SUPREME COURT OF THE UNITED STATES
No. 22–138
BILLY RAYMOND COUNTERMAN, PETITIONER v. COLORADO
ON WRIT OF CERTIORARI TO THE COURT OF APPEALS OF COLORADO
[June 27, 2023]
Justice Sotomayor, with whom Justice Gorsuch joins as to Parts I, II, III–A, and III–B, concurring in part and concurring in the judgment.
When the government seeks to punish speech based on its content, the First Amendment typically imposes stringent requirements. This ensures that the government, even when pursuing compelling objectives, does not unduly burden our Nation’s commitment to free expression. “From 1791 to the present, however, the First Amendment has permitted restrictions upon the content of speech in a few limited areas.” United States v. Stevens, 559 U. S. 460, 468 (2010) (internal quotation marks omitted). These categories must be “well-defined and narrowly limited” in light of the serious consequences that flow from carving out speech from ordinary First Amendment protections. Chaplinsky v. New Hampshire, 315 U. S. 568, 571 (1942).
“True threats” are one such category, and there is a tradition of criminalizing threats stretching back centuries. This includes punishing single utterances based on the message conveyed. One paradigmatic example of this would be writing and mailing a letter threatening to assassinate the President. Such laws are plainly important. There is no longstanding tradition, however, of punishing speech merely because it is unintentionally threatening. Instead, this Court’s precedent, along with historical statutes and cases, reflect a commonsense understanding that threatening someone is an intentional act. As to what intent is needed, “[t]raditionally, one intends certain consequences when he desires that his acts cause those consequences or knows that those consequences are substantially certain to result from his acts.” Tison v. Arizona, 481 U. S. 137, 150 (1987) (internal quotation marks omitted). This does not require showing that an individual intends to carry through with the threat. But it does require showing that an individual desires to threaten or is substantially certain that her statements will be understood as threatening.
Today, unfortunately, the Court unnecessarily departs from this traditional understanding. That is not to say that I disagree with the Court on everything. Far from it. I join the Court’s conclusion that some subjective mens rea is required in true-threats cases. I also agree that in this particular case, where petitioner was prosecuted for stalking that involved threatening statements, a mens rea of recklessness is amply sufficient. Where I part ways with the Court is that I would not reach the distinct and more complex question whether a mens rea of recklessness is sufficient for true-threats prosecutions generally. Further, requiring nothing more than a mens rea of recklessness is inconsistent with precedent, history, and the commitment to even harmful speech that the First Amendment enshrines. I therefore respectfully concur only in part and in the judgment.
I
As an initial matter, I do not believe that this Court should reach the question whether recklessness is sufficient for true-threats prosecutions. A key conceptual distinction is helpful for explaining why. On the one hand, there are statements that are objectively threatening. In some cases, such statements can be punished because they fall into the unprotected category of “true threats.” Yet such statements can also be punished if they fall into another category of unprotected speech, such as speech integral to criminal conduct. Or they might warrant less First Amendment protection for other reasons. On the other hand, there is the question of what constitutes the well-defined and longstanding category of unprotected true threats. It is with this latter question that I do not see the need to address whether a mens rea of recklessness is sufficient across the board.
First, the courts below did not address whether recklessness was sufficient to prosecute true threats and neither of the actual parties have advocated a recklessness standard. Colorado disclaimed the idea that recklessness was required, and petitioner asserted, correctly, that recklessness had not been raised under traditional principles of party presentation. The briefing on recklessness consists almost entirely of a few pages of an argument in the alternative at the tail end of an amicus brief filed by the United States.
Second, because petitioner was prosecuted for stalking involving threatening speech, this case does not require resort to the true-threats exemption to the First Amendment.
True-threats doctrine covers content-based prosecutions for single utterances of “pure speech,” which need not even be communicated to the subject of the threat. Watts v. United States, 394 U. S. 705, 707 (1969) (per curiam). The First Amendment would normally place strict limits on such prosecutions. So there is typically a need to determine whether the speech in question falls within the traditionally unprotected category of true threats.
This is not such a case, however. Petitioner was convicted for “stalking [causing] serious emotional distress” for a combination of threatening statements and repeated, unwanted, direct contact with C. W. 497 P. 3d 1039, 1043 (Colo. App. 2021).[1] This kind of prosecution raises fewer First Amendment concerns for a variety of reasons. Stalking can be carried out through speech but need not be, which requires less First Amendment scrutiny when speech is swept in. See, e.g., Rumsfeld v. Forum for Academic and Institutional Rights, Inc., 547 U. S. 47, 62 (2006). The content of the repeated communications can sometimes be irrelevant, such as persistently calling someone and hanging up, or a stream of “utterly prosaic” communications. Ante, at 1. Repeatedly forcing intrusive communications directly into the personal life of “an unwilling recipient” also enjoys less protection. Rowan v. Post Office Dept., 397 U. S. 728, 738 (1970). Finally, while there is considerable risk with a single intemperate utterance that a speaker will “accidentally or erroneously incur liability,” ante, at 7 (internal quotation marks and alterations omitted), that risk is far reduced with a course of repeated unwanted contact. Take, for example, petitioner continuously contacting C. W. despite her blocking him.
