarence Brandenburg intentionally and explicitly urges his audience to take specific and immediate violent action against African-Americans and Jews,95 he can be restricted according to the Brandenburg standard, but it would seem odd indeed to think that Brandenburg is entitled to compensation. But, to use another hypothetical (and decidedly counterfactual) scenario, if the hostile audience reactions against a modern-day Reverend Elton Cox96 are such that his otherwise protected demonstration must be curtailed, it seems less odd to think he might be entitled to something. Under existing doctrines and practices, however, Reverend Cox would get nothing. Civil rights actions would provide redress if the restriction were wrongful, but when the restrictions are rightful there is no route to a remedy, even if the injury to him—not being able to speak—is the same.

VI.  Conclusion

I have unsatisfyingly ended with a question to which I do not purport to provide an answer. Nor do I think that the question and the anomaly that generates it are the most important things to consider when we are addressing the kinds of issues that arise from the way in which free speech rights can be overridden. But the anomaly and the questions about how, if it all, to resolve them represent at least one potentially interesting corner of the larger question of free speech overrides generally. Given that the baseline free speech standards have become ever more speech protective, as the progression from Schenck to Brandenburg shows, it is easy to lose sight of the nonabsolute character of even the most highly speech-protective doctrines. But the hostile audience problem—no longer restricted to the epiphenomenal factual scenarios that characterized cases like Cantwell, Terminiello, and Feiner—is no longer an epiphenomenal problem, and considering the standards and consequences of the way in which free speech rights may be overridden turns out to be more germane than it was in the 1960s or even in the more recent past.

• 1See Norman Dorsen et al., Emerson, Haber, and Dorsen’s Political and Civil Rights in the United States 45 (4th ed. 1976).
• 2See, e.g., Brandenburg v. Ohio, 395 U.S. 444, 456–57 (1969) (Douglas, J., concurring); N.Y. Times Co. v. Sullivan, 376 U.S. 254, 293 (1964) (Black, J., concurring); Konigsberg v. State Bar of Cal., 366 U.S. 36, 61 (1961) (Black, J., dissenting); Barenblatt v. United States, 360 U.S. 109, 140–44 (1959) (Black, J., dissenting); Hugo Black, The Bill of Rights, 35 N.Y.U. L. Rev. 865 (1960); Hugo L. Black & Edmond Cahn, Justice Black and First Amendment “Absolutes,” A Public Interview, 37 NYU L. Rev. 549 (1962); Lucas A. Powe, Jr., Evolution to Absolutism: Justice Douglas and the First Amendment, 74 Colum. L. Rev. 371 (1974). In N.Y. Times Co. v. United States, 403 U.S. 713 (1971), Solicitor General Erwin Griswold, apparently addressing Justice Douglas, famously argued as follows: “You say that ‘no law’ means ‘no law’ and that should be obvious. I can only say, Mr. Justice, that to me it is equally obvious that ‘no law’ does not mean ‘no law,’ and I would seek to persuade the Court that that is true.” Transcript of Oral Argument in Times and Post Cases Before the Supreme Court, N.Y. Times, Jun. 27, 1971, at 24. Justice Black was of the opinion, however, that Griswold’s statement was addressed to him. See Gerald T. Dunne, Hugo Black and the Judicial Revolution 431 (1977).
• 3Alexander Meiklejohn, The First Amendment is an Absolute, 1961 Sup. Ct. Rev. 245 (1961).
• 4See, e.g., C. Edwin Baker, Human Liberty and Freedom of Speech 125–93 (1992); Zachary S. Price, Our Imperiled Absolutist First Amendment, 20 U. Pa. J. Con. L. 817 (2018); Solveig Singer, Reviving First Amendment Absolutism for the Internet, 3 Tex. Rev. L. & Pol. 279 (1999). The idea persists. See Tony Woodlief, Free Speech Absolutism Killed Free Speech, Wall St. J., Aug. 31, 2020, at A17.
• 5A great deal of communication, linguistic and otherwise, simply does not implicate the First Amendment at all. In my preferred terminology, such communication (much of which is “speech” in ordinary English) is not covered by the First Amendment, which is to be distinguished from those communications that are covered but wind up not being protected after the application of some First Amendment–inspired test. See Frederick Schauer, Out of Range: On Patently Uncovered Speech, 128 Harv. L. Rev. F. 346 (2015); Frederick Schauer, The Boundaries of the First Amendment: A Preliminary Exploration of Constitutional Salience, 117 Harv. L. Rev. 1765 (2004); Frederick Schauer, Every Possible Use of Language? in The Free Speech Century 33 (Geoffrey R. Stone & Lee C. Bollinger eds., 2019); Amanda Shanor, First Amendment Coverage, 93 NYU. L. Rev. 318 (2018).
