*70 A jury convicted David Ackell of one count of stalking in violation of 18 U.S.C. § 2261A. He now brings a First Amendment challenge to that statute, in addition to challenging the district court's jury instructions and arguing that insufficient evidence supported his conviction. Ackell's constitutional challenge does not succeed. We discern no error in the district court's jury instructions. And lastly, we hold that sufficient evidence supported Ackell's conviction. We therefore affirm.

I.

A.

We begin with an overview of the relevant facts. Because this appeal pertains in part to Ackell's motion for acquittal before the district court, "we recount the facts here 'in the light most favorable to the government.' " United States v. Fernández-Jorge , 894 F.3d 36 , 41 (1st Cir. 2018) (quoting United States v. Acevedo , 882 F.3d 251 , 257 (1st Cir. 2018) ).

Ackell and R.R. met online in 2012 during R.R.'s sophomore year of high school. To get around the requirements of the website on which they met-the now-defunct MyYearbook.com-R.R. held herself out as an eighteen-year-old, though she was actually only sixteen. Ackell's profile represented that he was twenty-one years old, but during his first conversation with R.R., he told her he was actually thirty-two. This was also false-Ackell was actually over forty at the time. The two began to regularly converse online. Eventually, Ackell told R.R. that if she sent him photos of herself, he would send her money in return. R.R. sent Ackell photos of herself partially clothed. She testified, though, that despite providing Ackell with a P.O. Box address, he never sent her money.

Around five months after R.R. and Ackell first began communicating online, Ackell proposed that they enter into a "dominant-submissive" relationship, in which R.R. would be "the submissive." R.R., who was now seventeen, did not know what this meant, so she did some research on the internet. R.R. testified that she came to understand that, under such an arrangement, Ackell would be "the boss," and that if he told her to "pose in a particular way ... [she] would pose in that way." Ultimately, R.R. agreed to enter into a relationship of this sort with Ackell. R.R. testified that, though her research into dominant-submissive relationships indicated that "[t]ypically there's supposed to be a safe word," she and Ackell did not have a safe word.

R.R. also testified that, after their dominant-submissive relationship commenced, Ackell began to treat her differently than before-and in a way that departed from her expectations about what the relationship would entail. For example, Ackell would call her "slave," or "caged butterfly," and insist that she address him as "owner" and tell him that she loved him. He also frequently demanded that R.R. send him sexually explicit photos of herself.

R.R. eventually told Ackell that she felt uncomfortable and wanted to end their dominant-submissive relationship. Ackell, *71 however, informed R.R. that she could not opt out of the relationship because she was "caged." Ackell also warned R.R. that if she stopped sending him photos, he would disseminate photos of her that he had saved among her friends, classmates, and family. R.R. testified that twice, she called Ackell "begging and pleading with him to ... delete all of [her] stuff and let [her] go." But, Ackell told her that he would not, because she was "trapped" and a "caged butterfly." In January of 2014, R.R. temporarily succeeded in terminating her relationship with Ackell after leading him to believe that her mother had discovered their relationship and was upset. Ackell resumed contacting her, though, and soon afterwards, R.R. told her father about her relationship with Ackell. R.R.'s father instructed her to take screenshots of her past conversations with Ackell and then delete those messages. Her father then contacted law enforcement.

B.

On July 29, 2015, a grand jury returned an indictment charging Ackell with one count of stalking. See 18 U.S.C. § 2261A(2)(B). Ackell moved to dismiss the indictment as insufficient, and on the grounds that § 2261A(2)(B) violates the First Amendment. On July 27, 2016, a grand jury returned a superseding indictment specifying that Ackell had committed the one count charged through "the sending of text messages, digital images and other electronic communications." Ackell renewed his original motion to dismiss as to the superseding indictment. The district court ordered the government to file a bill of particulars. See Fed. R. Crim. P. 7(f). But, it denied Ackell's motion to dismiss, finding the indictment "neither statutorily nor constitutionally deficient," and also rejecting his First Amendment challenge.

Ackell proceeded to trial. The jury found him guilty, and he then moved for a judgment of acquittal. See Fed. R. Crim. P. 29. The district court denied his motion, finding that sufficient evidence supported his conviction. The district court then sentenced him to thirty-three months of imprisonment. Ackell now appeals: (1) the district court's denial of his First Amendment challenge to the anti-stalking statute; (2) the district court's jury instructions; and (3) the district court's denial of his motion for acquittal.

II.

A.

As to Ackell's First Amendment challenge to the federal anti-stalking statute, he presses that § 2261A(2)(B) is both facially overbroad and a content-based restriction on speech that does not survive strict scrutiny. We consider these arguments sequentially, reviewing the district court's holding de novo because it involves only questions of law. See United States v. Floyd , 740 F.3d 22 , 38 (1st Cir. 2014).

1.

