I. Introduction

Few important areas of the law exhibit the unpredictability of free speech overbreadth cases. In a dramatic understatement, the Supreme Court has recognized that “[t]he concept of ‘substantial overbreadth’ is not readily reduced to an exact definition.”[1] In reality, overbreadth claims are often resolved in ways that seem readily contestable, if not arbitrary. This Article illustrates this systematic unpredictability and presents several suggestions for more consistently defensible results in overbreadth cases.

The basic logic of the free speech overbreadth doctrine has been expressed with at least some initial clarity. The Supreme Court has declared:

According to our First Amendment overbreadth doctrine, a statute is facially invalid if it prohibits a substantial amount of protected speech. . . . On the one hand, the threat of enforcement of an overbroad law deters people from engaging in constitutionally protected speech, inhibiting the free exchange of ideas. On the other hand, invalidating a law that in some of its applications is perfectly constitutional—particularly a law directed at conduct so antisocial that it has been made criminal—has obvious harmful effects.[2]

This formulation implicitly raises a distinction between speech and conduct and recognizes important value conflicts in the overbreadth cases and the need for interest balancing. The Court has thus declared, “In order to maintain an appropriate balance, we have vigorously enforced the requirement that a statute’s overbreadth be substantial, not only in an absolute sense, but also relative to the statute’s plainly legitimate sweep.”[3]

However this is to be fleshed out, the idea seems roughly to be that some limited degree of overbreadth of the regulation will not amount to unconstitutional overbreadth.[4] Unconstitutional overbreadth requires some substantial scope or severity of burdening constitutionally protected speech. But crucially, there must also be some disproportionality between that burdening and the constitutionally unprotected speech or conduct that is properly subjected to the regulation at issue.[5] That is, the adverse effect on constitutionally legitimate speech must be sufficiently large relative to the value of the regulation’s justified applications. The Court’s first task, though, is to determine the actual scope, or coverage, of the regulation at issue.[6]

Despite this and other authoritative formulations, however, the overbreadth cases generally manifest unpredictability, contestability, and apparent judicial arbitrariness in their resolutions. Merely by way of an initial illustration, consider the recent communications privacy case of State v. Lamoureux.[7]

The Lamoureux case involved a number of phone calls made by the defendant to more than one person with the apparent intention to “terrify, intimidate, threaten, harass, or offend”[8] a specific person, if not the recipient of the phone call.[9] The defendant argued that such a restriction amounts to a disfavored, content-based regulation of speech, focusing on the message or idea expressed by the speaker.[10]

The Montana Supreme Court’s response to the defendant’s argument oddly concluded that communications with the statutorily specified intention or purpose should be treated not as speech in the sense of conveying an idea, but instead, as mere conduct.[11] The conduct at issue, however, was then also curiously referred to as “speech . . . uttered with the purpose and specific intent of intimidating, threatening, or harassing another person.”[12] Thus, while the telephoned statements were “quite expressive,” they were “void of any social value whatsoever” and played “little to ‘no essential part of any exposition of ideas.’”[13] Such speech or mere conduct was not “communication of information or opinion safeguarded by the Constitution.”[14]

As to the overbreadth claim, the court held that the statute in question, regardless of the status of the defendant’s communication, was “narrowly tailored to control conduct without reaching a substantial amount of protected speech.”[15] The court assumed that the statutory requirement of a specific intent on the part of the speaker to “terrify, intimidate, threaten, harass, annoy, or offend” undermined any credible overbreadth claim.[16]

One might wonder, however, whether a “substantial” amount of speech deserving of constitutional protection could have—as one of several specific purposes or intentions—the intent to annoy, offend, or even intimidate in some respect. Presumably, much legitimate, partisan, political debate in an era of political polarization and fragmentation could, in part, have this character. A statutory specific intent requirement may thus not by itself be properly decisive against an overbreadth claim.

The court in Lamoureux concluded, however, that “[t]he statute does not suppress or infringe upon . . . any person’s[ ] freedom to engage in the uninhibited, robust, and wide-open expression of ideas or a suitable level of discourse within the body politic.”[17] Whether this result reflects the best application of the free speech overbreadth doctrine is, however, broadly controversial.

Consider, in the context of Lamoureux itself, the arguments of the opposing parties with respect to petitioner Lamoureux’s contentions in his petition—ultimately denied—for a writ of certiorari.[18] Lamoureux’s petition cited to a broad and persisting split in the relevant case law. In particular, Lamoureux argued that “courts have split over whether laws prohibiting speech intended to annoy or offend are facially overbroad.”[19] By Lamoureux’s own count, eight high courts at the state level and one federal appellate court have held such statutes to be facially overbroad, while three federal appellate courts and six state high courts have held the opposite.[20] The latter courts either find that the relevant statute regulates conduct, as opposed to speech, or that the statute’s impact on protected speech is somehow insufficient to meet the standards of the overbreadth doctrine.[21]

In contrast, though, the State of Montana argued that this apparent nine-versus-nine case split is, in reality, largely a matter of the level of generality or specificity at which one characterizes the universe of relevant statutes—with materially different statutes generating different analyses and outcomes.[22] As we shall see, though, the analyses adopted by courts that both find and decline to find unconstitutional overbreadth are often deeply questionable.[23] Let us then briefly consider the judicially recognized tests for the presence or absence of overbreadth.

II. The Tests for Overbreadth and Some Initial Problems

Let us simply assume, perhaps controversially, that some version of the overbreadth doctrine is legitimate.[24] We have already referred briefly to the underlying logic of overbreadth law.[25] The overbreadth doctrine is classically presented in the case of Broadrick v. Oklahoma.[26] Broadrick notes that the overbreadth doctrine,—confined to free speech law—loosens the familiar requirements of judicial standing.[27] The overbreadth doctrine permits “attacks on overly broad statutes with no requirement that . . . [the defendant’s] own conduct could not be regulated by a statute drawn with the requisite narrow specificity.”[28]

As a result, Broadrick observes, “[l]itigants . . . are permitted to challenge a statute not because their own rights of free expression are violated,”[29] but for the sake of a healthy culture of freedom of speech. The theory is one of expediting a judicial challenge to a statute that may unduly chill, or deter, speech by nonparties that is fully protected by the Constitution.[30] Broadrick thus concludes that “the possible harm to society in permitting some unprotected speech to go unpunished” may, in a given case, be “outweighed by the possibility that protected speech of others may be muted and perceived grievances left to fester.”[31]

The overbreadth doctrine thus requires some sort of hazy, speculative, and largely normative balancing, with rights and interests of various sorts on both sides of the balance.[32] The proper balance, according to Broadrick, particularly where expressive conduct—as distinct from pure or literal speech—is involved, requires that “the overbreadth of a statute must not only be real, but substantial as well, judged in relation to the statute’s plainly legitimate sweep.”[33]

Inevitably, any judicial test that focuses on the vague idea of ‘substantiality’ can provide only limited concrete guidance.[34] Any distinction between speech and symbolic conduct must somehow address the problem that some conduct may be much more distinctly expressive of an idea than other instances of conduct.[35] Are concepts such as the ‘muting’[36] of speech anything other than loose metaphors? Are all ‘muted’ grievances of equal weight in the constitutional balance? At what level of generality are the muted or ‘chilled’ grievances to be described?[37] Why is the judicial focus on the ‘plainly’ legitimate sweep[38] of the statute, as distinct from its legitimate sweep, whether plain or not? If there are genuine constitutional rights or strong interests on both sides of the balance; how is the balancing then to be carried out?