Given this, prosecuting threatening statements made as part of a course of stalking does not squarely present the hardest questions about the mens rea required to prosecute isolated utterances based solely on their content.[2] True-threats doctrine came up below only because of the lower courts’ doubtful assumption that petitioner could be prosecuted only if his actions fell under the true-threats exception. I do not think that is accurate, given the lessened First Amendment concerns at issue. In such cases, recklessness is amply sufficient. And I would stop there. There is simply no need to reach out in this stalking case to determine whether anything more than recklessness is needed for punishing true threats generally.
II
Lest there be any doubt, the First Amendment stakes around the definition of “true threats” are high indeed. The First Amendment’s mantle covers speech that is “vituperative, abusive and inexact.” Watts, 394 U. S., at 708. “It might be tempting to dismiss” seemingly low-value speech “as unworthy of … robust First Amendment protections.” Mahanoy Area School Dist. v. B. L., 594 U. S. ___, ___ (2021) (slip op., at 11). Yet “[m]ost of what we say to one another lacks ‘religious, political, scientific, educational, journalistic, historical, or artistic value’ (let alone serious value), but it is still sheltered from Government regulation.” Stevens, 559 U. S., at 479 (emphasis deleted). First Amendment vigilance is especially important when speech is disturbing, frightening, or painful, because the undesirability of such speech will place a heavy thumb in favor of silencing it. In response, the Court has upheld First Amendment rights in the context of gruesome animal cruelty videos, id., at 472; cross burning, Virginia v. Black, 538 U. S. 343, 347–348 (2003); hateful rhetoric in protests of the funerals of fallen soldiers, Snyder v. Phelps, 562 U. S. 443, 448–449, 458 (2011); and computer-generated images of child pornography, Ashcroft v. Free Speech Coalition, 535 U. S. 234, 239–240, 258 (2002).
The risk of overcriminalizing upsetting or frightening speech has only been increased by the internet. Our society’s discourse occurs more and more in “the ‘vast democratic forums of the Internet’ in general, and social media in particular.” Packingham v. North Carolina, 582 U. S. 98, 104 (2017) (citation omitted). “Rapid changes in the dynamics of communication and information transmission” have led to equally rapid and ever-evolving changes “in what society accepts as proper behavior.” Ontario v. Quon, 560 U. S. 746, 759 (2010). Different corners of the internet have considerably different norms around appropriate speech. Online communication can also lack many normal contextual clues, such as who is speaking, tone of voice, and expression. Moreover, it is easy for speech made in a one context to inadvertently reach a larger audience.
Without sufficient protection for unintentionally threatening speech, a high school student who is still learning norms around appropriate language could easily go to prison for sending another student violent music lyrics, or for unreflectingly using language he read in an online forum. “[A] drunken joke” in bad taste can lead to criminal prosecution. Perez v. Florida, 580 U. S. 1187 (2016) (Sotomayor, J., concurring in denial of certiorari). In the heat of the moment, someone may post an enraged comment under a news story about a controversial topic. Another person might reply equally heatedly. In a Nation that has never been timid about its opinions, political or otherwise, this is commonplace.
Many of this Court’s true-threats cases involve such charged political speech. See Black, 538 U. S., at 348–349 (Ku Klux Klan rally); Watts, 394 U. S., at 707 (antiwar protest); Rogers v. United States, 422 U. S. 35, 41–42, 47–48 (1975) (Marshall, J., concurring) (opposition to Nixon’s policies toward China). Amici give further contemporary examples of such speech from across the political spectrum. See, e.g., Brief for American Civil Liberties Union et al. as Amici Curiae 24–29. Much of this speech exists in a gray area where it will be quite hard to predict whether a jury would find it threatening. And the ubiquity of such speech raises the possibility of highly discretionary enforcement.
The burdens of overcriminalization will fall hardest on certain groups. A jury’s determination of when angry hyperbole crosses the line will depend on amorphous norms around language, which will vary greatly from one discursive community to another. Juries’ decisions will reflect their “background knowledge and media consumption.” Minnesota Voters Alliance v. Mansky, 585 U. S. ___, ___ (2018) (slip op., at 17). “[S]peakers whose ideas or views occupy the fringes of our society have more to fear, for their violent and extreme rhetoric, even if intended simply to convey an idea or express displeasure, is more likely to strike a reasonable person as threatening.” United States v. White, 670 F. 3d 498, 525 (CA4 2012) (Floyd, J., concurring in part and dissenting in part). Members of certain groups, including religious and cultural minorities, can also use language that is more susceptible to being misinterpreted by outsiders. And unfortunately yet predictably, racial and cultural stereotypes can also influence whether speech is perceived as dangerous. See, e.g., A. Dunbar, C. Kubrin, & N. Scurich, The Threatening Nature of “Rap” Music, 22 J. Psychol. Pub. Pol’y & L. 281, 281–282, 288–290 (2016).