Ackell does not claim that the conduct underlying his conviction was protected by the First Amendment. Rather, Ackell asserts that § 2261A(2)(B) cannot be applied to anyone because it is overbroad under the First Amendment, even though it has been constitutionally applied to him. "The traditional rule is that a person to whom a statute may constitutionally be applied may not challenge that statute on the ground that it may conceivably be applied unconstitutionally to others in situations not before the Court." United States v. Sayer , 748 F.3d 425 , 434-35 (1st Cir. 2014) (quoting New York v. Ferber , 458 U.S. 747 , 767, 102 S.Ct. 3348 , 73 L.Ed.2d 1113 (1982) ). The Supreme Court, however, *72 has "altered its traditional rules of standing" in a small number of contexts, "but only because of the most 'weighty countervailing policies.' " Broadrick v. Oklahoma , 413 U.S. 601 , 611-12, 93 S.Ct. 2908 , 37 L.Ed.2d 830 (1973) (quoting United States v. Raines , 362 U.S. 17 , 22-23, 80 S.Ct. 519 , 4 L.Ed.2d 524 (1960) ). This is the case with the First Amendment overbreadth doctrine. In this context, the Court has seen fit to slacken its standing requirements in response to the "concern that the threat of enforcement of an overbroad law may deter or 'chill' constitutionally protected speech-especially when the overbroad statute imposes criminal sanctions." Virginia v. Hicks , 539 U.S. 113 , 119, 123 S.Ct. 2191 , 156 L.Ed.2d 148 (2003). Thus, even when a law may be applied to a particular individual in a constitutionally unobjectionable way, if that individual can show that the law is facially overbroad-that is, that it "punishes a 'substantial' amount of protected free speech, 'judged in relation to the statute's plainly legitimate sweep,' "-the proper remedy is to "invalidate all enforcement of that law." Id. at 118-119 , 123 S.Ct. 2191 (quoting Broadrick , 413 U.S. at 615 , 93 S.Ct. 2908 ) ).

The Supreme Court has cautioned that "[r]arely, if ever, will an overbreadth challenge succeed against a law or regulation that is not specifically addressed to speech or to conduct necessarily associated with speech (such as picketing or demonstrating)." Hicks , 539 U.S. at 124 , 123 S.Ct. 2191 . Thus, we begin our analysis by ascertaining § 2261A(2)(B)'s aim, as well as its potential for punishing protected speech. See United States v. Williams , 553 U.S. 285 , 293, 128 S.Ct. 1830 , 170 L.Ed.2d 650 (2008) ("The first step in overbreadth analysis is to construe the challenged statute; it is impossible to determine whether a statute reaches too far without first knowing what the statute covers."). As is relevant here, § 2261A(2)(B) penalizes whoever:

with the intent to kill, injure, harass, intimidate, or place under surveillance with intent to kill, injure, harass, or intimidate another person, uses the mail, any interactive computer service or electronic communication service or electronic communication system of interstate commerce, or any other facility of interstate or foreign commerce to engage in a course of conduct that ... causes, attempts to cause, or would be reasonably expected to cause substantial emotional distress to [that] person [or an immediate family member, spouse, or intimate partner of that person.] 1

Hence, to properly secure a conviction under § 2261A(2)(B), the prosecution must prove that: (1) the defendant had the requisite intent; (2) the defendant "engage[d] in a course of conduct"; (3) the defendant used a facility of interstate commerce; and (4) the defendant's "course of conduct" "cause[d], attempt[ed] to cause, or would be reasonably expected to cause substantial emotional distress." A "course of conduct" is "a pattern of conduct composed of 2 or more acts, evidencing a continuity of purpose." 18 U.S.C. § 2266 (2). Ackell's First Amendment challenge *73 pertains to all but the "facility of interstate commerce" element.

By its own terms, § 2261A(2)(B) regulates not speech, but conduct-or, to be precise, "course[s] of conduct." "Conduct," of course, may also enjoy First Amendment protection if it is "sufficiently imbued with elements of communication." Texas v. Johnson , 491 U.S. 397 , 404, 109 S.Ct. 2533 , 105 L.Ed.2d 342 (1989) (quoting Spence v. Washington , 418 U.S. 405 , 409, 94 S.Ct. 2727 , 41 L.Ed.2d 842 (1974) ). Yet, "the overbreadth doctrine's concern with 'chilling' protected speech 'attenuates as the otherwise unprotected behavior that it forbids [legislatures] to sanction moves from "pure speech" toward conduct.' " Hicks , 539 U.S. at 124 , 123 S.Ct. 2191 (quoting Broadrick , 413 U.S. at 615 , 93 S.Ct. 2908 ).

Before continuing, it is important to note that we rejected an overbreadth challenge to a since-amended version of § 2261A in United States v. Sayer , 748 F.3d 425 , 436 (1st Cir. 2014). As part of the Violence Against Women Reauthorization Act of 2013, 2 Congress amended § 2261A so that it now differs in two ways from the version we considered in Sayer . See Pub. L. No. 113-4, § 107(b) (2013) (codified at 18 U.S.C. § 2261A ). First, the intent to "intimidate" is now listed among the offense's various possible mental states. Id. Second, the statute now proscribes engaging, with the requisite intent, in a course of conduct that "causes, attempts to cause, or would be reasonably expected to cause substantial emotional distress." Id. (emphasis added). While Sayer did not expressly state that it was treating § 2261A(2)(B)'s precursor as a statute regulating conduct rather than speech, its analysis suggests that it did. See 748 F.3d at 434-36 (applying Hicks , placing the burden on Sayer rather than on the government to show substantial overbreadth, and referring to the statute as one that "clearly targets conduct ..."). Hence, this court's reading of § 2261A(2)(B)'s predecessor in Sayer supports our conclusion here that § 2261A(2)(B) targets conduct rather than speech.