• 6See, e.g., Roe v. Wade, 410 U.S. 113, 155 (1973) (holding that the government must meet the “compelling interest” standard when fundamental rights under the due process clause are infringed); Grutter v. Bollinger, 539 U.S. 306, 325 (2003) (holding that race–based distinctions are permissible under the equal protection clause only if they serve a compelling governmental interest); Adarand Constructors, Inc. v. Peña, 515 U.S. 200, 237 (1995) (same); Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 532 (1993) (holding that restrictions targeted at religious practices are permissible only if narrowly tailored to serve a compelling governmental interest).
• 7See, e.g., Riley v. Nat’l Fed’n of Blind, Inc., 487 U.S. 781, 798 (1988) (restrictions on charitable solicitations); Williams–Yulee v. Fla. Bar, 575 U.S. 433, 445 (2015) (restrictions on speech of candidates in judicial elections); Sable Commc’n of Cal., Inc. v. FCC, 492 U.S. 115, 125 (1989) (restrictions on allegedly indecent speech); Reed v. Town of Gilbert, 576 U.S. 155, 162–63 (2015) (content–based restrictions generally). Slightly more complex is New York v. Ferber, 458 U.S. 747, 761 (1982), in which the Supreme Court used the language of “compelling” interest to justify restrictions on non–obscene child pornography, and thus announced the general permissibility of such restrictions, but did not require a showing of a compelling interest in particular applications.
• 8See Schenck v. United States, 249 U.S. 47, 52 (1919).
• 9For accounts of the events arising out of the Unite the Right rally in August 2017, see Hunton & Williams LLP, Final Report: Independent Review of the 2017 Protest Events in Charlottesville, Virginia (2017), https://www.huntonak.com/en/news/final-report-i‌nd‌e‌p‌e‌‌nd‌‌en‌t‌-‌r‌e‌view-of-the-2017-protest-events-in-charlottesville-virginia.html [‌p‌e‌r‌m‌a‌.‌c‌c‌/‌3‌7‌8‌7‌-‌L‌Y‌KV‌]‌; see also Complaint, Sines v. Kessler, 324 F. Supp. 3d 765 (W.D. Va. 2018) (No. 3:17–CV–00072); Kessler v. City of Charlottesville, No. 3:17CV00056, 2017 WL 34754071 (W.D. Va. 2017); Frederick Schauer, In the Shadow of the First Amendment, in Charlottesville 2017: The Legacy of Race and Inequity 65 (Louis P. Nelson & Claudrena Harold eds., 2018).
• 10Collin v. Smith, 578 F.2d 1197 (7th Cir. 1978).
• 11Forsyth Cty. v. Nationalist Movement, 505 U.S. 123 (1992).
• 12See Hunton & Williams LLP, supra note  NOTEREF _Ref42002380 \h  \* MERGEFORMAT 9 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300032003300380030000000 .
• 13Snyder v. Phelps, 562 U.S. 443 (2011).
• 14United States v. Stevens, 559 U.S. 460 (2010).
• 15Am. Booksellers Ass’n v. Hudnut, 771 F.2d 323 (7th Cir. 1985), aff’d, 475 U.S. 1001 (1986).
• 16See Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 799 (2011) (holding that restrictions on content of violent interactive videogames could be restricted only if the particular restriction was “necessary” to serve a “compelling interest”).
• 17Thus, each of the cases cited in supra notes  NOTEREF _Ref51763141 \h 10 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360033003100340031000000 ,  NOTEREF _Ref51763155 \h 11 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360033003100350035000000 ,  NOTEREF _Ref51763176 \h 13 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360033003100370036000000 ,  NOTEREF _Ref51763187 \h 14 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360033003100380037000000 ,  NOTEREF _Ref51763194 \h 15 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360033003100390034000000 , and  NOTEREF _Ref51763206 \h 16 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360033003200300036000000 was one in which the government’s justification for its attempted restriction was invalidated. For recent examples, see Iancu v. Brunetti, 139 S. Ct. 2294 (2019); Matal v. Tam, 137 S. Ct. 1744 (2017); Reed v. Town of Gilbert, 576 U.S. 155 (2015).