And then how are the courts to determine the relevant sweep, scope, or breadth of a statute in the first place? Do hypothetical cases ever count? Many hypothetical cases seem realistic. Do apparently realistic cases count more? Does each real case have the same constitutional weight as every other real case? Or are some real cases constitutionally weightier than others? Are some free speech interests or values weightier than others? To what degree, if any, should the judicial balancing consider whether the challenger could have readily expressed their idea in some way that clearly does not fall within the scope of the statutory prohibition? And to what extent could the legislature have promoted its statutory goals without impinging as much, or at all, on free speech rights and interests?

The courts often assert that the comparison and the balancing at stake will somehow be a matter of quantities or of numbers. Thus, the Supreme Court, in the animal-cruelty video case of United States v. Stevens, declared that “a law may be invalidated as overbroad if ‘a substantial number of its applications are unconstitutional, judged in relation to [its] plainly legitimate sweep.’”[39] Where these comparative numbers are to come from, at whatever level of granularity of description, is left unaddressed. Instead, the judicial accounting of cases and of applications is often largely a matter of imagery, metaphor, and what we might call emotive or intuitive judicial reaction.

Thus, the overbreadth cases drift from largely quantitative or numerical tests[40] to vaguer tests, such as whether the statute is “susceptible of regular application to protected expression.”[41] But, the application of a statute might be regular though infrequent, such as the case of the regular, but rare, return of a comet. And such a test focuses not even on the regularity of application but instead on whether the statute is merely “susceptible”[42] to such application.

The drift toward less quantitative test formulas then takes the form of asking whether “there is ‘a realistic danger that the statute itself will significantly compromise recognized First Amendment protections of parties not before the Court.’”[43] Here, the problem is that danger may be quite realistic without also being especially likely to occur. Most of us would, quite realistically, not stay in a building with a five or ten-percent chance of a roof collapse. And the vague notion of “compromising”[44] free speech protections may lack the precision of even a focus on the violation of a First Amendment right.

From there, the judicial drift continues to foreground imagery, metaphor, and emotive reaction, whether as a reaction to a supposedly quantitative test or as a partial substitute for such a test. Thus, the crucial statutory section in the Stevens animal-cruelty video case was characterized, whether subjectively or not, as “a criminal prohibition of alarming breadth.”[45] And more broadly, it has been said that “[a] statute is likely to be found overbroad if the criminal prohibition it creates is of ‘alarming breadth.’”[46] Similarly emotive, if not physiological in their nature, are characterizations of the degrees of overbreadth in given cases as “breathtaking.”[47]

More concretely, the courts are unclear on the role of particular examples in which constitutionally protected speech is swept up by the regulation, whether in contrast with examples of permissibly regulated speech or not. The parties would presumably wish to provide “numerous” examples of whatever weight, breadth, or significance to enhance their argument.[48] One might imagine that since the overbreadth doctrine focuses on the rights of unspecified parties not before the court,[49] the courts would welcome discussion of both actual and hypothetical examples of persons whose speech might well be deterred by the statute in question. After all, a silenced speaker may not be visible to the court.

But the cases are, in this respect, conflicting. The Fourth Circuit has observed that “the fear of chilling protected expression ‘has led courts to entertain facial challenges based merely on hypothetical applications of the law to nonparties.’”[50] Other courts may be willing to consider admittedly hypothetical cases of statutorily chilled speech if ultimately judging the hypothetical cases to be “weak” or “far-fetched,” and thus, insufficient to raise a meaningful overbreadth issue.[51] More restrictively, though, the First Circuit has declared that “in the absence of veridical examples we are not inclined to rely on hypotheticals.”[52] The problem with such restrictive approaches is that the required “substantial number”[53] of deterred speakers—or of deterred instances of speech—almost always requires consideration of cases that are unripe, and thus, hypothetical.

A final problem plaguing the overbreadth cases is the courts’ willingness to manipulate the distinction between speech and conduct—particularly for the sake of upholding the regulation in question. The courts typically recognize that some instances of what is classified as ‘conduct’ may deserve some degree of protection under the First Amendment.[54] This possibility depends, crucially, on whether the conduct is “sufficiently imbued with elements of communication.”[55]

But, it is then said that overbreadth doctrine concerns diminish as regulated activity moves from pure speech into the realm of conduct.[56] And it is hardly an abuse of language to refer to any act of social media communication as involving—if not entirely consisting of—conduct.[57] Repeated instances of social media communications, regardless of content, then become a “course of conduct.”[58] And commonly, the courts then characterize the charged acts as a ‘course of conduct,’ rather than repeated or continuing instances of speech of whatever sort.[59]

The problem here is that any such conclusion allows the court remarkably unconstrained discretion in taking seriously, or in dismissing, any overbreadth claim. As one court has observed, “Saying that restrictions on writing and speaking are merely incidental to speech is like saying that limitations on walking and running are merely incidental to ambulation.”[60] Thus, particularly in the overbreadth context, the “enterprise of labeling certain verbal or written communications ‘speech’ and others ‘conduct’ is unprincipled and susceptible to manipulation.”[61]

As it turns out, though, the adverse effects of these doctrinal problems can be limited by shifting the judicial focus to questions that ultimately matter the most. We can say, in a very brief and conclusory way, that what matters most in the overbreadth area is the constitutional value of the kinds of speech at stake; the availability of alternative unregulated ways of conveying one’s message; and the availability of regulatory regimes with either a diminished impact on protected speech or at least with a better coverage ratio of unprotected to protected speech. We take up these considerations below.

III. Overbreadth and a Judicial Sensitivity to Greater and Lesser Valued Categories of Speech

Close to the essence of free speech law is the strong aversion to regulating speech on the basis of anyone’s disagreement with some meaningful point the speaker seeks to make.[62] Much literal speech, though, does not rise to the level of making some interesting claim or asserting some idea sufficient to provoke any government authority to disagree on the merits.

Thus, it has been suggested that “[t]he purpose of the free-speech clause . . . is to protect the market in ideas . . . broadly understood as the public expression of ideas, narratives, concepts, imagery, opinions—scientific, political, or aesthetic—to an audience whom the speaker seeks to inform, edify, or entertain.”[63]

To the extent that any interpersonal communication evidently falls outside the scope of the purposes of freedom of speech,[64] such communication does not implicate the logic of constitutionally protecting speech.[65]

There is no doubt that speech, in whatever medium, that is threatening, vengeful, intimidating, alarming, abusive, terrifying, frightening, bullying, harassing, intentionally offensive, or merely annoying—particularly if repeated—is typically important to the speaker and, in a different way, to the recipient or target of such speech.[66] But mere private or personal importance in this sense need not translate into relevance to any of the broad purposes underlying the constitutional protection of speech.