At oral argument, Ackell insisted that § 2261A(2)(B) does target speech because it requires that a defendant have used "the mail, any interactive computer service or electronic communication service or electronic communication system of interstate commerce, or any other facility of interstate or foreign commerce." Ackell is correct that these enumerated facilities of interstate commerce are commonly employed to facilitate communication. Yet, while § 2261A(2)(B) could reach highly expressive conduct, it is plain from the statute's text that it covers countless amounts of unprotected conduct. For example, the government points out that:

a defendant could send envelopes of unknown white powder to the victim in the mail; he could send the victim nude photographs of herself; he could repeatedly infect the victim's computers with viruses; he could open unwanted on-line dating profiles under the victim's identity; he could take out unwanted loans in the victim's name; or he could arrange every day for deliveries to be made at the victim's home at all hours of the night.

As these examples illustrate, though the statute does name common means of communication among the possible facilities of interstate commerce one could use to commit the offense it defines, it does not necessarily follow that the statute targets speech. Moreover, we add that a defendant need not use the mail or the internet to violate the statute. The statute provides *74 these enumerated facilities of interstate commerce by way of example, but is also clear that one may take to "any other facility of interstate or foreign commerce" in violating it. See 18 U.S.C. § 2261A(2)(B). This further supports our conclusion that § 2261A(2)(B) does not target speech.

Moreover, our conclusion that § 2261A(2)(B) targets conduct is not inconsistent with our decision in March v. Mills , 867 F.3d 46 (1st Cir. 2017), where a similar analysis supported a different outcome. There, we considered a facial and as-applied challenge to a Maine statute forbidding, in simple terms, intentionally making noise to interfere with the operations of healthcare providers, after having been ordered by law enforcement not to do so. Id. at 51 . That statute implicated the First Amendment because it "restrict[ed] noisemaking even in public parks, plazas, sidewalks, [and] other traditional public fora." Id. at 53 (citing Hague v. Comm. for Indus. Org. , 307 U.S. 496 , 515-16, 59 S.Ct. 954 , 83 L.Ed. 1423 (1939) ). Similarly, in Cutting v. City of Portland , 802 F.3d 79 , 83 (1st Cir. 2015), the ordinance at issue proscribed sitting, standing, and staying at "median strips," which this court found were traditional public fora as "the people of Portland have used ... [them] for expressive purposes in much the same way that they have used parks and sidewalks." Finally, this line of cases is also consistent with the Supreme Court's holding in McCullen v. Coakley , --- U.S. ----, 134 S.Ct. 2518 , 189 L.Ed.2d 502 (2014). That case involved a challenge to a Massachusetts law creating a 35-foot "buffer zone" around reproductive health care facilities from which protesters were barred. Id. at 2526 . The Court highlighted that the law regulated access to public ways and sidewalks, which are "areas [that] occupy a 'special position in terms of First Amendment protection' because of their historic role as sites for discussion and debate." Id. at 2529 (quoting United States v. Grace , 461 U.S. 171 , 180, 103 S.Ct. 1702 , 75 L.Ed.2d 736 (1983) ). The Court explained that "even though the Act says nothing about speech on its face," because it restricted "access to traditional public fora" it was "subject to First Amendment scrutiny." Id. Section 2261A(2)(B), in contrast, does not implicate the interests that the First Amendment protects in a similar way.

We now turn to Ackell's arguments about the extent to which § 2261A(2)(B) criminalizes protected speech. Under Hicks , it is Ackell who bears the burden of demonstrating, "from the text of [the law] and from actual fact," that substantial overbreadth exists. Hicks , 539 U.S. at 122 , 123 S.Ct. 2191 (alteration in original) (quoting N. Y. State Club Ass'n v. City of New York , 487 U.S. 1 , 14, 108 S.Ct. 2225 , 101 L.Ed.2d 1 (1988).