• 18See Frederick Schauer, Costs and Challenges of the Hostile Audience, 94 Notre Dame L. Rev. 1671 (2019). For accounts of recent events, many on or near university campuses, see Jamal Greene, Constitutional Moral Hazard and Campus Speech, 61 Wm. & Mary L. Rev. 223 (2019); JD Hsin, Defending the Public’s Forum: Theory and Doctrine in the Problem of Provocative Speech, 69 Hastings L. J. 1099 (2018); Timothy E. D. Horley, Rethinking the Heckler’s Veto after Charlottesville, 104 Va. L. Rev. Online 8 (2018).
• 19See Boy Scouts of Am. v. Dale, 530 U.S. 640, 648 (2000) (using the exact language of “override”).
• 20See Chi., B. & Q. R. Co. v. City of Chicago, 166 U.S. 226 (1897); Missouri Pac. Ry. Co. v. Nebraska, 164 U.S. 403 (1896).
• 21U.S. Const. amend. V.
• 22Schenck v. United States, 249 U.S. 47, 52 (1919).
• 23249 U.S. 47 (1919).
• 24249 U.S. 211 (1919).
• 25249 U.S. 204 (1919); see also Schauer, supra note  NOTEREF _Ref42002380 \h  \* MERGEFORMAT 9 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300032003300380030000000 .
• 26For contrasting views on the relevance of speaker’s intent under the First Amendment, see Larry Alexander, Is There a Right to Freedom of Expression? 76 (2005); Larry Alexander, Free Speech and Speaker’s Intent, 12 Const. Comm. 24 (1995); Leslie Kendrick, Free Speech and Guilty Minds, 114 Colum. L. Rev. 1255 (2014); Leslie Kendrick, Speech, Intent, and the Chilling Effect, 54 Wm. & Mary L. Rev. 1633 (2013).
• 27See Thomas Healy, The Great Dissent: How Oliver Wendell Holmes Changed His Mind—and Changed the History of Free Speech in America 102–03 (2013). Indeed, the accompanying “shouting fire in a crowded theater” example was not even original with Holmes, having first appeared in the prosecutor’s closing argument in the Debs trial. Id. at 91.
• 28250 U.S. 616, 624 (1919) (Holmes, J., dissenting).
• 29Id. at 629 (“Even if I am technically wrong and enough can be squeezed from these poor and puny anonymities to turn the color of legal litmus paper . . . .”).
• 30See Richard Polenberg, Fighting Faiths: The Abrams Case, the Supreme Court, and Free Speech 212–18 (1987).
• 31See United States v. Carolene Prod. Co., 304 U.S. 144, 152–53 (1938); see also Ferguson v. Skrupa 372 U.S. 726 (1963); Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483 (1955).
• 32For information on GMOs, see Barbara de Santis et al., Case Studies on Genetically Modified Organisms (GMOs): Potential Risk Scenarios and Associated Health Indicators, 117 Food & Chemical Toxicology 36 (2018). For information on electronic cigarettes, see Jennifer Couzin–Frankel, How Safe is Vaping? New Human Studies Assess Chronic Harm to Heart and Lungs, Science Magazine, Science (Nov. 26, 2019), https://www.sciencemag.org/news/2019/11/how-safe-vaping-new-human-studies-assess-chronic-harm-heart-and-lungs [perma.cc/6MAP-N‌W‌ET‌‌]‌.
• 33[1]On the tolerance of the rational basis test for speculation, see Heller v. Roe, 509 U.S. 312, 320 (1993); Maria Ponomarenko, Administrative Rationality Review, 104 Va. L. Rev. 1399, 1399 (2018); John A. Robertson, Science Disputes in Abortion Law, 93 Tex. L. Rev. 1849, 1853 (2015). On the distinction between First Amendment standards and rationality review, see Felix T. Wu, The Commercial Difference, 58 Wm. & Mary L. Rev. 2005, 2036 n.145 (2017).
• 34Noto v. United States, 367 U.S. 290, 297–98 (1961); Scales v. United States, 367 U.S. 203, 229 (1961); Yates v. United States, 354 U.S. 298, 316 (1957); United States v. Spock, 416 F.2d 165 (1st Cir. 1969).
• 35See Frederick Schauer, Is It Better to Be Safe Than Sorry? Free Speech and the Precautionary Principle, 36 Pepperdine L. Rev. 301 (2009).
• 36See Frederick Schauer, Harm(s) and the First Amendment, 2011 Sup. Ct. Rev. 81 (2011).
• 37251 U.S. 466 (1920).
• 38Id. at 483, 486 (Brandeis, J., dissenting).
