People v. Burkman

Full title: PEOPLE OF THE STATE OF MICHIGAN, Plaintiff-Appellee, v. JACOB ALEXANDER…

Court: Supreme Court of Michigan

Date published: Jun 13, 2024

No. 164638 (Mich. Jun. 13, 2024)

Opinion

164638 164639

06-13-2024

PEOPLE OF THE STATE OF MICHIGAN, Plaintiff-Appellee, v. JACOB ALEXANDER WOHL, Defendant-Appellant. v. JOHN MACAULEY BURKMAN, Defendant-Appellant. PEOPLE OF THE STATE OF MICHIGAN, Plaintiff-Appellee,

Elizabeth T. Clement, Chief Justice

Syllabus

Argued November 9, 2023

John M. Burkman and Jacob A. Wohl were charged in the 36th district court with bribing or intimidating voters, MCL 168.932(a); conspiracy to bribe or intimidate voters, MCL 750.157a; and two counts of using a computer to commit a crime, MCL 752.796, for having designed and financed a robocall in 2020 targeting voters in Michigan who lived in areas with a significant Black population. The robocall asserted that voting by mail would result in the voter's personal information becoming part of a public database that would be used by the police to track down old warrants, by credit card companies to collect debt, and potentially by the Centers for Disease Control and Prevention (CDC) to track people for mandatory vaccines. MCL 168.932(a) prohibits a person from attempting, by means of bribery, menace, or other corrupt means or device, either directly or indirectly, to influence an elector in giving their vote, or to deter the elector from, or interrupt the elector in giving their vote at any election held in Michigan. The charges against defendants under MCL 168.932(a), which gave rise to the other charges, were brought on the alternative theories that their conduct constituted either a "menace" or an "other corrupt means or device" under that provision. The district court, Kenneth J. King, J., bound defendants over for trial in the Wayne Circuit Court following a preliminary examination after finding probable cause to believe that defendants had committed the charged offenses. Defendants moved to quash the bindovers, arguing that the robocall was not a "menace" or "other corrupt means or device" under MCL 168.932(a) and, even if it was, that MCL 168.932(a) was unconstitutional both facially and as applied to defendants. The circuit court, Margaret M. Van Houten, J., denied the motions. The Court of Appeals denied defendants' applications for leave to appeal, and defendants applied for leave to appeal in the Supreme Court. The Supreme Court, in lieu of granting leave to appeal, remanded the cases to the Court of Appeals for consideration as on leave granted. 508 Mich. 951 (2021). On remand, the cases were consolidated, and the Court of Appeals, LETICA, P.J., and RICK, J. (REDFORD, J., concurring in part and dissenting in part), affirmed. The majority concluded that the prosecutor had presented sufficient evidence to bind defendants over pursuant to a theory of "menace" under MCL 168.932(a) given that the term "menace" did not require a physical threat. In the alternative, the majority concluded that there was sufficient evidence to support the bindovers on the theory that defendants' conduct constituted attempted elector deterrence by "other corrupt means or device." The Court also held that defendants' conduct was not excluded from constitutional free-speech protections under the exception for true threats but was excluded under the speech-integral-to-criminal-conduct exception. Defendants sought leave to appeal in the Supreme Court, which granted leave to appeal and directed oral argument as to whether the Court of Appeals had properly interpreted MCL 168.932(a) and whether MCL 168.932(a) was unconstitutional on its face or as applied to defendants. 510 Mich. 968 (2022).

In an opinion by Chief Justice CLEMENT, joined by Justices BERNSTEIN, CAVANAGH, WELCH, and BOLDEN, the Supreme Court held:

The Court of Appeals erred by determining that defendants' conduct fell within the term "menace" as used in MCL 168.932(a) because that term requires the victim to reasonably believe that it is the speaker who will execute the threat, which was not the case here. However, the Court of Appeals correctly concluded that defendants' conduct fell within the statutory catchall term "other corrupt means or device." Regarding defendants' constitutional arguments, the Court of Appeals correctly held that defendants' conduct was not excluded from constitutional free-speech protections under the true-threat exception, but erred by holding that defendants' conduct was excluded from constitutional free-speech protections under the speech-integral-to-criminal-conduct exception. Defendants' conduct fell within the text of MCL 168.932(a), and that conduct is subject to constitutional free-speech protections. However, because MCL 168.932(a) posed a serious and realistic danger that it would encompass protected speech, the following limiting construction of the statute's catchall provision was adopted: when the charged conduct is solely speech and does not fall under any exceptions to constitutional free-speech protections, MCL 168.932(a)'s catchall phrase operates to proscribe that speech only if it is intentionally false speech that is related to voting requirements or procedures and is made in an attempt to deter or influence an elector's vote. The cases were remanded to the Court of Appeals for consideration of whether defendants' conduct fell within that limiting construction, and, if so, for consideration of defendants' remaining constitutional arguments.

1. Bindover is appropriate when, following a preliminary examination, the magistrate determines that there is probable cause to believe that the defendant has committed a felony. Defendants were charged with violating MCL 168.932(a), which prohibits a person from attempting, by means of bribery, menace, or other corrupt means or device, either directly or indirectly, to influence an elector in giving their vote, or to deter the elector from, or interrupt the elector in giving their vote at any election held in Michigan. Contrary to defendants' argument, "menace" does not require a physical component. Lay dictionary definitions supported an interpretation of the term that includes nonphysical threats, as did the Legislature's historical use of "menace." The caselaw defendants cited failed to establish that "menace" had acquired a different, specific meaning beyond its plain meaning. However, defendants' conduct was not encompassed by "menace" on another ground: namely, that the plain meaning of "menace" requires that the victim reasonably believe that it is the speaker who will execute the threat. When the victim has no such reasonable belief, there is no impetus for the victim to be compelled to comply with the terms of the threat.

2. Defendants' conduct was encompassed by the plain meaning of "other corrupt means or device" in MCL 168.932(a). The phrase "other corrupt means or device" operates as a catchall term, and so invokes the canon of ejusdem generis: in a statute in which general words follow a designation of particular subjects, the meaning of the general words will ordinarily be presumed to be and construed as restricted by the particular designation and as including only those things of the same kind, class, character, or nature as those specifically enumerated. Applying this canon, "bribery" and "menace"-the terms that precede the general catchall-both inform the definition of "corrupt means or device," because that term must also include "bribery" and "menace." The Court of Appeals' analysis of this phrase under the noscitur a sociis doctrine, under which a word or phrase is given meaning by its context or setting, failed to give proper weight to the word "other" in the statutory phrase "other corrupt means or device." The inclusion of "other" denotes that "other corrupt means or device" functions as a catchall, and "bribery" and "menace" thus serve as examples of corrupt means or devices. Drawing from lay and legal dictionary definitions, the statutory language "other corrupt means or device" was interpreted to mean any other depraved or immoral method or scheme of deterring or preventing someone from voting or influencing or interrupting someone in giving their vote. Under this definition, there was probable cause to believe that defendants violated MCL 168.932(a), because the prosecutor presented sufficient evidence to cause a person of ordinary prudence and caution to entertain a reasonable belief that defendants had attempted to deter Black metro-Detroiters from voting in the 2020 election by the immoral or depraved method of spreading misinformation regarding the consequences of voting and that defendants did so with racially based motives.