As discussed above, the text of the law is clear in that it targets conduct, specifically "conduct performed with serious criminal intent," rather than speech protected by the First Amendment. Sayer , 748 F.3d at 435 . And while we do acknowledge that the Supreme Court has not categorically foreclosed the possibility that a statute that does not facially regulate speech could be facially overbroad under the First Amendment, see Hicks , 539 U.S. at 124 , 123 S.Ct. 2191 , as we discuss below, Ackell has not met his burden of demonstrating that factually, the statute could apply to a substantial amount of protected speech, in an absolute sense and in relation to its many legitimate applications. See Hicks , 539 U.S. at 119-120 , 123 S.Ct. 2191 . 3

*75 Exceptions to the First Amendment's protection of expression exist in the case of a small number of "well-defined and narrowly limited classes of speech, the prevention and punishment of which have never been thought to raise any Constitutional problem." United States v. Stevens , 559 U.S. 460 , 468-469, 130 S.Ct. 1577 , 176 L.Ed.2d 435 (2010) (quoting Chaplinsky v. New Hampshire , 315 U.S. 568 , 571-72, 62 S.Ct. 766 , 86 L.Ed. 1031 (1942) ). Two of these classes are relevant here-"true threats," see Virginia v. Black , 538 U.S. 343 , 359, 123 S.Ct. 1536 , 155 L.Ed.2d 535 (2003), and "speech integral to criminal conduct," see United States v. Alvarez , 567 U.S. 709 , 717, 132 S.Ct. 2537 , 183 L.Ed.2d 574 (2012) (citing Giboney v. Empire Storage & Ice Co. , 336 U.S. 490 , 498, 69 S.Ct. 684 , 93 L.Ed. 834 (1949) ). " 'True threats' encompass those statements where the speaker means to communicate a serious expression of an intent to commit an act of unlawful violence to a particular individual or group of individuals." Black , 538 U.S. at 359 , 123 S.Ct. 1536 . Speech "integral to criminal conduct" is precisely what it sounds like, and it is not protected on First Amendment grounds "merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed." Giboney , 336 U.S. at 502 , 69 S.Ct. 684 .

Ackell and Amici argue that both the statute's intent element and harm element allow it to reach speech that is neither a true threat nor integral to criminal conduct. With respect to the intent element, they maintain that speech made with merely an intent to "harass" or "intimidate" cannot amount to a true threat. And as concerns the harm element, Ackell and Amici argue that the reasonable-person standard embedded in the statute's harm-caused element criminalizes protected speech by allowing for a conviction when no harm has actually occurred.

We begin by stating the obvious: § 2261A(2)(B) is not a statute that is valid under "no set of circumstances." See Stevens , 559 U.S. at 472 , 130 S.Ct. 1577 (quoting United States v. Salerno , 481 U.S. 739 , 745, 107 S.Ct. 2095 , 95 L.Ed.2d 697 (1987) ). For example, one could be convicted for undertaking a course of conduct, "with the intent to kill" that "causes [the victim] substantial emotional distress." That conviction would not be constitutionally problematic. It is uncontroversial that, insofar as that course of conduct involved speech, that speech would fall outside of the First Amendment's protections as a *76 true threat and/or speech integral to criminal conduct.

Nevertheless, Ackell and Amici coalesce around a number of similar hypothetical examples illustrating how § 2261A(2)(B) reaches protected speech. In essence, they stress that an individual who, with merely the intention to harass, twice directs speech on a matter of public concern at someone-say, via Twitter-that could be "reasonably expected to cause substantial emotional distress," would have violated § 2261A(2)(B) even if the "victim" did not actually suffer any emotional distress. In so arguing, they press that discourse on matters of public concern can often be vituperative. And, pointing to Snyder v. Phelps , 562 U.S. 443 , 131 S.Ct. 1207 , 179 L.Ed.2d 172 (2011), and Hustler Magazine, Inc. v. Falwell , 485 U.S. 46 , 108 S.Ct. 876 , 99 L.Ed.2d 41 (1988), they argue that speech of that sort nonetheless enjoys First Amendment protection. See Snyder , 562 U.S. at 458 , 131 S.Ct. 1207 (holding that members of the Westboro Baptist Church could assert a First-Amendment defense to liability for the tort of intentional infliction of emotional distress because their protest at the funeral of a deceased solider was "on a matter of public concern" and therefore entitled to "special protection"); Hustler Magazine, Inc. , 485 U.S. at 53 , 108 S.Ct. 876 ("Generally speaking the law does not regard the intent to inflict emotional distress as one which should receive much solicitude.... But in the world of debate about public affairs, many things done with motives that are less than admirable are protected by the First Amendment.").

First, Sayer takes much force out of Ackell's arguments concerning the statute's "intent to ... harass" language. There, the appellant did not directly attack this feature of the statute. 748 F.3d at 435 . Nonetheless, we take the opinion in Sayer to indicate that we must read "intent to ... harass," as referring to criminal harassment, see id. , which is unprotected because it constitutes true threats or speech that is integral to proscribable criminal conduct. We think that this logic would also apply to the term "intimidate" in the current version of the statute. Indeed, "interpreting the statute to avoid a serious constitutional threat," Zadvydas v. Davis , 533 U.S. 678 , 699, 121 S.Ct. 2491 , 150 L.Ed.2d 653 (2001), points to reading the statute as referring to "[i]ntimidation in the constitutionally proscribable sense of the word[, which] is a type of true threat," Black , 538 U.S. at 344 , 123 S.Ct. 1536 . Moreover, Ackell and Amici are both correct that the "reasonable person standard" embedded in the harm element would permit a conviction in the absence of any actual harm. Yet they fail to articulate why exactly that would violate the First Amendment-to say nothing of the even higher bar they must clear amid a facial overbreadth challenge.