While the typologies of the basic purposes underlying freedom of speech vary to some degree, the mainstream purposes are widely recognized.[67] Central among these recognized purposes is the desire to promote and facilitate the search for truth.[68] As well, freedom of speech is often considered essential to a well-functioning and meaningful, representative, democratic government.[69] And then there is the idea that freedom of speech may contribute to individual or collective self-realization, autonomous self-fulfillment, general development, and genuine flourishing.[70]

Now, it is certainly true in some instances that speech that is intentionally annoying, offensive, or otherwise provocative can rightly fall within the scope of protected speech. Classically, the Court recognized in Terminiello v. City of Chicago that it is:

a function of free speech under our system of government . . . to invite dispute. It may indeed best serve its high purpose when it induces a condition of unrest, creates dissatisfaction with conditions as they are, or even stirs people to anger. Speech is often provocative and challenging. It may strike at prejudices and preconceptions and have profound unsettling effects as it presses for acceptance of an idea.[71]

The problem, though, is that many cases involve threatening, vengeful, intimidating, alarming, abusive, terrifying, frightening, bullying, harassing, intentionally offensive, or merely annoying speech without also implicating the broad political concerns emphasized in Terminiello.[72] And more decisively, those many cases do not significantly implicate any of the basic values or purposes widely thought to justify constitutional protection for speech.[73] At most, it can be said that in some such cases, the free speech value of self-fulfillment and flourishing at stake is at least as much that of the victim or target of the speech as of the speaker.[74]

Thus, courts should decide overbreadth cases in ways that are not merely compatible with, but promote on balance, the consensually basic purposes of protecting freedom of speech. This does not mean that courts should simply hold the facts of the overbreadth case up, directly and immediately, against those basic purposes, looking for matches or conflicts. There can instead be a role for middle-range judicial tests that less directly link the basic purposes of freedom of speech with the circumstances of a speaker, and those of other potential speakers who may be affected.

A useful middle-range test asks whether the speech at issue implicates a matter of public interest and concern, or else a matter of merely private interest and concern.[75] Whether speech addresses a matter of public interest and concern can be clearly linked to the basic reasons for protecting freedom of speech in general.[76] The further necessary step is to then apply the public versus private concern test to the distinctive overbreadth context, where the relevant speech is not crucially that of the immediate speaker, but of other potentially affected speakers. That is, whether the potentially chilled speech by other persons would reflect merely private concerns, as distinct from any matter of public interest.

One complication is that the courts have not reached a consensus on the degree of free speech protection, if any, to be accorded to speech addressing matters of only personal or private interest. In non-overbreadth contexts, the Supreme Court has sometimes suggested that such speech should receive no distinctive free speech protection at all.[77] But the Court has also referred to merely reduced, or only rarely available, free speech protection for such speech.[78]

Fortunately, though, it is unnecessary for our purposes to decide whether the merely privately focused speech of persons who might be swept up by an overbroad statute should receive only minimal protection----or none at all. The point of the overbreadth doctrine, after all, is not to protect instances of speech of any nature, or perhaps, only minimal value.[79] And, it is clear from the case law that even constitutionally valuable commercial speech generally falls outside the scope of the overbreadth doctrine.[80]

On this basis, let us start with an underlying case in which overbreadth is alleged. And, let us assume that the speech in question amounts to an instance, or repeated instances, of some threatening, vengeful, intimidating, alarming, abusive, terrifying, frightening, bullying, harassing, intentionally offensive, or merely annoying speech. Now, given this characterization of such speech, the targeted speech itself will often be characterizable as speech that addresses matters of merely private interest.[81]

But what can we say about the most relevant cases in which the potential speakers might reasonably fear an unjust prosecution of their genuinely protected speech? Consider the position of a potential speaker who wishes to address some matter of public interest and concern. How would such a person react to typical cases seeking to invoke the overbreadth doctrine? In particular, would potential speakers reasonably feel chilled, intimidated, or anxious about raising matters of public interest and concern, given the denial of the overbreadth claim in typically litigated cases?

More concretely, how would such speakers react to a denial of an overbreadth claim in the case referred to above: State v. Lamoureux?[82] As the Montana Supreme Court reports the underlying events, it seems extremely unlikely that any speaker seeking to address a matter of public interest or concern would sense an impending shadow of repression being cast by the result in Lamoureux.

As reported, the defendant Lamoureux bore an intense and sustained hostility to his ex-wife that—to some degree—extended to one of her employees and to her father.[83] On one of several cited occasions, the defendant, while apparently “aggressive, angry, [and] drunk,” called the employee in question and allegedly demanded the phone numbers of one of his children and of his ex-wife’s father.[84] The employee responded that she was unable to provide the phone numbers, and the defendant allegedly reacted with profanity, threats, and repeated phone calls.[85] The employee then closed the store, alerted a neighboring store owner, and called the police.[86]

On another occasion, the defendant allegedly called his ex-wife’s father and announced that he was on his way to kill his ex-wife and that he intended to burn her jewelry-store building.[87] The defendant’s language was, on this occasion, again allegedly “profane, offensive, threatening, and harassing,” as well as credible.[88] The defendant was charged with violating a Montana state communications privacy statute, to which the defendant raised—among other defenses—an overbreadth claim.[89]

There seems, crucially, to be nothing in the speech or conduct of the defendant in Lamoureux that would suggest any communicative intent above and beyond personal or private grievances, however intensely felt.[90] Consider, in contrast, the position of virtually any other speaker whose primary intention is to convey, perhaps in emotionally fervent terms, some message directed at a matter of public interest and concern. Perhaps the matter of concern is climate change, racial equity, healthcare system design, or economic class opportunity. Would anyone with such primary intention reasonably fear repression of their speech based on the denial of an overbreadth claim in anything remotely like the Lamoureux case?[91] Hardly.

And, what is true of the Lamoureux case is true across a broad, but not entirely exhaustive, range of overbreadth cases. Consider, as a further typical example, the recent overbreadth case of United States v. Fredrickson.[92] In this case, the defendant was charged with violating a federal statute that prohibits “inducing sexually explicit videos from a minor.”[93] The defendant, a twenty-seven-year-old male, had developed a pattern of social media communications with a sixteen-year-old, out-of-state minor.[94] Allegedly at the defendant’s request, their conversations took a sexually explicit turn, with the minor female allegedly sending explicit videos.[95] Ultimately, according to the Seventh Circuit, the defendant’s cell phone was subjected to a law enforcement search, allegedly revealing that he had been recording and saving a number of images sent by the minor via Snapchat, amounting to “at least fifteen sexually explicit videos” of the minor in question.[96]

The statute in Fredrickson referred to “[a]ny person who . . . persuades, induces, entices, or coerces any minor to engage in . . . any sexually explicit conduct for the purpose of producing any visual depiction of such conduct . . . .”[97] Here again, we need not consider the arguments of the defendant and the prosecution regarding the charged conduct or the underlying circumstances. We instead merely ask whether even a risk-averse person seeking to address a public issue would reasonably feel a chill of intimidation caused by the denial of this defendant’s claim of overbreadth. Crucially, virtually any idea relating to matters of public interest and concern can be conveyed without remotely resembling any of the speech or conduct prosecuted in Fredrickson.

These examples represent a larger class of, at least loosely, similar cases in which the denial of an overbreadth claim would have essentially no adverse effect on the discourse surrounding public issues and would not risk impairing any of the values and purposes underlying the constitutional protection of speech.[98] Certainly, many such cases, do not involve images that would be classified as obscene or as child pornography.[99] And it seems clear that communication in the form of “sexting,” in general, is a common phenomenon.[100] However much this might otherwise bear on the case law, the point remains that typically, no one who might wish to address a matter of public interest or concern need feel constrained by a denial of an overbreadth claim in any of the above sorts of cases.