3. The First Amendment of the United States Constitution, made applicable to the states through the Fourteenth Amendment, provides that the government shall make no law abridging the freedom of speech, and Const 1963, art 1, § 5 provides similar free-speech protections. However, courts have held that there are several areas in which governmental restriction of speech is permissible, including true threats and speech integral to criminal conduct. 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. Excluded from this category are jests, hyperbole, or other statements whose context indicates no real possibility that violence will follow. The United States Supreme Court and Michigan courts have repeatedly referred to true threats as physical threats. Expanding the true-threat exception to encompass nonphysical threats would risk a significant chilling effect on one of the nation's most fundamental liberties, and that risk was not justified. Furthermore, a legally cognizable threat requires that the speaker, or someone within the speaker's control, be the person who executes the threat, and in these cases, the robocall stated that other third-party actors-police departments, credit card companies, and the CDC-would or likely would be performing the malevolent actions in question without any influence from or control by the purported speaker. Accordingly, the Court of Appeals' conclusion that defendants' conduct was not excluded from constitutional free-speech protections under the true-threat exception was affirmed.

4. Defendants' conduct was not excluded from constitutional free-speech protections under the speech-integral-to-criminal-conduct exception. This exception applies to speech or writing used as an integral part of conduct that violates a valid criminal statute. For this exception to apply, the speech must be integral to some conduct or scheme that is illegal in nature and independent of the speech that might be used to facilitate or accomplish the conduct or scheme. In these cases, the prosecutor did not allege that the creation of the robocall or the act of disseminating a robocall in Michigan was illegal per se, but instead alleged that the content of the robocall coupled with the defendants' intent to discourage voting was illegal, and an otherwise lawful robocall was the mechanism to broadcast defendants' message. Were the speech-integral-to-criminal-conduct exception to be applied to defendants' conduct in these cases, it would constitute the exclusion of speech from constitutional free-speech protections solely because of its content, and the Supreme Court declined to apply the speech-integral-to-criminal-conduct exception under those circumstances.

5. Defendants have not demonstrated that MCL 168.932(a) is unconstitutionally vague because the plain-language interpretations of the statutory terms provide fair notice of the conduct being proscribed and prevent arbitrary and discriminatory enforcement of the law. However, its catchall provision is unconstitutionally overbroad because it prohibits constitutionally protected conduct. The overbreadth doctrine exists to prevent the chilling of speech, but the mere fact that one can conceive of some impermissible applications of a statute is not sufficient to render it susceptible to an overbreadth challenge. Instead, a challenger must prove a realistic danger that the statute itself will significantly compromise recognized First Amendment protections of parties not before the Court. The statute's overbreadth must be substantial, not only in an absolute sense, but also relative to the statute's plainly legitimate sweep. The statutory provision at issue, MCL 168.932(a), prohibits a person from attempting "by means of bribery, menace, or other corrupt means or device" to influence, deter, or interrupt an elector's vote. The statute's catchall "or other corrupt means or device" is unconstitutionally overbroad because it poses a realistic danger of infringing constitutional free-speech protections and a substantial risk of chilling purely political speech, including statements designed to influence an elector's vote that are made via campaign speeches, rallies, door-to-door campaigning, flyers, and buttons. Although the term "corrupt" in the catchall provision limits MCL 168.932(a)'s scope, it remained likely that political speech was encompassed by "any other depraved or immoral method or scheme" in influencing or deterring votes. Therefore, the statute regulated substantially more political speech than its plainly legitimate sweep allowed. Because invalidating a statute on the basis of overbreadth should be avoided when possible, a limiting construction of MCL 168.932(a)'s catchall "other corrupt means or device" was imposed: when the charged conduct is solely speech and does not fall under any exceptions to constitutional free-speech protections, MCL 168.932(a)'s catchall phrase operates to proscribe that speech only if it is intentionally false speech that is related to voting requirements or procedures and is made in an attempt to deter or influence an elector's vote. The cases were remanded to the Court of Appeals to decide whether defendants' conduct fell within this limiting construction of MCL 168.932(a) and, if so, to resolve defendants' remaining constitutional arguments.

Affirmed in part, reversed in part, and remanded to the Court of Appeals for further proceedings.

Justice ZAHRA, joined by Justice VIVIANO, concurring in part and dissenting in part, agreed that defendants' conduct was not a "menace" within the meaning of MCL 168.932(a) but dissented from the remainder of the majority opinion, particularly the majority's limiting construction of MCL 168.932(a). Justice ZAHRA would have relied on traditional tools of statutory construction, including the canon of ejusdem generis, to conclude that MCL 168.932(a) uses "corrupt means" to proscribe: (1) conduct that is dishonest or without integrity, (2) that promises or threatens to take or not take action that would benefit or harm an individual, (3) for the purpose of causing that individual to take an action. Because the content of the robocall did not threaten or promise to take any action against or in support of its intended recipients, the dissemination of defendants' message was neither a menace nor accomplished by corrupt means under MCL 168.932(a). Further, Justice ZAHRA critiqued the majority opinion for violating the presumption of constitutional validity afforded to all statutes. Justice ZAHRA explained that the majority opinion construed MCL 168.932(a) so that it could find the statute unconstitutionally overbroad and then purport to save the statute with a limiting construction unsupported by recognized and accepted tools of interpretation. Justice ZAHRA concluded that the majority should have applied the presumption of constitutional validity to select a construction of "other corrupt means" that saved the statute from being unconstitutionally overbroad. Accordingly, he would have reversed the judgment of the Court of Appeals in total and remanded this matter to the trial court for the purpose of quashing the information with respect to both defendants. Because defendants' conduct was outside the scope of MCL 168.932(a), he would not have considered defendants' constitutional arguments. For these reasons, he concurred only in the conclusion that defendants' conduct was not a menace under MCL 168.932(a) and dissented in all other respects.

BEFORE THE ENTIRE BENCH

Justices: Brian K. Zahra David F. Viviano Richard H. Bernstein Megan K. Cavanagh Elizabeth M. Welch Kyra H. Bolden

OPINION

Elizabeth T. Clement, Chief Justice

After expressing a desire to "hi-jack" the 2020 election, defendants arranged for the dissemination of a recorded message to metro-Detroiters discouraging mail-in voting. Today, this Court is asked to determine whether defendants' conduct falls within Michigan's election-fraud statute, MCL 168.932(a), and whether the conduct nonetheless remains subject to the protection of the constitutional guarantees of free speech. We hold that defendants' conduct falls within the plain meaning of the catchall language in MCL 168.932(a), but that it is also subject to constitutional free-speech protections. We offer a limiting construction of MCL 168.932(a) to save the catchall from a facially overbroad reach under the First Amendment and remand to the Court of Appeals to decide whether defendants' conduct falls within our adopted limiting construction of MCL 168.932(a), and if so, for consideration of defendants' remaining constitutional arguments.

I. FACTUAL BACKGROUND AND PROCEDURAL HISTORY

The factual background of this case is largely undisputed. Leading up to the 2020 election, defendants created and caused the automatic dissemination of a prerecorded telephonic message (robocall) to various residents in the 313 area code. The robocall stated as follows:

Hi, this is Tamika Taylor from Project 1599, a civil rights organization founded by Jack Burman and Jacob Wohl. Mail-in voting sounds great, but did you know that if you vote by mail your personal information will be part of a public database that will be used by police departments to track down old warrants and be used by credit card companies to collect outstanding debts? The CDC¹ is even pushing to use records from mail-in voting to track people for mandatory vaccines. Don't be finessed into giving your private information to the man. Stay safe and be aware of vote by mail.