On the other hand, the internet has also made stalking and harassment even easier. Stalking can be devastating and dangerous. See Brief for First Amendment Scholars as Amici Curiae 7–8. Lives can be ruined, and in the most tragic instances, lives are lost. Ibid. Harassers can hide behind online anonymity while tormenting others. This happens in the context of intimate relationships and it happens with strangers. Overly constraining our society’s ability to respond to stalking would come at a real cost. For the reasons given, however, a mens rea standard for true threats would not hinder stalking prosecutions. See supra, at 3–5.
Even isolated threatening speech can do real harm. Such speech not only disrupts lives, it can silence the speech of others who become afraid to speak out. A mens rea requirement would not, however, present an uncommon or insurmountable barrier to true-threats prosecutions.[3] Nonetheless, under such a standard, there will be some speech that some find threatening that will not and should not land anyone in prison.
III
These high First Amendment stakes are further reason for caution when delineating the boundaries of what constitutes a true threat. In undertaking that analysis, the Court and I part ways on the order of operations. The Court begins by defining true threats as all objectively threatening speech, entirely independent of whether the speaker intended to be threatening, ante, at 6, and the lead dissent agrees, post, at 2–3 (opinion of Barrett, J.). The Court gets there by relying on this Court’s interpretation of the word “threat” in a federal statute. Ante, at 6 (citing Elonis v. United States, 575 U. S. 723, 733 (2015)). The Court declares all such speech categorically unprotected, and then asks what “buffer zone” is needed in order to protect other, unthreatening speech. See ante, at 4–7.
Respectfully, I see the analysis differently. The first step in the analysis should instead be to ask about the scope of the well-defined and narrow category of “true threats” as a constitutional matter. This Court has already warned about the danger of creating new categories of “unprotected speech” exempt from the ordinary First Amendment framework for balancing our society’s commitment to free expression with other interests. Stevens, 559 U. S., at 470. If courts were at liberty to redefine what counts as a “threat” or “defamation” at will, this would achieve the same results as creating new categories of unprotected speech.
Thus, the Court must first ask whether there is a longstanding tradition of punishing inadvertent threats as “true threats.” This Court’s prior definition of the word “threat” in a federal statute, looking primarily to dictionaries, Elonis, 575 U. S., at 733, does not tell us the scope of “true threats” for First Amendment purposes. Elonis itself made clear that it did “not … consider any First Amendment issues.” Id., at 740. Instead, a careful examination of this Court’s true-threats precedent and the history of threat crimes does not support a long-settled tradition of punishing inadvertently threatening speech.
A
A natural place to begin, one might think, would be with this Court’s most recent decision involving the First Amendment, mens rea, and true threats. Yet to read the Court’s decision, one would have little idea that in a seminal 2003 decision, this Court held that a threat conviction could not stand because of an insufficient mens rea requirement. See Black, 538 U. S. 343. Black plainly sets out a conception of true threats as including a mens rea requirement.
In Black, the Court confronted the constitutionality of a Virginia statute that prohibited burning a cross with intent to intimidate. Only part of the decision in Black is contained in a five-Justice majority opinion. The other relevant parts of the decision were written by the Members of that majority, who split into a four-Justice plurality and Justice Scalia’s partial concurrence in judgment.
The majority explained why a prohibition on cross burning with intent to threaten was constitutional, beginning by defining the category of true threats. “ ‘True threats,’ ” the majority explained “encompass those statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence.” Id., at 359 (emphasis added). However, “[t]he speaker need not actually intend to carry out the threat,” as true threats also include intimidation alone. Id., at 359–360. And “[i]ntimidation in the constitutionally proscribable sense of the word is a type of true threat, where a speaker directs a threat to a person or group of persons with the intent of placing the victim in fear of bodily harm or death.” Id., at 360 (emphasis added).
To the extent the Virginia statute covered intentionally threatening cross burning, it was thus tailored to cover only true threats. Critically, however, the statute also provided that “ ‘[a]ny such burning of a cross shall be prima facie evidence of an intent to intimidate.’ ” Id., at 348. In other words, the all-important intent requirement could be satisfied by the mere conduct itself.
Consistent with the majority’s definition of true threats, both the plurality and Justice Scalia agreed that the lack of a sufficient intent requirement meant that a conviction under the statute could not stand. Id., at 367, 379. For the plurality, the intent requirement was “the very reason why a State may ban cross burning” because it “distinguish[ed]” between the constitutionally unprotected true threat of burning a cross with intent to intimidate and “cross burning [as] a statement of ideology.” Id., at 365–366.[4] For Justice Scalia, the “plurality [was] correct in all of this.” Id., at 372 (opinion concurring in part, concurring in judgment in part, and dissenting in part). There was a constitutional need for a distinction between cross burning “ ‘intended to intimidate’ ” and cross burning as “ ‘a statement of ideology.’ ” Ibid. The plurality and Justice Scalia only parted ways as to whether to hold that the statute was “facially invalid,” id., at 367 (plurality opinion), or just that the jury instructions made it unclear “whether the jury has rendered its verdict (as it must)” with sufficient consideration of “intent to intimidate,” id., at 380 (opinion of Scalia, J.) (emphasis added).