• 39Zechariah Chafee, Freedom of Speech in War Time, 32 Harv. L. Rev. 932, 963 (1919). On Chafee’s connections with Hand and Holmes at the time, see Stephen Budiansky, Oliver Wendell Holmes: A Life in War, Law, and Ideas 369, 385, 393 (2019).
• 40252 U.S. 239 (1920).
• 41Id. at 255, 271, 272 (Brandeis, J., dissenting).
• 42Id. at 250.
• 43Id.
• 44254 U.S. 325 (1920).
• 45Id. at 335, 336, 338 (Brandeis, J., dissenting).
• 46274 U.S. 357, 374–77 (1927) (Brandeis, J., concurring). For more on the increasing divergence between Holmes and Brandeis on the exact limits of freedom of speech and thus on the precise understanding of “clear and present danger,” see Polenberg, supra note  NOTEREF _Ref42003731 \h 30 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300033003700330031000000 , at 265–71.
• 47268 U.S. 652 (1926).
• 48Id. at 672 (Holmes, J., dissenting).
• 49See id. at 673. Gitlow’s relaxed, so–called “bad tendency test” was based, in part, on the view that clear and present danger was the appropriate test for evaluating the prosecution of speech under a statute not aimed directly or specifically at speech as such, but that a test of less stringency was appropriate where the legislature, in targeting speech of a certain kind or with a certain effect, had already made a determination about the danger of the speech to which the statute was addressed. See Hans A. Linde, “Clear and Present Danger” Reexamined: Dissonance in the Brandenburg Concerto, 22 Stan. L. Rev. 1163 (1970); Yosal Rogat, Mr. Justice Holmes: Some Modern Views—The Judge as Spectator, 31 U. Chi. L. Rev. 213 (1964).
• 50341 U.S. 494 (1951).
• 51United States v. Dennis, 183 F.2d 201 (2d Cir. 1950).
• 52354 U.S. 298 (1957).
• 53367 U.S. 203 (1961).
• 54367 U.S. 290 (1961).
• 55395 U.S. 444 (1969). On the ins and outs of the Brandenburg test, see Larry Alexander, Inciting, Requesting, Provoking, or Persuading Others to Commit Crimes: The Legacy of Schenck and Abrams in Free Speech Jurisprudence, 72 S.M.U. L. Rev. 389, 392–95 (2019); Gerald Gunther, Learned Hand and the Origins of Modern First Amendment Doctrine: Some Fragments of History, 27 Stan. L. Rev. 719 (1975); Linde, supra note  NOTEREF _Ref42003896 \h  \* MERGEFORMAT 49 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300033003800390036000000 ; Frank R. Strong, Fifty Years of “Clear and Present Danger”: From Schenck to Brandenburg—and Beyond, 1969 Sup. Ct. Rev. 41 (1969).
• 56244 F. 535 (S.D.N.Y. 1917).
• 57See 395 U.S. at 447 (“[The] State [may not] forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.”).
• 58See, e.g., Higgins v. Ky. Sports Radio, LLC, 951 F.3d 728, 736 (6th Cir. 2020); United States v. Daley, 378 F. Supp. 3d 539, 555–56 (W.D. Va. 2019).
• 59See Eugene Volokh, Crime–Facilitating Speech, 57 Stan. L. Rev. 1095 (2005); James v. Meow Media, Inc., 300 F.3d 683 (6th Cir. 2002); Sanders v. Acclaim Entm’t, Inc., 188 F.2d 1264 (D. Colo. 2002).
• 60Not everything that is ex ante likely to happen actually happens, and thus Brandenburg would sometimes permit sanctions against a speaker urging immediate violent actions even if those actions did not in fact occur.
• 61See Wood v. Georgia, 370 U.S. 375 (1962); Pennekamp v. Florida, 328 U.S. 331 (1946); Bridges v. California, 314 U.S. 252 (1941).
• 62See Wood, 370 U.S. at 384–87; Pennekamp, 328 U.S. at 346; Bridges, 314 U.S. at 263; see also Craig v. Harney, 331 U.S. 367, 377 (1947).
• 63379 U.S. 559 (1965).
• 64Id. at 562–65.
• 65310 U.S. 88 (1940).
• 66See id. at 104–05. See also, in the same year, Carlson v. California, 310 U.S. 106, 113 (1940), and, a year later, Milk Wagon Drivers Union v. Meadowmoor Dairies, 312 U.S. 287 (1941). And, slightly earlier, Herndon v. Georgia, 295 U.S. 441, 447–48, 454 (1935) (Cardozo, J., dissenting), followed by Herndon v. Lowry, 301 U.S. 242, 261–64 (1937).