1 We assume "CDC" refers to the federal Centers for Disease Control and Prevention.

According to the prosecutor, the purpose of the robocall was to deter Black electors from voting in the 2020 general election by spreading misinformation regarding the consequences of mail-in voting. On the basis of the robocall content and other evidence uncovered by investigators, the prosecutor charged defendants with attempting to influence, deter, or interrupt electors, MCL 168.932(a); conspiracy to commit that offense, MCL 750.157a; and two counts of using a computer to commit a crime, MCL 752.796.

At defendants' preliminary examination, Jeffrey Campbell, an investigator with the Michigan Department of Attorney General, testified regarding various e-mails exchanged between defendants. In those e-mails, defendants discussed their desire to "hi-jack" the "boring" 2020 election and their later arrangements to disseminate the above-quoted robocall "to black neighborhoods in Milwaukee, Detroit, Philadelphia, Charlotte, Richmond, Atlanta, and Cleveland." After the robocall was implemented, defendants noted that the "robo [is] getting quite a bit of play on Twitter," and defendant Burkman remarked, "I love these robo calls getting angry black call backs, win or lose, the black robo calls was a great idea." In addition to Campbell's testimony, the prosecutor also presented the testimony of Khyla Craine, deputy to the Secretary of State and Chief Privacy Officer of the Michigan Department of State. Craine testified generally that the content of the robocall was false.²

2 More specifically, Craine testified that certain information contained within the Qualified Voter File was publicly accessible for a cost, but telephone numbers and e-mail addresses remained confidential. She opined that although creditors, law enforcement agencies, and the CDC could pay to access the publicly available information contained in the Qualified Voter File, she was not aware of those entities ever having done so. She also observed that other databases existed that would provide the same or similar information to those entities at a lower cost.

Following defendants' preliminary examination, the district court bound defendants over to the circuit court on all charges. There, defendants moved to quash the bindover and dismiss the charges. Defendants argued that the robocall was not a "menace" or "other corrupt means or device" under MCL 168.932(a), and-even if it was-that MCL 168.932(a) was unconstitutional both facially and as applied to defendants.³ The circuit court denied defendants' motion in an oral opinion. It reasoned that the prosecutor had presented sufficient evidence that defendants' conduct constituted a "menace" to bind defendants over and that the state's prohibition of this conduct under MCL 168.932(a) was constitutional given the state's compelling interest in protecting the right to vote and the narrow reach of MCL 168.932(a).

3 MCL 168.932(a) provides:

A person shall not attempt, by means of bribery, menace, or other corrupt means or device, either directly or indirectly, to influence an elector in giving his or her vote, or to deter the elector from, or interrupt the elector in giving his or her vote at any election held in this state.

Defendants applied for leave to appeal, and although the Court of Appeals initially denied their application, this Court later remanded the cases to the Court of Appeals as on leave granted. People v Wohl, 508 Mich. 951 (2021); People v Burkman, 508 Mich. 951 (2021). On remand, the Court of Appeals consolidated the cases and affirmed defendants' bindover in a published, split decision.

The majority agreed with the trial court that the prosecutor had presented sufficient evidence to bind defendants over pursuant to a theory of "menace" under MCL 168.932(a). People v Burkman, 341 Mich.App. 734, 751-752; 992 N.W.2d 341 (2022). In response to defendants' argument that their conduct could not constitute a "menace" because it was not a physical threat, the majority reasoned that dictionary definitions of the term did not include an accompanying physical component and neither caselaw nor the plain language of the statute indicated that the Legislature intended such a limitation. Id. Judge REDFORD dissented on this ground, opining that the caselaw did establish that "menace" requires a physical component. Id. at 766-767 (REDFORD, J., concurring in part and dissenting in part). Relying on People v Braman, 30 Mich. 460, 467-468 (1874) (opinion by GRAVES, C.J.), he also argued that, for speech to constitute a "menace," the speaker of the threat must be the person who would execute the threat. Id. at 767-768. Because the robocall did not involve a physical threat and because the robocall indicated that third parties would be the ones performing the threatened actions, Judge REDFORD would have held that bindover was erroneous under the "menace" theory. Id. at 766-768.

In the alternative, the Court collectively agreed that sufficient evidence was presented to support the bindover under the theory that defendants' conduct constituted attempted elector deterrence by "other corrupt means or device." Id. at 754-755 (majority opinion). Relying on dictionary definitions of those terms, the majority determined that the statutory terms were satisfied by the existence of probable cause to believe that "defendants intentionally disseminated a dishonest message with the depraved motive of deterring voting." Id. at 755. In so concluding, the majority also rejected defendants' argument that bindover was inappropriate because defendants had allegedly only attempted to deter electors from one method of voting and did not discourage those electors from casting their votes in person. Id. at 755-756. The majority reasoned that the robocall could have prevented electors who were unable to vote in person or unwilling to do so (given that the robocall occurred during the 2020 pandemic and in-person voting involved a serious risk to a voter's health) from casting their votes entirely. Id.

Next, the Court considered whether MCL 168.932(a) was an unconstitutional infringement of federal and state free-speech protections. Although defendants characterized their arguments as both facial and as-applied challenges, the Court chose to treat their arguments only as an as-applied challenge because defendants focused on the specific application of the statute to the facts at hand. Id. at 760. Under this characterization, the Court determined that defendants' conduct was not excluded from constitutional free-speech protections under the exception for true threats. Id. at 761. It reasoned that both the United States Supreme Court and this Court have described this exception as applying only to unlawful violence, and while the robocall "warned of harm to the listener's freedom, financial security, and bodily autonomy," this did not constitute a threat of unlawful violence. Id. at 761-762.

However, the Court held that defendants' conduct was excluded from constitutional free-speech protections under the speech-integral-to-criminal-conduct exception. Id. at 763-764. Under Giboney v Empire Storage &Ice Co, 336 U.S. 490, 498; 69 S.Ct. 684; 93 L.Ed.2d 834 (1949), that exception holds that "speech or writing used as an integral part of conduct in violation of a valid criminal statute" is not entitled to constitutional protection. As applied here, the Court reasoned that defendants' speech "was an integral part of conduct criminalized by MCL 168.932(a) and should not be constitutionally protected merely because the conduct was 'carried out by means of language.'" Id. at 764, quoting Giboney, 336 U.S. at 502. Having concluded that the trial court properly found probable cause to believe that defendants violated the statute and that defendants' conduct was excluded from constitutional free-speech protections, the Court affirmed the bindover. Id. at 739.

Defendants subsequently applied for leave to appeal in this Court. This Court granted leave to appeal and directed oral argument as to "(1) whether the Court of Appeals properly interpreted MCL 168.932(a); and (2) whether MCL 168.932(a) is unconstitutional on its face or as applied to the defendants." People v Burkman, 510 Mich. 968, 968-969 (2022).