The through-line is not hard to discern. First, unprotected true threats include a subjective mens rea requirement. Id., at 360 (majority opinion). Second, as a result, “Virginia’s statute does not run afoul of the First Amendment insofar as it bans cross burning with intent to intimidate.” Id., at 362 (majority opinion). Third, a conviction could not stand if it had categorically dispensed with that intent requirement, id., at 365–366 (plurality opinion), or if the jury had insufficiently considered “intent to intimidate,” id., at 380 (opinion of Scalia, J.).
In sum, all five Justices in the Black majority agreed that a true-threats prosecution could not stand under the First Amendment without a sufficient subjective mens rea requirement.[5]
B
In defining true threats as “statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence,” id., at 359, the Court in Black echoed the traditional understanding of threats. Historically, threat crimes covered the same kind of subjectively threatening speech Black invoked.
In reviewing this history, it is also vital to keep in mind the nature of the inquiry. Removing speech from normal First Amendment scrutiny is a major shift in the balance of expression and public interest that our Constitution generally strikes. The inquiry is therefore whether there is a “long-settled tradition” of prohibiting inadvertently threatening speech. Stevens, 559 U. S., at 469. None of the other opinions, however, identify a historical case that expressly raised the question whether a subjective mens rea is required and held that it is not. That is a remarkable thing when one considers that the sample size consists of decisions from both sides of the Atlantic across centuries.
There was a long tradition of crimes for threatening another person in order to extort them. See, e.g., 1796 N. J. Laws §57, p. 108. Colorado and the United States admit that this core category of threat crimes required intent.
Even beyond that, a subjective mens rea remained a key component of threat offenses. An 18th-century English statute made it a capital offense to “knowingly send any letter … threatening to kill or murder any of his Majesty’s subject or subjects” or to threaten arson. 27 Geo. II, c. 15, in 21 Eng. Stat. at Large 184 (1754). A leading treatise explained that the statute was “levelled against such whose intention it was [to] obtain their object by creating terror in [the victim’s] mind.” 2 W. Russell & D. Davis, Crimes & Misdemeanors *1845 (emphasis added).
Consistent with this, defendants were convicted of “knowingly, wilfully, and feloniously” sending threatening letters. Rex v. Tyler, 1 Mood. 428, 168 Eng. Rep. 1330 (1835); Rex v. Paddle, Russ. & Ry. 484, 168 Eng. Rep. 910 (1822) (indictment for “knowingly, unlawfully, wickedly, and feloniously” sending a threatening letter); see also King v. Girdwood, 1 Leach 142, 168 Eng. Rep. 173 (1776) (indictment for “feloniously” sending a threatening letter). “ ‘[K]nowingly and wilfully’ effecting any result applies to those who know that the acts performed will have that effect, and perform them with the intention that such shall be their operation.” 12 American and English Encyclopaedia of Law 522–524 (J. Merrill ed. 1890); see also J. Boag, Imperial Lexicon of the English Language 530 (1850) (defining “felonious” as “with the deliberate purpose to commit a crime”).
The necessary mens rea could sometimes be inferred from the content of the letter, but could be rebutted by other evidence. See King v. Philipps, 6 East 464, 475, 102 Eng. Rep. 1365, 1369 (1805). Courts thus considered “the threat intended to be made by the prisoner” and “what he meant by what he had written” in determining whether he had violated the statute. Regina v. Hill, 5 Cox 233, 235 (Crim. Cas. 1851); see also King v. John and Mary Hammond, 1 Leach 444, 446, 168 Eng. Rep. 324, 325 (1787) (describing the offense of sending a threatening letter “to the party whose fears the threat it contains was calculated to alarm”).
Threat laws in the United States were of a piece. Some state laws about threats expressly required maliciousness. See Me. Rev. Stat., Tit. 12, ch. 154, §26 (1840); 1884 La. Acts No. 64, §1, p. 86. Courts more generally emphasized the importance of a mens rea requirement. See, e.g., State v. Benedict, 11 Vt. 236, 239 (1839). The North Carolina Supreme Court, for example, singled out threats as quintessential examples of offenses where it is “necessary” to prove the “intent of the particular letter.” State v. Murphy, 84 N. C. 742, 743–744 (1881). And where state statutes may have been silent on intent to threaten, courts read such requirements in. See Commonwealth v. Morton, 140 Ky. 628, 631, 131 S. W. 506, 507–508 (1910) (letter must be “calculated to alarm, disturb, intimidate, or injure”); see also State v. Stewart, 90 Mo. 507, 512, 2 S. W. 790, 792 (1887) (jury instruction requiring that “ ‘defendant intended to threaten’ ”).