There is a difference between foreseeable—and even intended—annoyance or embarrassment as a result of vigorous contribution to a political debate, and annoyance or embarrassment—foreseen or intended—in the context of personal or private vengeance, hostility, score-settling, or any other personal form of gratification.[101]

The other side of the coin is that overbreadth claims should be taken seriously in cases in which the underlying speech—and most importantly, cases that implicate the speech of a number of other potential speakers—does, in fact, address matters of public interest or concern.[102] Some overbreadth cases crucially involve speech that is clearly political,[103] or arguably political in part.[104] And inevitably, there will be at least a few cases where any implication of speech on matters of public interest may be unclear.[105]

It is useful to remember, though, that even the presence of a clear and exclusive intent to convey a plainly political point should not decide an overbreadth claim. A true threat, for example, may be clearly political without triggering a serious claim for statutory overbreadth.[106] Therefore, the nature of any potentially chilled speech that addresses a matter of public interest or implicates the reasons speech is protected in the first place, remains crucial. But further progress on reforming overbreadth is also possible, as discussed immediately below.

IV. Overbreadth and the Alternatives Available to Other Speakers and to Government Regulators

The assumption underlying the overbreadth doctrine is that in the absence thereof, “[m]any persons, rather than undertake the considerable burden (and sometimes risk) of vindicating their rights through case-by-case litigation, will choose simply to abstain from protected speech . . . .”[107] On this assumption, the chilling of protected speech harms not only those intimidated speakers, “but society as a whole, which is deprived of an uninhibited marketplace of ideas.”[108] The overbreadth doctrine seeks to reduce this social cost.[109]

But the model of speech practices that this logic relies upon is grossly and unnecessarily crude and distortive. The crucial error is in assuming that potential speakers will, in response to the denial of an overbreadth claim, either censor themselves in response to a perceived chill (perhaps thereby leaving an otherwise unarticulated idea off the public agenda) or else speak their piece, risking costly litigation, if not a prison term. But it would be quite unreasonable for potentially affected speakers to embrace this crude binary.

The more reasonable course for potentially affected speakers would be to devote a moment’s reflection to possible, and indeed more optimal, ways of conveying the message they wish to communicate. An idea is not typically paired inseparably with a single way in which that idea can be meaningfully conveyed,[110] tone or emotion included.[111]

Thus, for anyone who might be affected by the denial of some other speaker’s overbreadth claim, the crucial question is one of realistic availability—at a comparable cost—of alternative speech channels, media, forms, or formats. It must be considered from the standpoint of each potential speaker’s priorities among the basic free speech values.[112] Speakers may differ as to their preferred target audience; the length and complexity of their message; the timeliness or permanence of the message’s availability; or the cost of sending, maintaining, or receiving the message—among other considerations.[113]

Thus, some potentially chilled speakers might prefer to disseminate their ideas by posting, at low cost, on publicly owned utility poles.[114] Such a preference may not be intense or well-considered; but suppose the denial of some other speaker’s overbreadth claim casts a shadow over the now-chilled potential speaker’s use of that channel. In that case, any reasonable potential speakers could assess whether there are alternative means of media to convey their messages—in an essentially undistorted and uncompromised form—that would not be vulnerable to any likelihood of prosecution.[115]

In a broad range of cases involving attempts to convey messages of public interest, there will ordinarily still be channels or formats available that do not fall under the shadow of the denial of any underlying overbreadth claim.[116]

Suppose, for example, a speaker wishes to communicate a contribution to a public issue. Bearing in mind the overbreadth claim denials in prior cases, though, that speaker wishes to reduce the risk of criminal prosecution. What the cases should deter, though, is typically speech involving credible threats of personal violence; the nonconsensual disclosure of private information or images; repeated personal stalking; revenge porn; the infliction of suffering merely for its own sake; or personal exploitation of a minor by an unrelated adult.[117]

Of course, potential speakers who are uninterested in communicating any message on a matter of public interest and concern may wish, primarily, to harass or frighten some targeted individual on personal grounds. In such a case, there may be no available way to carry out that intention without meaningfully risking prosecution. But the free speech interest of that speaker—if not that of the victim or target—is minimal to nonexistent.[118]

Consider, as one further example, the case of United States v. Waggy.[119] Waggy involved an “as-applied” challenge—distinct from a challenge on broader overbreadth.[120] But the case considers the question of remaining available speech channels with sufficient clarity for our purposes. The defendant in Waggy argued that the state telephone harassment statute was unconstitutional, as applied to him, “because he just ‘wanted to talk about his [Veterans Administration] medical care and the VA’s unpaid bills’ . . . .”[121]

Assuming this to be so, and further assuming this subject to be a matter of public interest and concern, consider the position of parties who might be affected had the defendant raised—and the court rejected—an overbreadth challenge. It was alleged in Waggy that the defendant repeatedly called the Spokane VA Medical Center seeking to speak with the director, but instead spoke several times with an executive secretary.[122] Allegedly, the defendant “demanded that the VA pay him $9.25 million or ‘get off [his] property.’”[123] It was further alleged that the defendant threatened to personally “seize the [VA] property and to ‘use force to defend himself.’”[124] Allegedly, the defendant called the same office back several times and voiced vulgar and insulting obscenities at a vocal level amounting to screaming, to a frightening effect.[125]

Applied in the overbreadth context, the speech “chill” at issue would prompt potential speakers to consider whether—in accordance with their own priorities among the basic free speech values—they could embrace some constitutionally adequate alternative means of communicating their public interest message.[126] More concretely, this would involve an assessment of whether their message could be conveyed without the alleged recourse to threats, intimidation, personally frightening speech, and personally directed, vulgar insults. This should be possible even for the emotionally fervent potential public interest speaker. And if not, the loss in free speech values for the directly affected persons will typically be minimal, if not negative.

Though there is another side of the alternative-options coin. Just as speakers may have realistic alternatives to avoid potential prosecution, legislatures may have alternatives when enacting statutory language. After all, the nature of whatever problem any statute seeks to address may be vague or indeterminate.[127] In practice, some legislative processes may tilt toward expanding the reach of a statute to sustain a broad coalition of support or to accommodate the best-organized factions.[128] And some legislatures may err on the side of legislative inclusiveness or the avoidance of “loopholes.”[129]

The classic technique by which courts sometimes address overexpansive legislative choices is that of judicial severance of one or more elements of a statute from the rest of the statute.[130] Where discrete words or phrases within a statute render that statute constitutionally suspect, courts may ask whether without such language the remaining statute is coherent and would have attracted legislative support.[131] The courts, in such a case, ask whether the legislature would have adopted, as an alternative, a statute without the constitutionally offending words or phrases.[132]

Thus, in response to an overbreadth challenge, the Ninth Circuit has recently applied the severability doctrine to excise particular statutory words from the federal Anti-Riot Act.[133] The court determined that while prohibitions against rioting or inciting to riot were permissible, the statutory prohibitions of generally urging, encouraging, or advocating a riot were not.[134] This judicial response essentially implies that the legislature had an adequate and constitutionally permissible legislative alternative available to it but declined to adopt it.

Beyond the question of severability, there is the much broader realm of legislative space within which any legislature might have navigated. Whether well-defined or only vaguely defined, any object of legislative attention can be addressed in a number of ways.[135] For example, even COVID-19 mandates could be supplemented by legislated taxes, insurance penalties, subsidies, and grants—all raising issues of various degrees of constitutional severity.[136] Just as speakers may steer clear of potential free speech controversies, so may legislative drafters by considering a range of alternative regulations.[137] Though it is hardly clear that any initially preferred statutory language is likely to be the most effective way of promoting one or more statutory goals in actual practice.[138]

Overbreadth cases can be minimized by charging defendants with the most evidently constitutionally valid statutory offenses, whether at the state or federal level. In a given case, chargeable statutory offenses, with varying degrees of vulnerability to an overbreadth challenge, may overlap to one degree or another.