II. WHETHER THE CONDUCT FALLS UNDER MCL 168.932(a)

Defendants challenge the district court's bindover decision. Bindover is appropriate where, following a preliminary examination, the magistrate determines that there is probable cause to believe that the defendant has committed a felony. MCL 766.13; People v Plunkett, 485 Mich. 50, 57; 780 N.W.2d 280 (2010). "Probable cause requires a quantum of evidence sufficient to cause a person of ordinary prudence and caution to conscientiously entertain a reasonable belief of the accused's guilt." People v Yost, 468 Mich. 122, 126; 659 N.W.2d 604 (2003) (quotation marks and citation omitted).⁴ Accordingly, to determine whether the district court's decision to bind defendants over was erroneous, this Court must determine whether there was probable cause to believe that defendants violated MCL 168.932(a), which requires an interpretation of that statutory provision.

4 This Court reviews district court decisions regarding bindover for an abuse of discretion, which occurs when the district court's decision falls outside the range of principled outcomes. People v Anderson, 501 Mich. 175, 182; 912 N.W.2d 503 (2018). However, to the extent that a bindover decision encompasses questions of law and statutory interpretation, this Court reviews those issues de novo. Id.

This Court's primary goal in interpreting a statute is to ascertain and give effect to the Legislature's intent. People v Gardner, 482 Mich. 41, 50; 753 N.W.2d 78 (2008). "If the statute's language is clear and unambiguous, we assume that the Legislature intended its plain meaning and we enforce the statute as written." People v Weeder, 469 Mich. 493, 497; 674 N.W.2d 372 (2004). In so doing, we assign each word and phrase its plain and ordinary meaning within the context of the statute, unless the word or phrase is a "term of art" with a "unique legal meaning." People v Thompson, 477 Mich. 146, 151-152; 730 N.W.2d 708 (2007); see also MCL 8.3a.

The statutory provision at hand, MCL 168.932(a), provides that "[a] person shall not attempt, by means of bribery, menace, or other corrupt means or device, either directly or indirectly, to influence an elector in giving his or her vote, or to deter the elector from, or interrupt the elector in giving his or her vote at any election held in this state." The charges against defendants were brought under alternative theories that their conduct constituted a "menace" or an "other corrupt means or device." We discuss each in turn.

A. "MENACE"

Defendants argue that the charges cannot be substantiated under a "menace" theory because "menace" requires the threat of physical force, which was not present here. Absent a statutory definition of the term "menace," and "[b]ecause our goal is to glean legislative intent from the plain meaning of statutory language, the dictionary is our first point of reference to determine the term's significance[.]" In re Erwin Estate, 503 Mich. 1, 9-10; 921 N.W.2d 308 (2018) (citations omitted). "Menace" has been defined as "a show of intention to inflict harm" or "one that represents a threat," Merriam-Webster's Collegiate Dictionary (11th ed); "something that threatens to cause evil, harm, etc.; threat," "a person whose actions or ideas are considered dangerous or harmful," or "to act as a threat; be threatening," Random House Webster's College Dictionary (1995); and "a threat" or "[t]he act of threatening," The American Heritage Dictionary (2d ed). While these definitions of "menace" undoubtedly include a threat of physical force, the definitions alone do not exclude a threat of nonphysical force.⁵

5 This remains true even if one chooses to consult only historical dictionaries in light of the mid-1800s enactment of the statute's predecessor, 1846 RS, ch 19, § 2. See, e.g., Johnson, A Dictionary of the English Language (1773) (defining "menace" as "[t]hreat"); Webster, An American Dictionary of the English Language (1828) (defining "menace" as "[a] threat or threatening; the declaration or show of a disposition or determination to inflict an evil; used of persons"); 2 Bouvier's Law Dictionary and Concise Encyclopedia (1914) (defining "menace" as "[a] threat; a declaration of an intention to cause evil to happen to another" and explaining that "[t]he word menace is not restricted to threats of violence to person and property nor to threats of accusing a person of crime; it includes a threat to accuse one of immoral conduct").

Defendants acknowledge these lay definitions of the term "menace," but they argue that the term "menace" has acquired a specialized meaning in Michigan caselaw as solely a physical threat. However, the caselaw cited by defendants fails to support this proposition. For example, in People v Doud, 223 Mich. 120, 129-131; 193 N.W. 884 (1923), this Court was presented with, and rejected, a defendant's claim that his conviction for felonious assault was erroneous because the prosecutor had not established his intent by extrinsic proof. This Court reasoned that the defendant, who had pointed a firearm at another person and threatened to shoot him if he did not leave the defendant's property, had "a declared purpose to injure, accompanied by acts manifesting a corporal hurt to one immediately menaced thereby," which sufficed to establish intent. Id. at 123. While this Court in Doud referred to a threat of physical harm as "menac[ing]," Doud does not answer the question whether, or stand for the principle that, "menace" includes only physical threats.

Similarly, in People v Reeves, 458 Mich. 236, 245; 580 N.W.2d 433 (1998), when considering whether the prosecutor had proved beyond a reasonable doubt all the elements of assault with intent to rob while unarmed, this Court summarized that the defendant had "approached [a delivery driver] with his hand inside [a] bag in a menacing fashion" and asked whether his job or his life was worth more. This Court thus referred to "menacing" in the context of a physical threat, but it did not define the term or decide whether the term excludes nonphysical threats. Further, in neither Reeves nor Doud was this Court considering a statute that used the term "menace." At most, these cases support the proposition that "menace" can include a threat of physical harm; they do not demonstrate that "menace" has acquired a specific meaning in Michigan caselaw that excludes nonphysical threats.

The Legislature's use of the term "menace" in other statutes also weighs against defendants' claim that the term has acquired a definition of only physical threats. Those statutes either refer to "menace" in the context of nonphysical threats or add qualifiers to specify that "menace" is meant in a physical sense-qualifiers that would not be needed if "menace" alone denoted a solely physical threat. See, e.g., MCL 29.23 ("The court may order the removal of occupancies of a building and the discontinuance of any use of the building that constitutes a fire hazard or menace to human life ...."); MCL 29.5 (providing that hazardous materials shall not be "handled or disposed of in a manner and by a method as not to constitute a fire hazard or a menace to the public peace, health, or safety, or endanger or cause loss, injury, or damage to persons or property"); MCL 28.721a (describing the sex offenders registration act and explaining that sex offenders "pose[] a potential serious menace and danger to the health, safety, morals, and welfare of the people"); MCL 791.233(1)(a) (describing the standards for parole as including that "the prisoner will not become a menace to society or to the public safety"). Defendants do not identify any caselaw resulting from these statutes that restricts the term "menace" to physical threats.