• 67For a thorough exploration of the idea of clear and present danger as a “fall back” approach, see Leslie Kendrick, On “Clear and Present Danger,” 94 Notre Dame L. Rev. 1653, 1655, 1662–63 (2019). For a concern about precisely this state of affairs, see Ronald J. Krotoszynski, Jr., The Clear and Present Dangers of the Clear and Present Danger Test: Schenck and Abrams Revisited, 72 S.M.U. L. Rev. 415 (2019).
• 68See generally Frederick Schauer, Costs and Challenges of the Hostile Audience, 94 Notre Dame L. Rev. 1671 (2019). Contemporary conflicts on college campuses have generated a recent and growing corpus of commentary. See, e.g., Suzanne B. Goldberg, Free Expression on Campus: Mitigating the Costs of Contentious Speakers, 41 Harv. J. L. & Pub. Pol’y 163 (2018); Darrell A. H. Miller, Constitutional Conflict and Sensitive Places, 28 Wm. & Mary Bill Rts. J. 459 (2019); Kevin Francis O’Neill, Disentangling the Law of Public Protest, 45 Loy. L. J. 411 (1999); Christina E. Wells, Free Speech Hypocrisy: Campus Free Speech Conflicts and the Sub–Legal First Amendment, 89 U. Colo. L. Rev. 533 (2018); see also Feiner v. New York, 340 U.S. 315, 320 (1951); Note, Freedom of Speech and Assembly: The Problem of the Hostile Audience, 49 Colum. L. Rev. 1118 (1949).
• 69On the activities of defendants Schenck, Debs, Frohwerk, and Abrams leading to their prosecutions, see Healy, supra note  NOTEREF _Ref51769659 \h 27 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600350031003700360039003600350039000000 ; Polenberg, supra note  NOTEREF _Ref42003731 \h 30 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300033003700330031000000 ; Geoffrey R. Stone, Perilous Times: Free Speech in Wartime: From the Sedition Act of 1798 to the War on Terrorism 135–234 (2004)
• 70See Brandenburg v. Ohio, 395 U.S. 444, 445 (1969); see also Steve Kissing, Brandenburg v. Ohio, Cincinnati Mag., Aug. 2001, at 14–15.
• 71See Volokh, supra note  NOTEREF _Ref42004353 \h  \* MERGEFORMAT 59 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300034003300350033000000 ; James v. Meow Media, Inc., 300 F.3d 683 (6th Cir. 2002); Sanders v. Acclaim Entm’t, Inc., 188 F.2d 1264 (D. Colo. 2002); see also Rice v. Paladin Enter., 128 F.3d 233 (4th Cir. 1997); Herceg v. Hustler Magazine, Inc., 814 F.2d 1017 (5th Cir. 1987); Olivia N. v. Nat’l Broad. Co., 178 Cal. Rptr. 888 (Ct. App. 1981). See generally David A. Anderson, Incitement and Tort Law, 37 Wake Forest L. Rev. 957 (2002); Leslie Kendrick, Note, A Test for Criminally Instructional Speech, 91 Va. L. Rev. 1973 (2005).
• 72340 U.S. 315 (1951).
• 73I put aside the complexities created by speakers who intentionally provoke or attract a hostile audience, and thus who can be said to encourage or desire angry listeners in just this sense.
• 74See Gregory v. Chicago, 394 U.S. 111 (1969); Cox v. Louisiana, 379 U.S. 536 (1965); Edwards v. South Carolina, 372 U.S. 229 (1963). Forsyth Cty. v. Nationalist Movement, 505 U.S. 123 (1992), which held unconstitutional an attempt by the county to require the speakers the bear the financial costs of increased security occasioned by the hostile audience, can be understood as reaffirming the basic thrust of Gregory, Cox, and Edwards, and thus as reaffirming the interment of Feiner.
• 75For more on recent events, of which that in Charlottesville is the most well–known, see the commentaries cited in note  NOTEREF _Ref42005773 \h  \* MERGEFORMAT 68 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000300035003700370033000000 .
• 76310 U.S. 296 (1940).
• 77See id. 311.
• 78337 U.S. 1 (1949).
• 79See id. at 4.
• 80See Glasson v. City of Louisville, 518 F.2d 899 (6th Cir. 1975); Christian Knights of the Ku Klux Klan Invisible Empire, Inc. v. District of Columbia, 751 F. Supp. 218, 220 (D. D.C. 1990).