Leading treatises also explained the importance of mens rea. See 25 American and English Encyclopaedia of Law 1071 (C. Williams ed. 1894) (when there is a question as to “whether or not the letter contains the threat alleged, the intent is a question for the jury”); see also 2 R. Anderson, Wharton’s Criminal Law and Procedure §803, pp. 659–660 (1957) (threats must be “intended to put the person threatened in fear of bodily harm”); 2 J. Bishop, Commentaries on the Criminal Law §1201, p. 664 (6th ed. 1877) (“The intent, both under the unwritten law and under the statutes, must be evil”).
Against that backdrop, I return to the inquiry at hand: whether there is a “long-settled” or “well-established” history of prosecuting inadvertently threatening speech. There is no line of cases or pattern of statutes affirmatively stating that an objective standard is sufficient.
C
Put together, Black and the history point to an intent requirement. When Black defined and analyzed true threats in terms of intent, there is no reason to think the Court used intent to mean anything less than its traditional definition of purpose or knowledge. See, e.g., Tison, 481 U. S., at 150. Nor would a recklessness standard play the necessary role of distinguishing between cross burning that is “ ‘intended to intimidate’ … and nonintimidating cross burning [that] cannot be prohibited.” 538 U. S., at 372 (opinion of Scalia, J.). Given the violent history of the symbol, it is hard to imagine that any politically motivated cross burning done within view of the public could be carried out without awareness of some risk a reasonable spectator would feel threatened. See id., at 388–391 (Thomas, J., dissenting). Recklessness, which turns so heavily on an objective person standard, would not have been enough.
As to the history, it is true that over time courts have often used a wide variety of terms to describe mental states. See, e.g., Morissette v. United States, 342 U. S. 246, 252 (1952). Yet “[t]he element of intent in the criminal law has traditionally been viewed as a bifurcated concept embracing either the specific requirement of purpose or the more general one of knowledge or awareness.” United States v. United States Gypsum Co., 438 U. S. 422, 445 (1978); see also Tison, 481 U. S., at 150; Carter v. United States, 530 U. S. 255, 270 (2000) (describing “feloniously” as equivalent to “ ‘intent’ ”). And at the very least, there is no well-settled history showing that it is enough for a defendant to be merely aware of some risk that their statements could be threatening. See, e.g., Borden v. United States, 593 U. S. ___, ___ (2021) (plurality opinion) (slip op., at 5) (recklessness requires awareness of a level of risk that “need not come anywhere close to a likelihood”). The history is, instead, replete with the enduring and commonsense pairing of threats and intent.
D
The Court, eschewing Black and history, instead reaches its result based on the need for a “buffer zone” drawn by analogy to other categories of unprotected speech. Ante, at 4. For the reasons above, I do not think we can leap ahead to this question. With that caveat, I agree with the Court that precedent in other areas of unprotected speech and concerns about chilling support a subjective mens rea requirement for true threats. Yet these same chilling concerns only further buttress the conclusion that true threats should be limited to intentionally threatening speech. Indeed, in the concurrence by Justice Marshall that the Court invokes, ante, at 9–10, he advocated “requir[ing] proof that the speaker intended his statement to be taken as a threat,” based on concerns about punishing “pure speech.” Rogers, 422 U. S., at 47–48. In determining the appropriate mens rea, the Court analogizes to three categories of traditionally unprotected speech: incitement, obscenity, and defamation. None of these warrants expanding the narrow boundaries of true threats.
1
Speech inciting harm is the closest cousin to speech threatening harm. Both incitement and threats put other people at risk, and both “sprin[g] from [Justice] Holmes’s ‘clear and present danger’ test.” G. Blakey & B. Murray, Threats, Free Speech, and the Jurisprudence of the Federal Criminal Law, 2002 B. Y. U. L. Rev. 829, 1069 (2002). Like true threats, incitement’s scope is defined in terms of both intention and effect, covering speech “[1] intended to produce, and [2] likely to produce, imminent disorder.” Hess v. Indiana, 414 U. S. 105, 109 (1973) (per curiam).
Despite their similar nature and source, the Court today draws a hard line between the two. Incitement requires “ ‘inten[t].’ ” Ante, at 8. While for threats, the speaker need only be “aware that others could regard his statements as threatening violence and delive[r] them anyway.” Ante, at 11 (internal quotation marks omitted). The Court justifies this asymmetry by the idea “that incitement to disorder is commonly a hair’s-breadth away from political ‘advocacy,’ ” ante, at 13, and the lead dissent says much the same, post, at 7 (opinion of Barrett, J.). These opinions offer little basis for distinguishing threats on this ground, as this Court’s own cases show time and again how true-threats prosecutions sweep in political speech. See Black, 538 U. S., at 348–349; Watts, 394 U. S., at 707 (antiwar protest); Rogers, 422 U. S., at 41–42 (Marshall, J., concurring) (opposition to Nixon’s policies toward China).[6] Not only that, but incitement itself is often only a hair’s-breadth away from threats.