In sum, we are not convinced by defendants' argument that "menace" requires a physical component.⁶ The lay dictionary definitions, from which we draw the plain meaning of "menace," all support an interpretation of the term open to nonphysical threats, as does the Legislature's historical use of "menace." The caselaw cited by defendants fails to establish that "menace" has acquired a different, specific meaning beyond its plain meaning.⁷

6 Defendants also argue that other Michigan statutes have referred to both threats and menace and that pursuant to the canon against surplusage, those terms must have unique meanings. See Baker v Gen Motors Corp, 409 Mich. 639, 665; 297 N.W.2d 387 (1980) ("Every word of a statute should be given meaning and no word should be treated as surplusage ...."). Defendants cite as an example People v Lyons, 197 Mich. 64, 66; 163 N.W. 484 (1917), which discussed Michigan's former pandering statute. That statute prohibited a person from, among other things," 'tak[ing] or detain[ing] a female with the intent to compel her by force, threats, menace or duress to marry him ....'" Id., quoting 1915 CL 15494. But "[t]he canon against surplusage is not an absolute rule," and legislatures are known to occasionally employ redundancies in drafting statutes. Marx v Gen Revenue Corp, 568 U.S. 371, 385; 133 S.Ct. 1166; 185 L.Ed.2d 242 (2013). Legislatures may "engage[] in the retrograde practice of stringing out synonyms and near-synonyms" in a "belt-and-suspenders approach" to ensure the inclusion of all relevant conduct, and so the application of the canon against surplusage must always "be applied with judgment and discretion, and with careful regard to context." Scalia & Garner, Reading Law: The Interpretation of Legal Texts (St. Paul: Thomson/West, 2012), pp 176-177, 179. See also Posner, Statutory Interpretation-in the Classroom and in the Courtroom, 50 U Chi L Rev 800, 812 (1983). When considering the entirety of the pandering statute at issue in Lyons, judgment and discretion lead one to conclude that the drafters were engaging in such a belt-and-suspenders approach rather than choosing each word carefully for its unique and independent meaning:

Any person who shall procure a female inmate for a house of prostitution; or who shall induce, persuade, encourage, inveigle or entice a female person to become a prostitute; or who by promises, threats, violence or by any device or scheme, shall cause, induce, persuade, encourage, take, place, harbor, inveigle or entice a female person to become an inmate of a house of prostitution or assignation place, or any place where prostitution is practiced, encouraged or allowed; or any person who shall, by promises, threats, violence, or by any device or scheme, cause, induce, persuade, encourage, inveigle or entice an inmate of a house of prostitution or place of assignation to remain therein as such inmate; or any person who by promises, threats, violence, by any device or scheme, by fraud or artifice, or by duress of person or goods, or by abuse of any position of confidence or authority, or having legal charge, shall take, place, harbor, inveigle, entice, persuade, encourage or procure any female person to enter any place within this state in which prostitution is practiced, encouraged or allowed, for the purpose of prostitution, or to inveigle, entice, persuade, encourage or procure any female person to come into this state or to leave this state for the purpose of prostitution; or who takes or detains a female with the intent to compel her by force, threats, menace or duress to marry him or to marry any other person or to be defiled; or upon the pretense of marriage takes or detains a female person for the purpose of sexual intercourse; or who shall receive or give or agree to receive or give any money or thing of value for procuring or attempting to procure any female person to become a prostitute or to come into this state or leave this state for the purpose of prostitution, shall be guilty of pandering ....[1915 CL 15494.]

The statute's drafters repeatedly included a list of synonyms or near synonyms as operative verbs, and the use of "menace" and "threat" in the latter portion of the statute does not appear to be an exception to this practice. Moreover, the Legislature did not use both the term "threat" and the term "menace" in the statute at hand. Defendants also pursue this argument in the context of other states' election-fraud statutes that list both "threat" and "menace" in their prohibitory phrases. But defendants fail to produce any caselaw wherein those states reasonably applied the canon against surplusage to arrive at distinct and unique meanings for "threat" and "menace."

7 Given this conclusion, we need not address the prosecutor's argument that the threat of mandatory vaccination presents a physical threat.

While we disagree with defendants' contention that their conduct does not constitute "menace" by virtue of presenting a nonphysical threat, we nonetheless conclude that defendants' conduct is not encompassed by "menace" on another ground. We agree with Judge REDFORD, dissenting in part below, that "menace" requires that the victim reasonably believe that the speaker will execute the threat. Where the victim has no such reasonable belief, there is no impetus for the victim to be compelled to comply with the terms of the threat. This limitation is supported by the dictionary definitions of "menace" and "threat," which indicate that the person making the threat must be the one who intends to inflict the injury. See, e.g., 2 Bouvier's Law Dictionary and Concise Encyclopedia (1914) (defining "menace" in part as "a declaration of an intention to cause evil to happen to another"); Black, A Dictionary of Law (1891) (defining "menace" in part as "the declaration or show of a disposition or determination to inflict an evil or injury upon another"); Black's Law Dictionary (11th ed) (defining "threat" in part as "[a] communicated intent to inflict harm or loss on another or another's property"). As the United States Court of Appeals for the Sixth Circuit remarked in a discussion of the true-threat exception to the First Amendment, "a person who informs someone that he or she is in danger from a third party has not made a threat, even if the statement produces fear." New York ex rel Spitzer v Operation Rescue Nat'l, 273 F.3d 184, 196 (CA 2, 2001). See also Braman, 30 Mich. at 467-468 (opinion by GRAVES, C.J.) (in the context of an extortion charge, reasoning that "the accusation menaced is to be one threatened to come from the party threatening, and not exclusively from some other").⁸

8 Although People v Watson, 307 Mich. 378, 381-382; 11 N.W.2d 926 (1943), identified the opinion of Chief Justice GRAVES as a dissenting opinion, Braman actually involved an equally divided court, with Chief Justice GRAVES joined by Justice CAMPBELL on one opinion, and Justice COOLEY joined by Justice CHRISTIANCY on the other.

Here, defendants' robocall message stated that, if an elector cast their vote by mailin ballot, the elector's information would be used by police departments to track down old warrants, by credit card companies to collect outstanding debts, and potentially by the Centers for Disease Control and Prevention (CDC) to track persons for mandatory vaccines. Thus, the parties who would perform the threatened action would be police departments, credit card companies, and the CDC-not defendants or the speaker on the robocall. Further, there is no implication that defendants have any sort of control over those entities such that the robocall message could be considered an indirect threat-i.e., that defendants could have instructed those entities to follow through on the threatened actions. Accordingly, the robocall does not constitute a "menace."

In sum, while we agree with the prosecutor that "menace," as used in MCL 168.932(a), is not limited to solely physical threats, we hold that defendants' conduct here does not constitute a "menace" because "menace" requires that the person making the threat will be the party who performs the threat or causes the threat to be performed.

B. "OTHER CORRUPT MEANS OR DEVICE"

The prosecutor also brought charges against defendants on a theory that their conduct constitutes an "other corrupt means or device" by which defendants attempted, "either directly or indirectly, to influence an elector in giving his or her vote, or to deter the elector from, or interrupt the elector in giving his or her vote at any election held in this state." MCL 168.932(a).

Like "menace," "corrupt means or device" is not defined by the statute. However, the phrase "other corrupt means or device" operates as a catchall term, and so invokes the canon of ejusdem generis. Ejusdem generis provides that "in a statute in which general words follow a designation of particular subjects, the meaning of the general words will ordinarily be presumed to be and construed as restricted by the particular designation and as including only those things of the same kind, class, character or nature as those specifically enumerated." People v Smith, 393 Mich. 432, 436; 225 N.W.2d 165 (1975). Applying this canon here, "bribery" and "menace"-the terms that precede the general catchall-both inform the definition of "corrupt means or device[s]," because that term must also include "bribery" and "menace."