Consider, for example, an apparently difficult overbreadth case in which the defendant was charged under the federal Anti-Riot Act. In United States v. Massey, the court rejected the defendant’s plausible claims that the Anti-Riot Act was facially overbroad.[139] The court engaged in an extensive discussion of the conflicting federal court of appeals law on the alleged overbreadth of the statute in question.[140] The defendant’s speech was directly focused on the impending civil disorders in the Chicago Loop area on August 9–10, 2020.[141]

While the court’s overbreadth discussion in Massey was thoughtful, it was not necessary. The case opinion itself states that after the defendant posted the charged Facebook messages in question, he “broke into and stole merchandise from four different stores in the Loop and across the near north side of Chicago.”[142]

In this case, state and federal prosecutors had a range of entirely viable alternative regulatory approaches available to them. One obvious alternative to this federal statutory speech prosecution that invited a complex overbreadth challenge, would have been to prosecute for state law violations of burglary, destruction of property, theft, or disorderly conduct. Presumably, such prosecutions would not have been beyond the resources and capacities of the state criminal-justice system; and therefore, no overbreadth issues would have arisen.

In this case, and many others, jurisdictions have a range of chargeable offenses available that are equal in gravity and in likely cost of prosecution. Some offenses will involve communicative elements and possible overbreadth issues, but others will not. To a significant degree, the complications of the overbreadth doctrine can thus be limited by the reasonable exercise of prosecutorial choices.

More narrowly, federal and state actors can reduce the significance of overbreadth issues by reflecting on questions of intent and mens rea. For example, instances of speech may be, to many, offensive or annoying.[143] But at least on matters of public interest and concern, such speech is typically protected and the overbreadth doctrine then becomes relevant.[144] But if the statutory interests and purposes are compatible with a narrow and demanding intent requirement, overbreadth issues can be minimized.

In State v. Richards, for example, the court considered the possible overbreadth of a state telephone harassment statute.[145] But, at least as construed by the court, the statute in question required the speaker to have had a sole and exclusive speech intent to annoy or offend, among other possible effects on a listener.[146] Such a construction clearly reduces, if not eliminates, any significant overbreadth issues. Potential speakers on a public issue have an intent that at the very least cannot be proven by the prosecution, beyond a reasonable doubt, to be exclusively to annoy or offend some audience. By definition, a speaker who wishes to address a matter of public interest and concern will also have the intent to convey some substantive idea or message, however vague.[147] In such cases, the government bears the burden of proving the impossible: that a speaker who intends to communicate a public interest message instead intends, solely and exclusively, to annoy or offend some listener. Any conceivable overbreadth issue is minimized by the legislative choice to narrow the required intent.

A final kind of legislative choice among alternatives should be noted. As it stands, the overbreadth doctrine creates a perverse incentive for legislative drafters. Recall that “a statute’s overbreadth [must] be substantial, not only in an absolute sense, but also relative to the statute’s plainly legitimate sweep.”[148] This proportionality requirement, however vague and metaphorical, invites legislative manipulation.

Clearly, one way to reduce the chances of unconstitutional overbreadth is for legislators to gratuitously add to the legitimate sweep—plain or not—of the statute. Any additional non-speech and plainly unprotected behavior that can be brought within the scope of the regulation obviously increases the ratio of unproblematic applications to the problematic applications of the statute. By the Court’s own test, a statute becomes less vulnerable to an overbreadth challenge as the statutory scope expands to include more unprotected behaviors.[149] There is, however, no legitimate public interest in incentivizing such gratuitous, and perhaps perverse, legislative behavior. The constitutional case law should not encourage any such legislative behavior.

V. Conclusion

At this point, the case law on the overbreadth doctrine has descended, in large measure, into dubious logic and arbitrary results. But this state of affairs is also largely remediable. First, there is no need for the courts to engage in the practice of manipulating the admittedly imprecise distinction between speech and conduct. Nor need courts rely on constitutional tests that incorporate psychological notions such as ‘alarming’ or ‘breathtaking’ degrees of overbreadth. Courts should instead pay more attention to the practically manageable distinction between speech that addresses some matter of public interest and speech that implicates matters of only personal or private interest. Both courts and potential speakers should also better consider the presence of reasonable and unregulated alternative ways in which to convey messages of public interest. Courts and legislatures should also attend to the responsible use of narrow and demanding specific intent requirements in statutes that regulate speech. And finally, the courts should not incentivize the legislative drafting of statutes that unnecessarily restrict realms of clearly prohibitable conduct just for the sake of increasing the ratio of prohibitable conduct to protected speech within the scope of the statute.

  1. Members of City Council of L.A. v. Taxpayers for Vincent, 466 U.S. 789, 800 (1984). This language was picked up in Acosta v. City of Costa Mesa, 718 F.3d 800, 811 (9th Cir. 2013). For a recent confirmation, see Dream Defs. v. DeSantis, 559 F. Supp. 3d 1238, 1282 (N.D. Fla. 2021). One minor terminological complication is that the courts are not consistent as to whether the constitutional problem is one of the regulation’s overbreadth, of its ‘substantial’ overbreadth, or else of its ‘excessive’ or otherwise unconstitutional overbreadth. See infra Part II.

  2. United States v. Williams, 553 U.S. 285, 292 (2008).

  3. Id.

  4. See id.

  5. See id.

  6. Id. at 293.

  7. State v. Lamoureux, 485 P.3d 192 (Mont. 2021).

  8. Id. at 196, 200.

  9. Id. at 200–01.

  10. Id. For controversial background, see the Court’s discussion of this general concept in Reed v. Town of Gilbert, 576 U.S. 155 (2015).

  11. Lamoureux, 485 P.3d at 200. For a similarly odd and unnecessary distinction between speech and conduct in an as-applied context, see United States v. Waggy, 936 F.3d 1014, 1019–20 (9th Cir. 2019) (citing United States v. Sayer, 748 F.3d 425, 435 (1st Cir. 2014)). For a terse response, see Ludwig Wittgenstein, Philosophical Investigations § 546, at 155 (P.M.S Hacker & Joachim Schulte eds., G.E.M Anscombe et al. trans., Wiley-Blackwell rev. 4th ed. 2009) (“Words are also deeds.”).

  12. Lamoureux, 485 P.3d at 200.

  13. Id. (quoting the historic, if today not much influential, logic of Chaplinsky v. New Hampshire, 315 U.S. 568, 572 (1942)).

  14. Id. (quoting Chaplinsky, 315 U.S. at 572).

  15. Id.

  16. Id. at 198–99 (citing State v. Dugan, 303 P.3d 755 (Mont. 2013)). Crucially, an overbreadth claimant, in effect, concedes the legitimacy of their conviction under a hypothetical, constitutionally proper statute but seeks to be allowed to raise the speech rights of other parties not before the court. This accommodation is largely for the sake of an early judicial assessment that the regulation at issue may unduly deter others from engaging in constitutionally protected speech. See, e.g., id. at 198 (citing State v. Nye, 943 P.2d 96, 102 (Mont. 1997)).

  17. Id. at 199 (emphasis added). The statute also requires the use of not necessarily ‘obscene’ but at least ‘lewd’ or ‘profane’----and thus commonly protected----language. Id. at 198. See, e.g., Cohen v. California, 403 U.S. 15 (1971); Mahanoy Area Sch. Dist. v. B.L., 141 S. Ct. 2038 (2021).