• 81For an example, see the thorough and complex opinions on both sides of the issue in Bible Believers v. Wayne Cty., 805 F.3d 228 (6th Cir. 2015).
• 82Indeed, Feiner itself is slightly unclear on the issue. Irving Feiner had disobeyed several police requests (and then, seemingly, orders) to stop speaking before he was finally arrested. See Feiner v. New York, 340 U.S. 315, 317–18 (1951). The question remains, and neither Feiner nor any of the subsequent cases answer it, whether the standards for a non–punitive order (the disregard of which might then provide the basis for punishment) are or should be different from the standards applicable to an immediate arrest or citation.
• 83See, e.g., Va. Code Ann. § 18.2–406 (West 2018).
• 84I do not ignore the extent to which—especially these days—speakers, protesters, picketers, paraders, and demonstrators often deliberately provoke the hostile audience, and often do so in the hopes of a violent reaction. But this is not and need not always be so. Sometimes, not surprisingly, speakers prefer not to be assaulted, and sometimes prefer that violence not occur as a result of their activities.
• 85See R. George Wright, The Heckler’s Veto Today, 68 Case West. Res. L. Rev. 159 (2017). The phrase originated in Harry Kalven, Jr., The Negro and the First Amendment 140–65 (1965) and made its first appearance in the United States Reports in Brown v. Louisiana, 383 U.S. 131 (1966).
• 86See Maureen E. Brady, The Damagings Clauses, 104 Va. L. Rev. 341 (2018) (explaining how the law in some states provides compensation even for impairments that do not rise to the level of takings).
• 87The prevailing view now is that property is best understood as a “bundle of rights” and not a physical thing. See Shane Nicholas Glackin, Back to Bundles: Deflating Property Rights, Again, 20 Leg. 1 (2014) (defending the bundle of rights account). But the longstanding lay belief that property is defined by its physical presence has its contemporary academic defenders. See J.E. Penner, The “Bundle of Rights” Picture of Property, 43 UCLA L. Rev. 711 (1996).
• 88This explanation—the monetization explanation—seems odd, however. If the state wrongly deprives someone of her free speech rights, she can bring a civil rights action to seek monetary compensation for what she has lost. See generally John C. Jeffries, Jr., et al., Civil Rights Actions: Enforcing the Constitution (4th ed., 2018). And if this is possible, then it is difficult to see why there could not similarly be a monetary value attached to a rightful restriction.
• 89124 N.W. 221 (Minn. 1910) (suggesting, even if not directly holding, that a shipowner who justifiably saved his ship in a storm at the cost of damage to someone else’s dock would owe compensation to the dock-owner). Somewhat similar is Ploof v. Putnam, 71 A. 188 (Vt. 1908), concluding that the shipowner in an analogous situation was not liable in trespass. For commentary on these cases and the issues they raise, see George C. Christie, The Defense of Necessity Considered from the Legal and Moral Points of View, 48 Duke L. J. 975 (1999); John C.P. Goldberg & Benjamin C. Zipursky, The Strict Liability in Fault and the Fault in Strict Liability, 85 Fordham L. Rev. 743, 765 n.89 (2016).
• 90See Judith Jarvis Thomson, The Realm of Rights 84–86, 93–96 (1990); Judith Jarvis Thomson, Rights, Restitution, and Risk: Essays in Moral Theory 59–60, 71–72, 76–77 (1986).
• 91F.M. Kamm, Intricate Ethics: Rights, Responsibilities, and Permissible Harm 249–60 (2007).
• 92Walter Sinnott–Armstrong, Moral Dilemmas 44–53 (1988).
• 93This is the term used by Thomson, The Realm of Rights, supra note  NOTEREF _Ref42064205 \h 90 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000360034003200300035000000 , and Sinnott–Armstrong, supra note  NOTEREF _Ref42064214 \h 92 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000360034003200310034000000 . Kamm, supra note  NOTEREF _Ref42064221 \h 91 08D0C9EA79F9BACE118C8200AA004BA90B02000000080000000D0000005F00520065006600340032003000360034003200320031000000 , calls it “negative residue.”
• 94See also Rex Martin & James W. Nickel, Recent Work on the Concept of Rights, 17 Am. Phil. Q. 165 (1980).
• 95This was not the exact language he used. See Brandenburg v. Ohio, 395 U.S. 444, 446 (1969).
• 96From Cox v. Louisiana, 379 U.S. 536 (1965).