The Court of Appeals erroneously rejected this analysis as a noscitur a sociis analysis. "Contextual understanding of statutes is generally grounded in the doctrine of noscitur a sociis: '[i]t is known from its associates[.]' This doctrine stands for the principle that a word or phrase is given meaning by its context or setting." Koontz v Ameritech Servs, Inc, 466 Mich. 304, 318; 645 N.W.2d 34 (2002) (quotation marks, citations, and brackets omitted). Under noscitur a sociis, "[w]hen several nouns or verbs or adjectives or adverbs-any words-are associated in a context suggesting that the words have something in common, they should be assigned a permissible meaning that makes them similar." Reading Law, p 195. The Court of Appeals majority reasoned as follows:

We decline the request to apply noscitur a sociis in order to achieve defendants' goal of equating "corrupt means or device" with menace or bribery. To do so, we would fail to give effect to every word, phrase and clause in MCL 168.932(a) and render "corrupt means or device" surplusage or nugatory. [Burkman, 341 Mich.App. at 753.]

The quoted analysis fails to give proper weight to the word "other" in the statutory phrase "other corrupt means or device." Had the statute merely said "by means of bribery, menace, or corrupt means or device," the majority's analysis might have been viable, given that the rule against surplusage generally advises that each item in a list should be interpreted to have a unique meaning lest it be rendered superfluous or nugatory. Baker, 409 Mich. at 665.⁹ However, the statute instead says, "by means of bribery, menace, or other corrupt means or device." (Emphasis added.) The inclusion of "other" denotes that "other corrupt means or device" functions as a catchall, and, as described above, "bribery" and "menace" thus serve as examples of "corrupt means or device[s]."

9 But see our discussion in note 6 of this opinion regarding the judicial discretion necessary in applying the rule against surplusage.

Keeping in mind that the definition of "corrupt means" must include "bribery" and "menace," we turn to the dictionary definitions of the relevant terms for guidance. See Erwin, 503 Mich. at 9-10. The term "corrupt" is generally defined as involving "[d]epravity, perversion, or taint; an impairment of integrity, virtue, or moral principle" or "[t]he act of doing something with an intent to give some advantage inconsistent with official duty and the rights of others[.]" Black's Law Dictionary (7th ed) (defining "corruption"). See also The American Heritage Dictionary (2d ed) ("[m]arked by immorality and perversion; depraved"); Merriam-Webster's Collegiate Dictionary (11th ed) ("morally degenerate and perverted: DEPRAVED" or "characterized by improper conduct (as bribery or the selling of favors)"). These definitions focusing on amorality are consistent with and encompass "bribery" and "menace." They are also consistent with Michigan caselaw defining "corrupt" for purposes of determining whether a person committed the common-law crime of misconduct in office, which penalizes "corrupt behavior by an officer in the exercise of the duties of his office while acting under color of his office." People v Coutu, 459 Mich. 348, 354; 589 N.W.2d 458 (1999) (quotation marks and citation omitted). In that context, Michigan caselaw interprets "corrupt intent" as acting "with a sense of depravity, perversion or taint." People v Perkins, 468 Mich. 448, 456; 662 N.W.2d 727 (2003) (quotation marks and citation omitted). See also People v Waterstone, 296 Mich.App. 121, 138; 818 N.W.2d 432 (2012) (interpreting "corrupt behavior" in the misconduct-in-office context as "intentional, purposeful, deliberate, and knowing wrongful behavior").¹⁰

10 The dissent argues that these definitions of "corrupt" would render the statute's inclusion of "menace" and "bribery" meaningless surplusage because the inclusion of "corrupt means or device" alone would be sufficient to prohibit "menace" and "bribery." But the Legislature's inclusion of "menace" and "bribery" is useful insofar as these terms guide the courts to define "corrupt" in this context of moral depravity rather than, for example, in the context of unreliable data. And, as discussed above, the Legislature's use of the phrase "other corrupt means or device" (emphasis added) indicates that "menace" and "bribery" must be encompassed by "corrupt means or device."

Finally, in interpreting "means or device," we note the following relevant definitions: "[s]omething that helps to attain an end; an instrument; a cause," or "[a] scheme or trick to deceive; a stratagem or artifice, as in the law relating to fraud," Black's Law Dictionary (7th ed) (defining "means" and "device," respectively); or "an agency, instrument, or method used to attain an end," or "a plan, scheme, or procedure for effecting a purpose," Random House Webster's College Dictionary (1995) (same).

Drawing from these dictionary definitions, we interpret the statutory language "other corrupt means or device" as any other depraved or immoral method or scheme of deterring or preventing someone from voting or influencing or interrupting someone in giving their vote.¹¹ Under this definition, we conclude that there is probable cause to believe that defendants violated MCL 168.932(a). Plunkett, 485 Mich. at 57. The prosecutor presented sufficient evidence to cause a person of ordinary prudence and caution to entertain a reasonable belief that defendants attempted to deter Black metro-Detroiters from voting in the 2020 election by the immoral or depraved method of spreading misinformation regarding the consequences of voting and that defendants did so with racially based motives. See Yost, 468 Mich. at 126. Defendants discussed their desire to "hi-jack this boring election" and arranged for the distribution of a robocall specifically to "black neighborhoods" with the call stating that the consequences of mail-in voting would include voter information being used by police departments to effectuate old warrants, by credit card companies to collect outstanding debts, and (potentially) by the CDC to support mandatory vaccination efforts-information that Deputy Legal Director Craine testified was untrue. Defendants expressed pleasure in receiving "angry black call backs" regarding the message, and they initially denied their involvement when contacted regarding the robocall. Given the targeted nature of the robocall, defendants' e-mails, and the content of the robocall, one could reasonably believe that the robocall was a depraved attempt to deter Black electors from voting in the 2020 election.¹² To the extent that questions may remain regarding defendants' intentions in distributing the robocall and whether the information in the robocall was false, we would emphasize that a bindover requires only that an ordinary person would entertain a reasonable belief of guilt, and that standard has been met here. Id.

11 The dissent argues that we have adopted an "exceedingly broad construction" of the statutory language "other corrupt means or device" and instead offers an exceedingly narrow construction of the same. The dissent's proposed definition of "other corrupt means or device" is so specific that it would likely not encompass any means or device other than the specifically listed terms of menace and bribery-and, accordingly, would not encompass actions that the Legislature intended to prohibit by MCL 168.932(a). The Legislature's use of catchall terms in a statute is intended to save it "from spelling out in advance every contingency in which a statute could apply," 2A Singer & Singer, Sutherland Statutory Construction (7th ed), § 47:17, p 386, and too narrowly construing a catchall term eliminates this intended benefit. For example, under the dissent's narrow construction of the catchall term, it would not be a violation of MCL 168.932(a) for a person to engage in a robocall campaign that falsely informs voters that the polls will remain open three hours later because such action does not promise or threaten to take or not take action that would benefit or harm an individual. Further, adopting such a narrow construction renders the catchall surplusage because the catchall is fully encompassed by the specific terms of the statute, and this Court is inclined to avoid such constructions. See People v Rea, 500 Mich. 422, 428; 902 N.W.2d 362 (2017) (explaining that this Court should avoid a statutory interpretation that would render any part of a statute surplusage).

12 Defendants also argue that their conduct does not fall under the terms of the statute because the robocall only allegedly attempted to deter electors from casting their ballots by mail and electors remained able to exercise their right to vote in person. However, MCL 168.932(a) prohibits more than successfully preventing an elector from casting a vote at all; it also prohibits attempts to deter, influence, or interrupt an elector from casting a vote. From the evidence admitted thus far, one could reasonably conclude that defendants attempted to deter certain electors (i.e., Black electors) from casting their votes at all, especially given that the 2020 election occurred during the COVID-19 pandemic and that some electors, whether because of the COVID-19 pandemic or other reasons, were only able to vote by mail. That those who were affected by the robocall may have voted in person regardless of the robocall does not negate defendants' alleged attempt to deter those electors.