  18. Petition for Writ of Certiorari, Lamoureux, 485 P.3d 192 (No. 21-427), cert. denied, 142 S. Ct. 860 (2022).

  19. Id. at 1.

  20. Id.

  21. Id.

  22. See Brief in Opposition to Petition for Writ of Certiorari at 3, Lamoureux, 485 P.3d 192 (No. 21-427), cert. denied, 142 S. Ct. 860 (2022).

  23. See infra Parts II–IV.

  24. For an expression of doubt in this regard, see Ams. for Prosperity Found. v. Bonta, 141 S. Ct. 2373, 2389–91 (2021) (Thomas, J., concurring) (first quoting United States v. Sineneng-Smith, 140 S. Ct. 1575, 1583 (2020) (Thomas, J., concurring); and then quoting Borden v. United States, 141 S. Ct. 1817, 1836 (2021) (Thomas, J., concurring)) (raising issues of separation of powers, standing, and advisory opinions).

  25. See supra notes 2–6 and accompanying text. For a concise review, see United States v. Henson, 705 F. App’x 348, 352 (6th Cir. 2017) (unpublished) (discussing United States v. Williams, 553 U.S. 285, 292–93 (2008)).

  26. Broadrick v. Oklahoma, 413 U.S. 601 (1973).

  27. Id. at 612.

  28. Id. (quoting Dombrowski v. Pfister, 380 U.S. 479, 486 (1965)).

  29. Id.

  30. See id.

  31. Id.

  32. See id. at 615.

  33. Id.

  34. For background, see generally R. George Wright, Substantial Burdens in the Law, 46 Sw. L. Rev. 1 (2016).

  35. For background, see generally R. George Wright, What Counts as “Speech” in the First Place?: Determining the Scope of the Free Speech Clause, 37 Pepp. L. Rev. 1217 (2010).

  36. See supra note 31 and accompanying text.

  37. See supra note 31 and accompanying text.

  38. See supra note 33 and accompanying text.

  39. United States v. Stevens, 599 U.S. 460, 473 (2010) (quoting Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 449 n.6 (2008)).

  40. See, e.g., id.

  41. See, e.g., United States v. Hansen, 25 F.4th 1103, 1109–10 (9th Cir. 2022) (quoting City of Houston v. Hill, 482 U.S. 451, 467 (1987)).

  42. Id.

  43. Id. at 1109 (quoting Members of City Council of L.A. v. Taxpayers for Vincent, 466 U.S. 789, 801 (1984)).

  44. Taxpayers for Vincent, 466 U.S. at 801.

  45. Stevens, 559 U.S. at 474. This alarm was raised despite the statute’s express exclusion of speech of numerous sorts of typical value categories. Id. at 465. For a less alarmed response to the statute, see, e.g., Steven H. Shiffrin, The Dark Side of the First Amendment, 61 UCLA L. Rev. 1480, 1488–92 (2014).

  46. Ex parte Thompson, 442 S.W.3d 325, 350 (Tex. Crim. App. 2014) (quoting Stevens, 559 U.S. at 474).

  47. Id. (quoting Ex parte Lo, 424 S.W.3d 10, 25 n.71 (Tex. Crim. App. 2013)); State v. Chen, 615 S.W.3d 376, 384 (Tex. App.—Houston [14th Dist.] 2020, pet. granted), vacated by Ex parte Chen, No. PD-0096-21, PD-0097-21, 2022 WL 17171098 (Tex. Crim. App. Nov. 23, 2022) (along with the assertion that the statute “has the potential to sweep up large swaths of protected speech”). For the metaphor of one class of statutory applications ‘dwarfing’ another, see United States v. Miselis, 972 F.3d 518, 541 (4th Cir. 2020).

  48. Hansen, 25 F.4th at 1110.

  49. See supra note 30 and accompanying text.

  50. Miselis, 972 F.3d at 530 (quoting Preston v. Leake, 660 F.3d 726, 738 (4th Cir. 2011)).

  51. See, e.g., United States v. Bonin, 932 F.3d 523, 537 (7th Cir. 2019).

  52. United States v. Ackell, 907 F.3d 67, 77 (1st Cir. 2018).

  53. Id. at 74 n.3.

  54. See, e.g., Texas v. Johnson, 491 U.S. 397, 404 (1989).

  55. Id. (quoting Spence v. Washington, 418 U.S. 405, 409 (1974)). See generally Wright, supra note 35.

  56. See Virginia v. Hicks, 539 U.S. 113, 124 (2003); Broadrick v. Oklahoma, 413 U.S. 601, 615 (1973); Ackell, 907 F.3d at 73; United States v. Blair, No. ELH-19-00410, 2021 WL 4339132, at *24 (D. Md. Sept. 23, 2021).

  57. See, e.g., Ackell, 907 F.3d at 73.

  58. Id.

  59. See, e.g., courts categorizing such messages as courses of conduct under various state and federal statutes, Hicks, 539 U.S. at 123–24; United States v. Gonzales, 905 F.3d 165, 190 & n.10 (3d Cir. 2018); Ackell, 907 F.3d at 73; United States v. Conlan, 786 F.3d 380, 386 (5th Cir. 2015); United States v. Osinger, 753 F.3d 939, 944 (9th Cir. 2014); United States v. Petrovic, 701 F.3d 849, 856 (8th Cir. 2012); United States v. Shrader, 675 F.3d 300, 312 (4th Cir. 2012); United States v. Storme, No. 20 CR 650, 2021 WL 3674611, at *4 (N.D. Ill. Aug. 19, 2021). In an as-applied case, see United States v. Waggy, 936 F.3d 1014, 1019–20 (9th Cir. 2019).

  60. Wollschlaeger v. Governor of Fla., 848 F.3d 1293, 1308 (11th Cir. 2017) (en banc).

  61. Id. (quoting King v. Governor of N.J., 767 F.3d 216, 228 (3d Cir. 2014)).

  62. See, e.g., R. George Wright, A Variable Number of Cheers for Viewpoint-Based Regulations of Speech, 96 Notre Dame L. Rev. Reflection 82 (2021), https://scholarship.law.nd.edu/ndlr_online/vol96/iss2/ [https://perma.cc/GZ8B-WEJH]; R. George Wright, Sharpening the Focus of Free Speech Law: The Crucial Role of Government Intent, 9 Tex. A&M L. Rev. 405 (2021).

  63. Swank v. Smart, 898 F.2d 1247, 1250–51 (7th Cir. 1990).

  64. For mainstream accounts of the basic aims, purposes, and values thought to be promoted by accommodating freedom of speech, see generally Kent Greenawalt, Free Speech Justifications, 89 Colum. L. Rev. 119 (1989), as well as Frederick Schauer, Free Speech: A Philosophical Enquiry (1982).

  65. See Swank, 898 F.2d at 1250–51; see also Trejo v. Shoben, 319 F.3d 878, 887 (7th Cir. 2003) (“‘Casual chit-chat between two persons or otherwise confined to a small group . . . is not protected’ under the First Amendment”) (quoting Swank, 898 F.2d at 1251) (detailing a professional discipline case involving interpersonal comments and phone calls of a social nature); H’s Bar, LLC v. Berg, No. 20-cv-1134, 2020 WL 6827964, at *4 (S.D. Ill. Nov. 21, 2020); Marshall v. Town of Merrillville, 228 F. Supp.3d 853, 864 (N.D. Ind. 2017); Robert Post, Recuperating First Amendment Doctrine, 47 Stan. L. Rev. 1249, 1255 (1995).

  66. See Swank, 898 F.2d at 1251.

  67. See supra note 64. For a classic treatment, see John Stuart Mill, On Liberty 115–28 (Gertrude Himmelfarb ed. 1974) (1859).