Second, the examples to which Ackell and Amici point in Snyder and Hustler Magazine, Inc. are regulated pursuant to laws that are far afield from the text of § 2261A (i.e., tort law). And while the government could not rule out that some activity analogous to those cases could be covered, nothing suggests to us, and Ackell has not demonstrated, that it certainly would be covered.

Finally, there is only one example of the statute, in its previous version, actually having been applied to protected conduct. See United States v. Cassidy , 814 F.Supp.2d 574 (D. Md. 2011) (finding § 2261A unconstitutional "as applied" to defendant who was anonymously harassing a religious leader via Twitter and a blog). However, just as the government is reluctant to state that Ackell's hypotheticals could be prosecuted under § 2261A(2), so *77 too the government states that "it is not clear that the evidence [in Cassidy ] would have met the Rule 29 standard for one of the required criminal intents if there had been a trial." And in Sayer , in which the defendant also pointed to Cassidy as an unconstitutional application of the cyberstalking statute, we held that one District Court precedent combined with a list of hypotheticals did not result in the defendant showing that the statute was substantially overbroad. Thus, in the absence of veridical examples, we are not inclined to rely on hypotheticals. See New York v. Ferber , 458 U.S. 747 , 781, 102 S.Ct. 3348 , 73 L.Ed.2d 1113 (1982) (Stevens, J., concurring) ("Hypothetical rulings are inherently treacherous and prone to lead us into unforeseen errors; they are qualitatively less reliable than the products of case-by-case adjudication.").

Ultimately-while acknowledging that § 2261A(2)(B) could have an unconstitutional application, and remaining cognizant of the chilling-effect-related concerns inherent in declining to invalidate a statute that can be applied to violate the First Amendment-we are unconvinced that we must administer the "strong medicine" of holding the statute facially overbroad. See Williams , 553 U.S. at 293 , 128 S.Ct. 1830 (quoting L.A. Police Dep't v. United Report Publ'g Corp. , 528 U.S. 32 , 39, 120 S.Ct. 483 , 145 L.Ed.2d 451 (1999) ). 4 The statute does not, on its face, regulate protected speech, or conduct that is necessarily intertwined with speech or expression. Should situations arise where the statute is applied to courses of conduct that are sufficiently expressive to implicate the First Amendment, we are confident that as-applied challenges will properly safeguard the rights that the First Amendment enshrines.

2.

Ackell also asserts that § 2261A(2)(B) is an impermissible content-based restriction on speech that is not sufficiently narrowly tailored to vindicate a compelling government interest. In assessing arguments of this stripe, the Supreme Court has instructed that we must first "consider[ ] whether a law is content neutral on its face before turning to the law's justification or purpose." Reed v. Town of Gilbert , --- U.S. ----, 135 S.Ct. 2218 , 2228, 192 L.Ed.2d 236 (2015) (emphasis in original). As concerns step one, a law is content-based if it "target[s] speech based on its communicative content." Id. at 2226 . But, implicit in this is also a "step zero": does the law in question target speech at all? See McCullen , 134 S.Ct. at 2529-31 (considering whether the challenged law was a content-based regulation only after determining that the law "restricts access to traditional public fora and is therefore subject to First Amendment scrutiny"); March , 867 F.3d at 53-54 . And we have already answered this question in the negative. See supra § II.A.1.

Amici presses that the statute "facially penalizes protected speech based on its *78 content and viewpoint." For support, Amici first cites Boos v. Barry , in which the Supreme Court found a District of Columbia ordinance prohibiting, within 500 feet of a foreign embassy, signs "tend[ing] to bring that foreign government into 'public odium' or 'public disrepute,' " to be content-based because it "regulate[d] speech due to its potential primary impact." 485 U.S. 312 , 315, 321, 108 S.Ct. 1157 , 99 L.Ed.2d 333 (1988). Amici also relies on Justice Kennedy's concurrence in Matal v. Tam , which advanced that "[a] law found to discriminate based on viewpoint is an 'egregious form of content discrimination[.]' " --- U.S. ----, 137 S.Ct. 1744 , 1766, 198 L.Ed.2d 366 (2017) (Kennedy, J., concurring) (quoting Rosenberger v. Rector & Visitors of Univ. of Va. , 515 U.S. 819 , 829-30, 115 S.Ct. 2510 , 132 L.Ed.2d 700 (1995) ). Tam involved an ultimately successful First Amendment challenge to a provision of the Lanham Act that prohibited "the registration of trademarks that may 'disparage ... or bring ... into contemp[t] or disrepute' any 'persons, living or dead.' " Id. at 1751 (majority opinion) (alteration and ellipses in original) (quoting 15 U.S.C. § 1052 (a) ).

These arguments, however, presuppose that § 2261A(2)(B) targets speech at all. Yet, comparing § 2261A(2)(B) to the laws at issue in Boos and Tam -which facially regulate pure speech or highly expressive conduct-provides further support for the notion that § 2261A(2)(B) does not. As a result, it cannot be so that § 2261A(2)(B) is an impermissible content- or viewpoint-based restriction on speech.

B.