We conclude that defendants' conduct is encompassed by the plain-language meaning of "other corrupt means or device" in MCL 168.932(a) and therefore affirm the Court of Appeals judgment in that regard.¹³

13 In so holding, we also reject defendants' request to apply the rule of lenity to grant them their requested relief. The rule of lenity requires that courts mitigate punishment when a criminal statute is unclear, but because MCL 168.932(a) is unambiguous, that rule is inapplicable here. See People v Denio, 454 Mich. 691, 699; 564 N.W.2d 13 (1997).

III. WHETHER THE CONDUCT IS SUBJECT TO CONSTITUTIONAL FREE-SPEECH PROTECTIONS

Having concluded that defendants' conduct is encompassed by MCL 168.932(a), we must next consider defendants' constitutional challenges to the statute. Statutes are presumed to be constitutional, and the party challenging a statute has the burden of showing the contrary. People v Sands, 261 Mich.App. 158, 160; 680 N.W.2d 500 (2004). A facial challenge alleges that a statute is unconstitutional "on its face," meaning that, in general, "the challenger must establish that no set of circumstances exists under which the [statute] would be valid." United States v Salerno, 481 U.S. 739, 745; 107 S.Ct. 2095; 95 L.Ed.2d 697 (1987); see also Bonner v City of Brighton, 495 Mich. 209, 223 n 26; 848 N.W.2d 380 (2014). However, in the First Amendment context, a facial challenge may be sufficient if it establishes that the statute prohibits constitutionally protected speech or conduct and is thus overbroad. See Grayned v City of Rockford, 408 U.S. 104, 108; 92 S.Ct. 2294; 33 L.Ed.2d 222 (1972). An as-applied challenge, on the other hand, "alleges a present infringement or denial of a specific right or of a particular injury in process of actual execution of government action." Bonner, 495 Mich. at 223 n 27 (quotation marks and citation omitted).

Here, defendants argue that the government's prohibition of conduct through MCL 168.932(a) violates federal and state constitutional rights to free speech.¹⁴ The First Amendment of the United States Constitution, made applicable to the states through the Fourteenth Amendment, provides that the government shall "make no law . . . abridging the freedom of speech ...." U.S. Const, Am I. Likewise, the Michigan Constitution provides that "[e]very person may freely speak, write, express and publish his views on all subjects, being responsible for the abuse of such right; and no law shall be enacted to restrain or abridge the liberty of speech or of the press." Const 1963, art 1, § 5. "If there is a bedrock principle underlying the First Amendment, it is that the government may not prohibit the expression of an idea simply because society finds the idea itself offensive or disagreeable." Texas v Johnson, 491 U.S. 397, 414; 109 S.Ct. 2533; 105 L.Ed.2d 342 (1989). See also Operation Rescue, 273 F.3d at 196 ("It is worth reinforcing that we must tolerate even views that upset our most heartfelt and deeply held convictions.")

14 Defendants raise both facial and as-applied challenges to MCL 168.932(a). Although the Court of Appeals below characterized defendant's arguments as raising only as-applied challenges, we agree with defendants that they have sufficiently presented facial arguments as to vagueness and overbreadth in their applications for leave to appeal and their briefs before both this Court and the Court of Appeals. We address defendants' as-applied challenges regarding whether their conduct falls within constitutional free-speech protections in Part III(A) and defendants' facial challenges regarding vagueness and overbreadth in Part III(B).

A law that restricts or proscribes speech or conduct on the basis of the message or idea it conveys is considered content-based. Generally, content-based restrictions on speech "are presumptively unconstitutional and may be justified only if the government proves that they are narrowly tailored to serve compelling state interests"-i.e., that they survive strict scrutiny. Reed v Town of Gilbert, 576 U.S. 155, 163; 135 S.Ct. 2218; 192 L.Ed.2d 236 (2015). However, the United States Supreme Court and this Court have recognized several specific areas where restriction of speech is permissible and not subject to this framework, including obscenity, defamation, fraud, incitement, true threats, and speech integral to criminal conduct. United States v Stevens, 559 U.S. 460, 468-469; 130 S.Ct. 1577; 176 L.Ed.2d 435 (2010).

A. WHETHER DEFENDANTS' CONDUCT IS SUBJECT TO AN EXCEPTION TO CONSTITUTIONAL FREE-SPEECH PROTECTIONS

The prosecutor asserts that defendants' conduct falls under the true-threat and speech-integral-to-criminal-conduct exceptions and therefore is not entitled to constitutional free-speech protections. The Court of Appeals held that the true-threat exception did not apply in this case, but that the speech-integral-to-criminal-conduct exception did apply. We discuss each in turn.

1. THE TRUE-THREAT EXCEPTION

As stated, true threats are historically excepted from constitutional free-speech protections. Id." '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." Virginia v Black, 538 U.S. 343, 359; 123 S.Ct. 1536; 155 L.Ed.2d 535 (2003). See also TM v MZ, 326 Mich.App. 227, 239; 926 N.W.2d 900 (2018). Excluded from this category are jests, hyperbole, or other statements whose context indicates no real possibility that violence will follow. Counterman v Colorado, 600 U.S. 66, 74; 143 S.Ct. 2106; 216 L.Ed.2d 775 (2023). The exclusion of true threats from constitutional protections is intended to "protect[] individuals from the fear of violence" and "from the disruption fear engenders," while also protecting people "from the possibility that the threatened violence will occur." Black, 538 U.S. at 359 (quotation marks and citation omitted).

Defendants argue that this exception includes only physical threats, and so the nonphysical threats of their robocall are not encompassed within it. The prosecutor argues that this exception has never been limited to physical threats and that the justification for the exception applies with equal force to nonphysical threats.¹⁵

15 The prosecutor argues that the robocall still fulfills the true-threat exception even if it is limited to physical threats because the robocall's "subtle threat of mandatory vaccination could be viewed as a threat that a battery would be committed on the listener if he or she participated in mail-in voting." The robocall stated that "[t]he CDC is even pushing to use records from mail-in voting to track people for mandatory vaccines." This statement is significantly less definite than the statements related to police departments' and credit card companies' use of the mail-in voting information. For the statement to be realized, vaccination would have to be required, and the CDC would have to be successful at "pushing to use records from mail-in voting to track people" for those mandatory vaccinations. Because of the extent of these conditions-and because this statement would also not qualify as a "true threat" given that a third party would be effectuating the threat, as discussed in detail below-we decline to conclude that this statement fulfills the true-threat exception.

As the Court of Appeals noted, the United States Supreme Court has repeatedly referred to true threats as physical threats. See Black, 538 U.S. at 359 (noting that true threats include "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") (emphasis added); id. at 360 ("Intimidation 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 victims in fear of bodily harm or death.") (emphasis added); id. (referring to the "fear of violence") (emphasis added; quotation marks and citation omitted); Counterman, 600 U.S. at 74 ("True threats are 'serious expression[s]' conveying that a speaker means to 'commit an act of unlawful violence.' ") (quotation marks and citation omitted; alteration in original); id. at 80 (referring to "true threats of violence"); RAV v St Paul, 505 U.S. 377, 388; 112 S.Ct. 2538; 120 L.Ed.2d 305 (1992) (referring to "threats of violence"). Our state courts have done similarly. See, e.g., People v Johnson, 340 Mich.App. 531, 547; 986 N.W.2d 672 (2022) (holding that, to satisfy the true-threat exception, the jury should have been instructed "that the prosecution was required to prove beyond a reasonable doubt that defendant meant to express a serious intent to kill or injure [the victim]"); People v Byczek, 337 Mich.App. 173, 190-191; 976 N.W.2d 7 (2021); TM v MZ (On Remand), 326 Mich.App. 227, 239; 926 N.W.2d 900 (2018).