  68. See, e.g., William P. Marshall, In Defense of the Search for Truth as a First Amendment Justification, 30 Ga. L. Rev. 1 (1995); Frederick Schauer, Reflections on the Value of Truth, 41 Case W. Res. L. Rev. 699 (1991).

  69. See, e.g., Alexander Meiklejohn, Free Speech and Its Relation to Self-Government (1948); Ashutosh Bhagwat, The Democratic First Amendment, 110 Nw. U. L. Rev. 1097 (2016).

  70. See, e.g., Floyd Abrams, The Soul of the First Amendment 22 (2017); C. Edwin Baker, Autonomy and Free Speech, 27 Const. Comment. 251 (2011).

  71. Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949); see also Bible Believers v. Wayne Cnty., 805 F.3d 228, 249–50 (6th Cir. 2015) (en banc).

  72. See Terminiello, 337 U.S. at 3–4.

  73. See supra notes 67–70 and accompanying text.

  74. See, e.g., Mary Anne Franks, “Revenge Porn” Reform: A View from the Front Lines, 69 Fla. L. Rev. 1251, 1266–68 (2017); R. George Wright, Cyber Harassment and the Scope of Freedom of Speech, 53 U.C. Davis L. Rev. Online 187, 190 (2020).

  75. See, e.g., the public-employee, speech-discipline case of Pickering v. Bd. of Educ., 391 U.S. 563, 571–72 (1968). For development of the concept, see generally R. George Wright, Speech on Matters of Public Interest and Concern, 37 DePaul L. Rev. 27 (1987).

  76. See, e.g., for the Court’s discussion of the value of public debate for promoting self-government and “a meaningful dialogue of ideas” in Snyder v. Phelps, 562 U.S. 443, 451–52 (2011) (quoting Dun & Bradstreet, Inc. v. Greenmoss Builders, Inc., 472 U. S. 749, 760).

  77. See, e.g., Bd. of Cnty. Comm’rs v. Umbehr, 518 U.S. 668, 675 (1996); Waters v. Churchill, 511 U.S. 661, 668 (1994) (plurality opinion).

  78. See, e.g., Phila. Newspapers, Inc. v. Hepps, 475 U.S. 767, 774 (1986); Order Denying Certiorari, 470 U.S. 1009, 1012 (1985) (Brennan, J., dissenting from denial of certiorari); Snyder, 562 U.S. at 451–52; United States v. Fleury, 20 F.4th 1353, 1364 (11th Cir. 2021); State v. Katz, 179 N.E.3d 431, 452 (Ind. 2022).

  79. See, e.g., In re J.J.M., 265 A.3d 246, 286 (Pa. 2021) (Todd, J., concurring) (“The . . . overbreadth doctrine does not simply say that laws restricting speech should not prohibit too much speech. It says that laws restricting speech should not prohibit too much speech that is protected by the First Amendment.”) (quoting United States v. Heineman, 767 F.3d 970, 978–79 (10th Cir. 2014)).

  80. See, e.g., Bd. of Trs. of State Univ. v. Fox, 492 U.S. 469, 481 (1989); Vill. of Hoffman Ests. v. Flipside, Hoffman Ests., 455 U.S. 489, 497 (1982) (“[T]he overbreadth doctrine does not apply to commercial speech.”); Melendez v. City of New York, 16 F.4th 992, 1011 n.37 (2d Cir. 2021) (quoting Flipside, 455 U.S. at 497).

  81. See, for example, the cyber harassment case of United States v. Osinger, 753 F.3d 939, 948 (9th Cir. 2014) (“[T]he public has no legitimate interest in the private sexual activities of [the victim] or in the embarrassing facts revealed about her life . . . .”) (quoting United States v. Petrovic, 701 F.3d 849, 856 (8th Cir. 2012)).

  82. See supra notes 7–16 and accompanying text; State v. Lamoureux, 485 P.3d 192 (2021).

  83. Lamoureux, 485 P.3d at 196–97.

  84. Id. at 196.

  85. Id.

  86. Id.

  87. Id. at 196–97.

  88. Id.

  89. Id. at 196 n.1, 197 (citing Mont. Code Ann. § 45-8-213(1)(a) (2017)).

  90. See supra notes 83–89 and accompanying text.

  91. See supra notes 83–89 and accompanying text.

  92. United States v. Fredrickson, 996 F.3d 821 (7th Cir. 2021).

  93. Id. at 822–23. The offense was also referred to as the “sexual exploitation of a minor.” Id. at 823.

  94. Id. at 823.

  95. Id.

  96. Id.

  97. Id. (quoting 18 U.S.C. § 2251(a)).

  98. For further such examples, see, e.g., United States v. Ackell, 907 F.3d 67, 70–71 (1st Cir. 2018) (describing how the stalking defendant warned the seventeen-year-old in question “that if she stopped sending him photos, he would disseminate photos of her that he had saved among her friends, classmates, and family”); United States v. Storme, No. 20 CR 650, 2021 WL 3674611, at *1, *3 (N.D. Ill. Aug. 19, 2021) (relying unduly on a speech versus conduct distinction) (“After breaking up with his girlfriend in May 2020, Defendant . . . began an online campaign to intimidate and harass her,” along with disseminating nonconsensually obtained personal information); State v. Cartwright, 487 P.3d 737, 744 (Idaho 2021) (regarding a statute prohibiting the enticement of persons under age sixteen to engage in prohibited sexual acts); State v. Bonner, 61 P.3d 611, 612–13 (Idaho Ct. App. 2002) (“Police determined that he had secretly videotaped the girl in various states of undress by standing on the stepstool to make a videotape recording through a gap in the blinds covering a window.”); State v. Katz, 179 N.E.3d 431, 439 (Ind. 2022) (deciding a revenge porn case in which the defendant was alleged to have, “unbeknownst to his girlfriend . . . captured cell phone video of her performing oral sex on him, [and] then sent it to another person”).

  99. See Miller v. California, 413 U.S. 15 (1973); see also New York v. Ferber, 458 U.S. 747 (1982); Ashcroft v. Free Speech Coal., 535 U.S. 234 (2002).

  100. See, e.g., United States v. Henson, 705 F. App’x 348, 353 (6th Cir. 2017) (unpublished) (referring to reports that approximately “one in five teenagers [have engaged] in [some form of] sexting”).

  101. See, e.g., Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949); see supra notes 82–100 and accompanying text; infra notes 102–104 and accompanying text.

  102. See, in the course of upholding an overbreadth claim, State v. Chen, 615 S.W.3d 376, 380, 383, 385 (Tex. App.—Houston [14th Dist.] 2020, pet. granted), vacated and remanded, No. PD-0096-21, PD-0097-21, 2022 WL 17171098 (Tex. Crim. App. Nov. 23, 2022).

  103. See, e.g., the classic Vietnam War-era military draft recruiting case of Gooding v. Wilson, 405 U.S. 518 (1972) (state disorderly conduct statute, as authoritatively construed, held to be overbroad); United States v. Miselis, 972 F.3d 518 (4th Cir. 2020) (addressing the federal Anti-Riot Act, as often deployed in explicitly political contexts); United States v. Rundo, 990 F.3d 709 (9th Cir. 2021). For a discussion of Gooding, see Harry Kalven, Jr., A Worthy Tradition: Freedom of Speech in America 110–11 (Jamie Kalven eds., 1988).