Take the seminal incitement case NAACP v. Claiborne Hardware Co., 458 U. S. 886 (1982). During a civil rights boycott, NAACP leader Charles Evers, brother of the murdered civil rights hero Medgar Evers, gave a series of heated speeches. See id., at 898–902. He intoned that “boycott violators would be ‘disciplined’ ” and that “ ‘[i]f we catch any of you going in any of them racist stores, we’re gonna break your damn neck.’ ” Id., at 902. The Court acknowledged that in this charged context, these speeches “might have been understood as inviting an unlawful form of discipline or, at least, as intending to create a fear of violence.” Id., at 927. Yet inflammatory and threatening as these speeches were, they did not constitute incitement. That was because “there [was] no evidence—apart from the speeches themselves—that Evers authorized, ratified, or directly threatened acts of violence.” Id., at 929. His speeches were thus not “ ‘directed to inciting or producing imminent lawless action’ ” and he had not “specifically intended to further an unlawful goal.” Id., at 925, n. 68, 928.
Under a recklessness rule, Claiborne would have come out the other way. So long as Evers had some subjective awareness of some risk that a reasonable person could regard his statements as threatening, that would be sufficient. It would be quite troubling indeed to adopt a rule rendering this Court’s admirable defense of the First Amendment wrongly decided. Nor is Claiborne the only example. The foundational incitement case, Brandenburg v. Ohio, 395 U. S. 444 (1969) (per curiam), extended First Amendment protections to armed Klan members uttering racial slurs, a warning that “there might have to be some revengeance taken,” and plans for a “ ‘four hundred thousand strong’ ” march in two cities. Id., at 446. Then, as now, there would be at least some risk that a reasonable resident of those cities could feel threatened.
These concrete examples illustrate a more general principle. Speech inciting imminent and dangerous unlawful activity will reasonably be threatening to those who would be harmed by that illegality. In all such cases, whether seminal decisions by this Court or guilty pleas that barely see the inside of a courtroom, the Court’s decision effectively downgrades to recklessness the mens rea required for incitement of unlawful force; prosecutors could now simply charge such offenses as true threats. This is particularly worrisome because the standard for recklessness decreases the lower the “social utility” of the conduct. 1 W. LaFave, Substantive Criminal Law §5.4(f) (3d ed. 2018). That is a troubling standard for juries in a polarized nation to apply in cases involving heated political speech. This collateral damage can be avoided, however, if intent to threaten is understood as part of a true threat, just like intent to incite is part of incitement.
2
While obscenity is a step further afield of true threats and incitement, examination of this Court’s obscenity case law further supports an intent requirement for prosecutions of true threats.
The Constitution “ ‘requires proof of scienter’ ” in part “ ‘to compensate for the ambiguities inherent in the definition of obscenity.’ ” Hamling v. United States, 418 U. S. 87, 123 (1974). This is in line with this Court’s more general observation that “vagueness” of “content-based regulation of speech” is of “special concern” when it comes to “criminal statute[s].” Reno v. American Civil Liberties Union, 521 U. S. 844, 871–872 (1997).[7]
Specifically, the Court has held that a “knowledge” mens rea is sufficient for obscenity: “It is constitutionally sufficient that the prosecution show that a defendant had knowledge of the contents of the materials he distributed, and that he knew the character and nature of the materials.” Hamling, 418 U. S., at 123. This ensures that “not innocent but calculated purveyance of filth … is exorcised.” Id., at 122 (internal quotation marks omitted). While the Court today asserts that this Court has “never determined the precise mens rea” for obscenity, ante, at 13, n. 6, the Court has cited a knowledge standard approvingly for half a century, see Hamling, 418 U. S., at 123; Elonis, 575 U. S., at 739.[8] Applying that standard to threats, the “ ‘calculated purveyance’ of a threat would require that [a defendant] know the threatening nature of his communication.” Id., at 739.
The considerations that drove this Court to approve a higher mens rea for obscenity apply here as well. With obscenity, the ambiguity comes partly from the reliance on “ ‘contemporary community standards’ ” to define what is obscene. Hamling, 418 U. S., at 129. Such a standard is notoriously amorphous, and will change a great deal between communities and over time. The same chilling concerns apply to true threats. A recklessness standard based on what a reasonable person could find threatening will depend on ever-shifting community norms around language and when heated speech crosses the line from overly aggressive to criminal. See supra, at 5–7.[9]
3
Finally, the Court relies heavily upon this Court’s framework for defamation. Specifically, the Court analogizes to the “reckless disregard” standard for defamation of public figures or punitive damages for certain claims involving private figures. New York Times Co. v. Sullivan, 376 U. S. 254, 279–280 (1964).