Ackell purports to bring four different challenges to the district court's jury instructions. Two of those, however, are merely a repackaging of his First Amendment challenge to § 2261A(2)(B), so we need not consider them here. This leaves us with Ackell's arguments that the district court erred in (1) failing to instruct the jury that it had to be unanimous as to which specific acts formed Ackell's "course of conduct," and (2) in "instruct[ing] the jury that a course of conduct can 'attempt to cause' substantial emotional distress." Ackell duly preserved these arguments below. Thus, "[w]e consider de novo whether an instruction embodied an error of law, but we review for abuse of discretion whether the instructions adequately explained the law or whether they tended to confuse or mislead the jury on the controlling issues." United States v. Gray , 780 F.3d 458 , 464 (1st Cir. 2015) (quoting United States v. Symonevich , 688 F.3d 12 , 24 (1st Cir. 2012) ).

1.

Ackell's proposed jury instructions would have impressed upon the jury that

you may not find Mr. Ackell guilty of the charged offense unless you unanimously agree on which two or more text messages, digital images, and other electronic communications to R.R. form the course of conduct. By that I mean that it is not sufficient if you all agree that two or more of the texts, digital images, or electronic communications in evidence form the course of conduct, but cannot agree on which two.

The district court, however, declined to adopt these instructions, and rather instructed the jury that "you are not required to agree unanimously on which two or more acts constitute the course of conduct."

A jury's verdict-that is, its decision as to whether or not it finds the defendant guilty-must be unanimous. See Fed. R. Crim. P. 31(a). So too must a jury unanimously agree that the prosecution proved each element of the charged offense(s).

*79 Richardson v. United States , 526 U.S. 813 , 817, 119 S.Ct. 1707 , 143 L.Ed.2d 985 (1999). Unanimity is not necessary, in contrast, as to "the brute facts that constitute those elements." United States v. Lee , 317 F.3d 26 , 36 (1st Cir. 2003). "Thus, if a jury is confronted with divergent factual theories in support of the same ultimate issue, courts generally have held that the unanimity requirement is met as long as the jurors are in agreement on the ultimate issue (even though they may not be unanimous as to the precise theory)." Id. ; see also Schad v. Arizona , 501 U.S. 624 , 630-31, 111 S.Ct. 2491 , 115 L.Ed.2d 555 (1991). Consequently, a unanimity instruction was only necessary here if the specific acts making up Ackell's course of conduct were "elements" of the offense that § 2261A(2)(B) codifies. In other words, we ask whether those specific acts are "fact[s] strictly necessary to define the conduct prohibited under the statute of conviction." Lee , 317 F.3d at 37 .

And they plainly are not. Nothing in § 2261A(2)(B)'s text indicates that the acts comprising the "course of conduct" are themselves elements. Indeed, it is not § 2261A(2)(B) that defines a "course of conduct" as comprising "2 or more acts evidencing a continuity of purpose," but rather a separate statutory section. See 18 U.S.C. § 2266 (2). The most natural reading of the statute suggests that the relevant element is the existence of a course of conduct. The specific two-plus acts comprising that course of conduct are, in turn, the sort of "brute facts" for which unanimity is not required. See United States v. LaPlante , 714 F.3d 641 , 647 (1st Cir. 2013) (holding that, in the context of a prosecution for mail fraud, jury unanimity was not necessary as to "which particular false statement alleged in the indictment was used to carry out the fraud"); Lee , 317 F.3d at 36-41 (holding that, to properly convict the defendant under 18 U.S.C. § 1029 (a)(3), which prohibits possessing fifteen or more counterfeit credit cards, the jury did not need to be unanimous as to which fifteen credit cards (out of a possible twenty-two) the defendant had possessed); United States v. Verrecchia , 196 F.3d 294 , 298-300 (1st Cir. 1999) (holding that, to properly find a defendant guilty of possessing a firearm after having been convicted of a felony, under 18 U.S.C. § 922 (g), the jury need not be unanimous as to the particular firearm that the defendant possessed).

Ackell argues that Richardson compels the opposite result. In that case, the Supreme Court considered 21 U.S.C. § 848 , which forbids engaging in a "continuing criminal enterprise." Richardson , 526 U.S. at 815 , 119 S.Ct. 1707 . That statute provides, in relevant part, that

[A] person is engaged in a continuing criminal enterprise if-

(1) he violates any provision of [the federal drug laws, i.e. ,] this subchapter or subchapter II of this chapter the punishment for which is a felony, and

(2) such violation is a part of a continuing series of violations of [the federal drug laws, i.e. ,] this subchapter or subchapter II of this chapter-

Id. (alterations in original) (quoting 21 U.S.C. § 848 ). The question before the Court was whether the occurrence of a "series of violations" was an element, or whether each violation comprising that series was itself a separate element requiring jury unanimity. Id. at 817-18 , 119 S.Ct. 1707 . The Court concluded that each constituent violation was a distinct element. Id. at 824 , 119 S.Ct. 1707 .

In so holding, it reasoned that "[t]he words 'violates' and 'violations' are words that have a legal ring. A 'violation' is not simply an act or conduct; it is an act or conduct that is contrary to law."