  104. For one such possible case, see Lewis v. New Orleans, 415 U.S. 130, 131–32 (1974).

  105. For one approach to reducing the number and significance of such cases, see generally Wright, supra note 75.

  106. See, e.g., United States v. Dutcher, 851 F.3d 757 (7th Cir. 2017) (alleging repeated threats against the life of President Obama); see also United States v. Baker, 514 F. Supp. 3d 1369 (N.D. Fla. 2021) (finding that a threat to corral those storming the state Capitol was a “true threat”). A true, and therefore, prohibitable threat may implicate the marketplace of ideas and fall, however perversely, within the scope of broadly conceived political discourse. For discussion, see State v. Jorgenson, 946 N.W.2d 596, 602–03, 605 (Minn. 2020); Renee Griffin, Note, Searching for Truth in the First Amendment’s True Threat Doctrine, 120 Mich. L. Rev. 721 (2022).

  107. Virginia v. Hicks, 539 U.S. 113, 119 (2003). It may also be important to ask whether the deterred speech is likely to be redundant. Other undeterred speakers may express a deterred speaker’s ideas more widely, articulately, and forcefully, thereby placing the idea in question on the public issue agenda.

  108. Id.

  109. Id. But at some cost: perhaps the cost of not applying the speech restriction to speech that could and should be restricted—for the sake of some significant social good—by a constitutionally legitimate regulation. Id.

  110. See, e.g., FCC v. Pacifica Found., 438 U.S. 726, 743 n.18 (1978) (plurality opinion) (“A requirement that indecent language be avoided will have its primary effect on the form, rather than the content, of serious communication.”). For critical discussion, see Heidi Kitrosser, From Marshall McLuhan to Anthropomorphic Cows: Communicative Manner and the First Amendment, 96 Nw. Univ. L. Rev. 1339, 1356–59 (2002).

  111. For a discussion, see generally R. George Wright, The Unnecessary Complexity of Free Speech Law and the Central Importance of Alternative Speech Channels, 9 Pace L. Rev. 57 (1989). See also Members of City Council of L.A. v. Taxpayers for Vincent, 466 U.S. 789, 812 (1984) (“[A] restriction on expressive activity may be invalid if the remaining modes of communication are inadequate.”).

  112. For background, see supra notes 64–70 and accompanying text.

  113. See Wright, supra note 111, at 63, 79.

  114. See Taxpayers for Vincent, 466 U.S. at 802, 812 (describing the type of speech utilized by the underlying speakers who were seeking to invoke the overbreadth doctrine).

  115. See id. at 812.

  116. See supra Part III.

  117. See generally the factual circumstances cited as examples supra Part III.

  118. See supra notes 77–78 and accompanying text.

  119. United States v. Waggy, 936 F.3d 1014 (9th Cir. 2019).

  120. Id. at 1018. The court of appeals majority took the odd and unnecessary view that the defendant’s language in the case amounted to conduct rather than speech. Id. at 1019–20.

  121. Id. at 1015, 1018; see also Wash. Rev. Code § 9.61.230(1)(a)–(b) (applying to federal land within the territory of the state).

  122. Waggy, 936 F.3d at 1016–17.

  123. Id. at 1016.

  124. Id.

  125. Id. at 1016–17.

  126. See supra notes 64–70 and accompanying text.

  127. Consider, in a different context, whether COVID-19 legislation should be aimed at slowing the disease’s transmission; reducing the spread; reducing the number of cases or their severity; reducing deaths; or reducing the impact on vulnerable groups.

  128. See, classically, Mancur Olson, Jr., The Logic of Collective Action (1965); see also Mancur Olson, The Rise and Decline of Nations: Economic Growth, Stagflation, and Social Rigidities (1982).

  129. See Peter H. Schuck, Why Government Fails So Often: And How It Can Do Better 224 (Princeton Univ. Press) (2014).

  130. For background, see Barr v. Am. Ass’n of Pol. Consultants, 140 S. Ct. 2335, 2354 (2020); United States v. Booker, 543 U.S. 220, 258–59 (2005); United States v. Rundo, 990 F.3d 709, 720–21 (9th Cir. 2021); United States v. Miselis, 972 F.3d 518, 541–43 (4th Cir. 2020); Nat’l Mining Ass’n v. Zinke, 877 F.3d 845, 865–66 (9th Cir. 2017); United States v. Massey, No. 21 CR 142, 2022 WL 79870, at *2–3 (N.D. Ill. Jan. 7, 2022); State v. Jones, 492 P.3d 433, 445 (Kan. 2021).

  131. See Barr, 140 S. Ct. at 2353–54; Booker, 543 U.S. at 258–59; Nat’l Mining Ass’n, 877 F.3d at 864–65; Massey, 2022 WL 79870 at *3; Jones, 492 P.3d at 931–32. As opposed to the complete invalidation of the entire statute, see Rundo, 990 F.3d at 720; Miselis, 972 F.3d at 543–44.

  132. See Barr, 140 S. Ct. at 2353; Booker, 543 U.S. 220 at 258–59; Rundo, 990 F.3d at 720–21; Miselis, 972 F.3d at 541–43; Nat’l Mining Ass’n, 877 F.3d at 864–65; Massey, 2022 WL 79870 *3; Jones, 492 P.3d at 931–32.

  133. Rundo, 990 F.3d at 720–21, 720 nn.20–21 (construing the Anti-Riot Act, 18 U.S.C. §§ 2101(a), 2102(b)).

  134. See id. at 720–21.

  135. As an example, see e.g., supra note 127.

  136. Consider, for example, state statutory lottery programs that were intended to incentivize vaccination, as discussed in Anica C. Law et al., Lottery-Based Incentives and COVID-19 Vaccination Rates in the U.S., 182 JAMA Internal Med. 235, 235 (2022), https://www.jamanetwork.com/journals/amainternalmedicine/fullarticle/2787782 [https://perma.cc/7EZY-VPR6].

  137. In particular, potential speakers do not normally face a binary choice between risky speech and silence; some prudent and otherwise appropriate alternative expression may be available. See supra notes 110–126 and accompanying text.

  138. For background, see Schuck, supra note 129.

  139. United States v. Massey, No. 21 CR 142, 2022 WL 79870, at *2, *4 (N.D. Ill. Jan. 07, 2022).

  140. Id. at *2–3.

  141. Id. at *1.

  142. Id.

  143. See, e.g., Snyder v. Phelps, 562 U.S. 443, 459 (2011); Terminiello v. City of Chicago, 337 U.S. 1, 4 (1949); Cohen v. California, 403 U.S. 15, 21 (1971).

  144. See Snyder, 562 U.S. at 459; Terminiello, 337 U.S. at 4; Cohen, 403 U.S. at 21.

  145. State v. Richards, 896 P.2d 357, 360–61 (Idaho Ct. App. 1995) (including references to an “intent to annoy . . . or offend”).

  146. Id. at 362.

  147. See, e.g., the flag burning case of Texas v. Johnson, 491 U.S. 397, 405–06 (1989).

  148. United States v. Williams, 553 U.S. 285, 292 (2008) (second emphasis added); see also Ams. for Prosperity Found. v. Bonta, 141 S. Ct. 2373, 2387 (2021) (quoting United States v. Stevens, 559 U.S. 460, 473 (2010)); Virginia v. Hicks, 539 U.S. 113, 118–19 (2003); Broadrick v. Oklahoma, 413 U.S. 601, 615 (1973).

  149. Perhaps some minimal intimation of this possibility is suggested by In re A.J.B., 929 N.W.2d 840 (Minn. 2019) (comparing the relative “sweep” of state and federal stalking statutes).