Yet while civil defamation may be “the best known and best theorized example” of unprotected speech, ante, at 8, the same does not go for criminal prosecution of defamation. It is true that this Court in 1964 invalidated a prosecution for criminal libel for failing to apply the Sullivan standard, which covers “only those false statements made with a high degree of awareness of their probable falsity.” Garrison v. Louisiana, 379 U. S. 64, 75 (1964). Yet the Court expressed strong skepticism of the very concept of criminal prosecutions for libel and noted the salutary trend of its “virtual disappearance.” Id., at 69–70. The Court approvingly cited the Model Penal Code’s recommendation that criminal libel be limited to speech likely to cause a breach of the peace and “calculated” to do so. Id., at 70. This is not a promising theoretical springboard for determining the mens rea required to criminalize other speech.
If the Court were correct that the Sullivan standard is the appropriate analogy, however, then this standard should guide how to analyze recklessness in true-threats prosecutions. The generic formulation of recklessness requires that an individual disregard a relatively unspecified level of risk that the harm in question will occur. See Borden, 593 U. S., at ___ (plurality opinion) (slip op., at 5). Within that potentially broad range, Sullivan provides a more definite and demanding level of risk, reflecting the First Amendment concerns at stake. The Court has “made clear that the defendant must have made the false publication with a high degree of awareness of probable falsity or must have entertained serious doubts as to the truth.” Harte-Hanks Communications, Inc. v. Connaughton, 491 U. S. 657, 667 (1989) (internal quotation marks and ellipsis omitted). This makes sense. Allowing liability for awareness of a small chance that a story may be false would undermine the very shield Sullivan erects.
For similar reasons, after today’s ruling, future courts grappling with how to articulate the appropriate level of recklessness in true-threats cases would be well served to consult the Sullivan standard. The equivalent to Sullivan for true threats would require a high degree of awareness that a statement was probably threatening or serious doubts as to the threatening nature of the statement. This could avoid the chilling that would arise from a more amorphous and easily satisfied standard.
4
This Court’s various frameworks for unprotected speech do not speak with one voice, as perhaps befits the First Amendment. The above survey does not, however, give reason to depart from the traditional understanding of true threats. To the contrary, this case law supports keeping true threats within their traditional bounds. Incitement similarly requires intent. The same chilling concerns that have led this Court to approve a knowledge requirement for obscenity are present with true threats. And to the extent the civil defamation context is relevant, at the very least, it points to a precise and demanding form of recklessness.[10]
IV
Maintaining true threats doctrine within its traditional boundaries will guard against the overcriminalization of a wide range of political, artistic, and everyday speech based on its content alone. This does not mean that unintentionally threatening communications are exempt from regulation, far from it. As explained above, there are far fewer First Amendment concerns with stalking laws that punish repeated, targeted, unwanted conduct and accompanying speech. For that reason, recklessness is quite sufficient. As to true threats, intent is neither an unusual nor an insurmountable bar. “[C]ourts and juries every day pass upon knowledge, belief and intent … having before them no more than evidence of … words and conduct, from which, in ordinary human experience, mental condition can be inferred.” American Communications Assn. v. Douds, 339 U. S. 382, 411 (1950). *** I agree with the Court’s conclusion that the First Amendment requires a subjective mens rea in true-threats cases, and I also agree that recklessness is amply sufficient for this case. Yet I would stop there, leaving for another day the question of the specific mens rea required to prosecute true threats generally. If that question is reached, however, the answer is that true threats encompass a narrow band of intentional threats. Especially in a climate of intense polarization, it is dangerous to allow criminal prosecutions for heated words based solely on an amorphous recklessness standard. Our society has often concluded that an intent standard sets a proper balance between safety and the need for a guilty mind, even in cases that do not involve the First Amendment. Surely when the power of the State is called upon to imprison someone based on the content of their words alone, this standard cannot be considered excessive. Because I part ways with the Court on this score, I respectfully concur only in part and in the judgment.
- ↑ The statute of conviction applies to someone who “[r]epeatedly follows, approaches, contacts, places under surveillance, or makes any form of communication with another person … in a manner that would cause a reasonable person to suffer serious emotional distress and does cause that person … serious emotional distress.” Colo. Rev. Stat. §18–3–602(1)(c) (2022).
- ↑ For these reasons, stalking prosecutions that do not rely on the content of communications would raise even fewer First Amendment concerns.
- ↑ Intent requirements are common, including for incitement that results in actual violence, not just the threat of it. See infra, at 15–17. For that reason there are longstanding frameworks for determining when someone is not guilty by reason of insanity, and when delusions do (and do not) defeat a showing of intent. See, e.g., 1 W. LaFave, Substantive Criminal Law §§7.1(a), (b) (3d ed. 2018); 2 id., §9.2.