*80 Id. at 818 , 119 S.Ct. 1707 (citing Black's Law Dictionary 1570 (6th ed. 1990) ). Because § 2261A contains no requirement that the acts making up the course of conduct themselves be illegal, this cuts sharply against the notion that those underlying acts are elements and not brute facts. Ackell also fails to point to any historical or fairness-related considerations comparable to those that pushed the Court in Richardson toward the conclusion that it was correct to treat each violation as an element. See id. at 818-22 , 119 S.Ct. 1707 . And we also note that, unlike § 2261A(2)(B), which does not itself define "course of conduct," the statute at issue in Richardson expressly specified what constitutes a "continuing criminal enterprise." See 21 U.S.C. § 848 (a)(1)-(2). This suggests that Congress did not intend for each act making up the course of conduct for which liability attaches to be a distinct element. See Lee , 317 F.3d at 38-39 (giving weight to Congress's apparent intent that the specific identity of the fifteen-plus counterfeit credit cards not be treated as an element).

Given all of this, Ackell has failed to convince us that the district court erred in denying his request for a unanimity instruction.

2.

Ackell also contends that the district court erred in instructing the jury that it needed to find that Ackell's course of conduct "caused substantial emotional distress to R.R., attempted to cause substantial emotional distress to R.R., or would be reasonably expected to cause substantial emotional distress to R.R." Ackell's proposed instruction would have omitted the words "attempted to cause substantial emotional distress to R.R." In declining to adopt that proposed instruction, the district court acknowledged that the statute's provision that a course of conduct may "attempt to cause substantial emotional distress" is "linguistically odd." Nonetheless, it elected to "instruct the jury in a way that tracks the statute as closely as possible."

Because the jury instructions tracked the statute's language-meaning that they cannot have embodied an error of law-we take Ackell to object to the district court's choice of words in instructing the jury. Our review, therefore, is for abuse of discretion. Gray , 780 F.3d at 464 . It is true that one does not usually think of "courses of conduct" as having volition. This does make the statute's provision that a defendant may be convicted for engaging in course of conduct that "attempts to cause ... substantial emotional distress," rather peculiar. See 18 U.S.C. § 2261A(2)(B). We note, however, that Ackell only challenges the district court's rejection of his proposed instructions. He does not, for example, object to the statute's wording on due process, or void-for-vagueness grounds. And nowhere in his First Amendment challenge to the statute does he contend that this feature renders it unconstitutional. The only question before us, then, is whether the district court abused its discretion in giving jury instructions that precisely tracked the statute's wording. We hold that it did not. See United States v. Hall , 434 F.3d 42 , 56 (1st Cir. 2006) (finding no abuse of discretion when "[t]he court's instruction closely tracked the language of the statute, which is a strong indicator that the charge passes muster").

C.

Lastly, we turn to Ackell's challenge to the district court's denial of his motion for acquittal. We review a district court's denial of a Rule 29 motion de novo , viewing the evidence in the light most favorable to the jury's guilty verdict.

*81 United States v. Santos-Soto , 799 F.3d 49 , 56-57 (1st Cir. 2015). We will affirm unless "the evidence is so scant that a rational factfinder could not conclude that the government proved all the essential elements of the charged crime beyond a reasonable doubt." United States v. Rodríguez-Vélez , 597 F.3d 32 , 39 (1st Cir. 2010) (emphasis in original).

Ackell argues that the government failed to introduce sufficient evidence to prove the intent and harm elements of § 2261A(2)(B). We consider these arguments in turn, concluding that both lack merit.

1.

With regard to the intent element, Ackell maintains, as he did before the district court, that until a conversation taking place over January 27 and 28, 2014-when, he concedes, R.R. made clear to him that she wanted to terminate their relationship-he "subjectively and even reasonably believed until that moment that he and [R.R] were in a consensual dominant/submissive fantasy relationship, and that any previous mild protestations she had made were in her role as the submissive." Thus, Ackell's argument goes, up until that date, he could not have had the "intent to kill, injure, harass, [or] intimidate" R.R. And, Ackell adds, the government failed to present sufficient evidence of any acts post-dating January 27-28, 2014 that could provide a basis for his conviction. But even if we accept that Ackell-believing that his conduct was taking place within the bounds of a consensual "dominant/submissive" relationship-could not have formed the requisite mental state until that date, we, like the district court, still conclude that his sufficiency-of-the-evidence challenge fails. We explain why.

The government introduced into evidence as Exhibits One and Two the screenshots that R.R. captured, at her father's direction, of conversations with Ackell. Exhibit One depicted an undated conversation between R.R. and Ackell on the smartphone messaging application Kik. Exhibit Two depicted a text-message conversation beginning on January 27, 2014-the conversation that purportedly alerted Ackell to R.R.'s desire to leave the relationship. The screenshots of that conversation in Exhibit Two indicate that the absolute latest moment at which Ackell could have realized that R.R. was not a consenting participant was when he-in response to R.R.'s plea that he delete the photos of her in his possession-asked "[s]ubmissive a lie as well?" To this, R.R. responded "I have a tendency to tell people what they want to hear. You wanted to hear I like to be submissive, which is o[nly] 25% true. I[']m being honest because I feel bad." R.R.'s subsequent messages amid that conversation underscore her desire to terminate the relationship, containing statements such as "I still need this to just go away, please, all of it. I just need[ ] it to go before [I] go crazy."