- ↑ The lead dissent asserts that the Black plurality’s decision was based on how the statute “ ‘ignore[d] all of the contextual factors that are necessary to decide whether a particular cross burning’ was covered by the statute.” Post, at 9 (opinion of Barrett, J.) (quoting 538 U. S., at 367 (plurality opinion)). But some context is missing from this reading itself. The full sentence is “all of the contextual factors that are necessary to decide whether a particular cross burning is intended to intimidate.” Id., at 367 (emphasis added). The plurality was thus concerned with context to the extent it was relevant to the mens rea requirement needed to render the statute constitutional. Id., at 365–366.
- ↑ According to the Court today and the lead dissent, however, Black somehow managed not to say anything about the First Amendment mens rea requirement for true-threats prosecutions—while striking down a true-threat conviction under the First Amendment for an insufficient mens rea requirement. On this reading, Black only discussed intent because “the statute involved in the case required a showing of intent.” Ante, at 6, n. 3; post, at 9, n. 4 (discussion of intent was “a reference to the statutory requirements for a conviction, not the constitutional requirements”). This puzzling interpretation does not explain why an illusory mens rea requirement in a Virginia law would pose any First Amendment problems if the Amendment did not impose a mens rea requirement of this kind. After all, “[w]hy would the First Amendment care how a jury goes about finding an [intent] element that is a matter of indifference to the Amendment?” United States v. Heineman, 767 F. 3d 970, 980 (CA10 2014). The obvious answer, from Black’s reasoning to its holding, is that such a mens rea requirement was necessary for the statute to target true threats.
- ↑ Nor is this limited to decisions by this Court. Threats cases sweep in political speech. See, e.g., State v. Taylor, 379 N. C. 589, 590, 866 S. E. 2d 740, 744 (2021). Incitement cases can sweep in nonpolitical speech. See, e.g., Rice v. Paladin Enterprises, Inc., 128 F. 3d 233, 264, n. 11, 267 (CA4 1997). And still other cases show how incitement and threats can often go hand in hand. See, e.g., State v. Caroll, 456 N. J. Super. 520, 544–545, 196 A. 3d 106, 120–121 (App. Div. 2018).
- ↑ Analogously, the Court’s civil defamation case law recognizes that heightened liability can require a heightened mens rea; even as to nonpublic figures, a higher standard must be met for punitive damages in certain cases. See, e.g., Gertz v. Robert Welch, Inc., 418 U. S. 323, 349–350 (1974).
- ↑ The Court has held, however, that recklessness is sufficient for child pornography. See Osborne v. Ohio, 495 U. S. 103, 115 (1990). This Court has emphasized time and again how child pornography is “a special case” because “[t]he market for child pornography [is] ‘intrinsically related’ to the underlying abuse” and thus “ ‘an integral part of the production of such materials, an activity illegal throughout the Nation.’ ” United States v. Stevens, 559 U. S. 460, 471 (2010) (quoting New York v. Ferber, 458 U. S. 747, 759, 761 (1982)); see also Osborne, 495 U. S., at 110–111. Child pornography, with its integral ties to separate criminal conduct, is not a strong analogue for threats, which can be fleeting statements in total isolation from any other criminality (though it is a stronger analogy to threats as part of an unlawful course of stalking). Yet the Court’s decision today puts child pornography on a First Amendment par with overheated political speech or violent song lyrics.
- ↑ There is a further safeguard in obscenity cases. Something is obscene if “taken as a whole, [it] lacks serious literary, artistic, political, or scientific value.” Ashcroft v. American Civil Liberties Union, 535 U. S. 564, 574 (2002) (internal quotation marks omitted). An intent requirement can provide a similar safeguard for threats. As Virginia v. Black, 538 U. S. 343 (2003), explained, requiring intent distinguishes between speech intended to intimidate and speech intended to express a political statement. Id., at 365–366 (plurality opinion); id., at 372 (opinion of Scalia, J.).
- ↑ The lead dissent headlines its analysis by pointing to this Court’s case law on “fighting words.” Post, at 3–4 (opinion of Barrett, J.). This is an unlikely candidate for a broader theory of the First Amendment. For “nearly three-quarters of a century … the Court has never … upheld a fighting words conviction” and “[t]he cumulative impact of [the Court’s] decisions is to make it unlikely that a fighting words law could survive.” E. Chemerinsky, The First Amendment 1094 (6th ed. 2019). It is not hard to see why such convictions would be unlikely to pass First Amendment muster; the leading case involved a Jehovah’s Witness distributing literature who was arrested for breach of the peace for calling a public official a “ ‘damned Fascist.’ ” Chaplinsky v. New Hampshire, 315 U. S. 568, 569, 573–574 (1942). Drawing upon a conviction like the one in Chaplinksy as the proper model for criminalizing political speech is proof itself of the serious risks with the lead dissent’s approach. In any event, as to the question at hand, when such breach of the peace offenses involved threats, intent to threaten was required. See 2 R. Anderson, Wharton’s Criminal Law and Procedure §803, pp. 659–660 (1957).