Ackell did not respond to this information by agreeing to terminate his relationship with R.R. Instead, he pressured her to accept an arrangement under which he would retain control over her, and would not delete any of the saved photos of her, until February 28. He also instructed R.R. to send him another photo of herself. This conversation, which carried over to January 28, 2014, only ended when R.R. told Ackell that her mother had found out about their interactions and was upset. And only then did Ackell tell R.R. "[y]ou won't hear from me again." But this did not prove to be true. The screenshots of R.R.'s text messages with Ackell show that on February 9, 2014 Ackell texted R.R. "[c]heck your Kik please."

*82 What followed this is not perfectly clear from the record. Yet-as the district court also recognized-a rational factfinder could have concluded that the Kik conversation between Ackell and R.R. contained in Exhibit One took place after Ackell asked R.R. to check Kik on February 9th, 2014. For one, R.R. testified that this was the case. Moreover, a number of things about the Kik messages depicted in Exhibit One provide further support for this timeline. For example, in response to Ackell's demand that R.R. send him a photo of herself, R.R. said that she would not "take any pictures like I did before, we[']ve talked about it." A rational factfinder could understand this to refer back to the January 27-28 conversation. During this conversation, Ackell also referred to his having "let [R.R.] go," which he characterized as a "[m]istake." This too could be reasonably understood as a reference to Ackell agreeing on January 28 that he would not contact R.R. again. Finally, one of the screenshots in Exhibit One shows a timestamp in the Kik application-indicating that the messages below that timestamp were from "Today @ 8:20 PM." And R.R. testified that her father told her to take screenshots of her conversations with Ackell after R.R. had told Ackell that her mother had found out about him-which is how the conversation in Exhibit Two concludes. Given all of this, a rational factfinder could well have concluded that the conversation in Exhibit One occurred after the conversation in Exhibit Two.

And it is beyond dispute that Ackell's statements both in the latter part of the conversation in Exhibit Two and throughout the conversation in Exhibit One would allow a reasonable factfinder to conclude that he had the requisite intent to violate § 2261A(2)(B). Once again, Exhibit Two shows that, after R.R. clearly communicated to Ackell that she no longer wished to be a "submissive" to him, Ackell attempted to pressure R.R. to agree to continue the relationship for another month. He also told her-in explicit terms-to send him another revealing picture of herself. The conversation in Exhibit One likewise clearly evinces an intent to "injure, harass, [or] intimidate" her. See 18 U.S.C. § 2261A. Indeed, that conversation began with Ackell ordering R.R. "Get in your room. Top off. Show me those tits." When R.R. expressed her unwillingness to do so, Ackell responded "I'll trade you. Want that??? You are MINE. You will do as told." When asked at trial what Ackell meant by "trade," R.R. explained that "[t]here are certain forums and sites online where you can trade people's photos and all of the[ir] information," and that Ackell was threatening to "trade me to somebody else who would do what he was doing to me." As Ackell requested more photos-providing explicit instructions as to what he wished for them to depict-R.R. told him "I feel uncomfortable," but Ackell again threatened to "trade" R.R. Later in the conversation, R.R. told Ackell that she felt suicidal. Ackell then sent R.R. photos of another young girl, who Ackell said he would "cage" in replacement of R.R. if he had to trade her. Ackell told R.R. that if he "caged" this girl, he would have sex with her and force her to have sex with a dog. R.R. asked how old the girl in the photos was, and Ackell told her that she was fourteen. R.R. testified that this is what led her to seek help from her father.

And so, even were we to accept Ackell's argument that he could not have formed the proper intent until January 27, 2014, he still cannot prevail. This evidence provides a substantial basis for a rational factfinder to conclude beyond a reasonable doubt that Ackell engaged in a course of conduct with the requisite intent.

2.

Ackell also argues that "the government failed to adduce sufficient evidence *83 from which a rational factfinder could conclude beyond a reasonable doubt that [R.R.] actually suffered substantial emotional distress[.]" Even setting aside that the statute requires only that Ackell's course of conduct "would be reasonably expected to cause substantial emotional distress," see 18 U.S.C. § 2261A(2)(B), this argument is impossible to square with the record. Most strikingly, R.R. testified that she considered committing suicide as a means of escaping from her relationship with Ackell. She expressed as much to Ackell during multiple conversations as well. Thus, we think it clear that a rational factfinder could have concluded beyond a reasonable doubt that R.R. suffered substantial emotional distress as the result of Ackell's conduct.

We therefore hold that the district court did not err in rejecting Ackell's Rule 29 motion.

III.

Ackell's First Amendment, instructional, and sufficiency-of-the-evidence challenges all fail. The district court's judgment is therefore affirmed .