The prosecutor is correct that this caselaw does not definitively exclude nonphysical threats from the constitutional exception and that nonphysical threats may engender fear in victims just as physical threats do. See Black, 538 U.S. at 359 (noting that the purpose of the true-threat exception is to protect persons from the fear and actual imposition of violence). But, with one apparent exception, no court to date has extended the true-threat exception to encompass nonphysical threats, and we decline to do so today.¹⁶ Such an expansion of the true-threat exception would risk a significant chilling effect on one of this nation's most fundamental liberties. Expansion of the true-threat doctrine would create "substantial costs in discouraging the 'uninhibited, robust, and wide-open' debate that the First Amendment is intended to protect." Rogers v United States, 422 U.S. 35, 47-48; 95 S.Ct. 2091; 45 L.Ed.2d 1 (1975) (Marshall, J., concurring) (citation omitted). More specifically, "[t]he speaker's fear of mistaking whether a statement is a threat; his fear of the legal system getting that judgment wrong; his fear, in any event, of incurring legal costs-all those may lead him to swallow words that are in fact not true threats." Counterman, 600 U.S. at 78. We are not convinced at this time that this risk is justified such that nonphysical threats should be excluded from constitutional protection.

16 At least, no court has done so whose decision we find persuasive. The prosecutor cites Nat'l Coalition on Black Civic Participation v Wohl, 498 F.Supp.3d 457 (SDNY, 2020), in support for the expansion of the true-threat exception to nonphysical threats. In Nat'l Coalition, a group of plaintiffs sued the defendants here-and others-regarding the same robocall at issue here under claims premised in federal law. That decision granted the plaintiffs' request for injunctive relief requiring the defendants to send a curative message to all recipients of the robocall communicating that court's findings regarding the call (i.e., that it contained misinformation) and otherwise restraining defendants from engaging in or causing anyone else to engage in robocalls or similar forms of communication without the written express consent of each recipient or approval of the court. In reaching this disposition, the court concluded that the true-threat exception included nonphysical harm. Id. at 480. The court reasoned that Black and other caselaw did not affirmatively exclude nonphysical threats as true threats and noted that the United States Court of Appeals for the Second Circuit had recently approvingly cited a law review article's proposition that a threat of nonphysical harm could constitute a true threat if it was likely to cause substantial emotional disturbance. Id. at 479-480. Nat'l Coalition, however, never received appellate review. And "[a] decision of a federal district court judge is not binding precedent in either a different judicial district, the same judicial district, or even upon the same judge in a different case" in federal court, Camreta v Greene, 563 U.S. 692, 708 n 7; 131 S.Ct. 2020; 179 L.Ed.2d 1118 (2011) (quotation marks and citation omitted), nor is it binding precedent in Michigan courts, Linsell v Applied Handling, Inc, 266 Mich.App. 1, 16; 697 N.W.2d 913 (2005). Further, we do not find the Nat'l Coalition decision persuasive. While the decision offers thorough reasoning regarding why defendants' robocall may have engendered reasonable fear and apprehension by the recipients of the call, see Nat'l Coalition, 498 F.Supp. at 483-484, it fails to engage in any analysis regarding the United States Supreme Court's repeated advisements of caution as to the expansion of First Amendment exceptions, as we discuss later in this opinion. Further, the Second Circuit case it cites, United States v Turner, 720 F.3d 411, 420 (CA 2, 2013), concerned that circuit's objective test of "whether an ordinary, reasonable recipient who is familiar with the context of the [communication] would interpret it as a threat of injury." (Quotations marks and citation omitted; alteration in original.) And it appears to us that Turner's citation of the law review article in question was meant to establish that the alleged threat is surveyed through an objective lens and not by an analysis of the speaker's subjective intent; there is little indication that Turner's citation of the law review article was instead meant to establish that nonphysical threats are included in the true-threat exception. Id. at 420 & n 4. Further, the threat at issue in Turner was a physical one-the defendant published a blog stating that three federal judges should be killed, id. at 421-422-and so, even if the Second Circuit did intend such a meaning from that citation, it is merely dictum, Roberts v Auto-Owners Ins Co, 422 Mich. 594, 596; 374 N.W.2d 905 (1985).

Further, even if we were so inclined, the prosecutor's argument that defendants' conduct is a true threat would still fail for the same reason that the prosecutor's statutory argument regarding "menace" fails: a legally cognizable threat requires that the speaker, or someone within the speaker's control, be the person who executes the threat. See Operation Rescue, 273 F.3d at 196; Braman, 30 Mich. at 467-468 (opinion by GRAVES, C.J.). And here, the robocall stated that other third-party actors-police departments, credit card companies, and the CDC-would or likely would be performing the malevolent actions in question without any influence from or control by the purported speaker. Accordingly, we affirm the Court of Appeals' conclusion that defendants' conduct is not excluded from constitutional free-speech protections under the true-threat exception.

2. THE SPEECH-INTEGRAL-TO-CRIMINAL-CONDUCT EXCEPTION

Next, the prosecutor argues-and the Court of Appeals agreed-that defendants' conduct is excluded from constitutional free-speech protections under the speech-integral-to-criminal-conduct exception. The United States Supreme Court first recognized this exception in Giboney, 336 U.S. 490. In that case, a union of Kansas City area ice peddlers sought to obtain agreements from wholesale ice distributors in the area that the distributors would not sell to nonunion ice peddlers. Id. at 492. One distributor, Empire Storage and Ice Company, refused to do so because such an agreement was a crime under Missouri law. Id. at 492-493. In response to Empire's refusal, union members picketed its place of business and caused a drastic reduction in Empire's business. Id. at 493. When Empire sought an injunction against the picketing, the union responded that its actions were protected under the First Amendment. Id.

The United States Supreme Court rejected the union's argument when it affirmed the trial court's issuance of the summons below. The Court explained that the picketers' "sole, unlawful immediate objective" was to induce Empire to violate the law and reasoned that constitutional free-speech rights do not encompass "speech or writing used as an integral part of conduct in violation of a valid criminal statute." Id. at 498, 502. The Court elaborated that "it has never been deemed an abridgement of freedom of speech or press to make a course of conduct illegal merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed." Id. at 502.

Since Giboney, courts and legal scholars have expressed the need to apply the speech-integral-to-criminal-conduct exception with caution. Buchanan v Crisler, 323 Mich.App. 163, 186; 922 N.W.2d 886 (2018). Otherwise," '[u]nder the broadest interpretation'" of the exception," 'if the government criminalized any type of speech, then anyone engaging in that speech could be punished because the speech would automatically be integral to committing the offense.'" Id., quoting United States v Matusiewicz, 84 F.Supp.3d 363, 369 (D Del, 2015). Accordingly, state and federal legislatures could supersede constitutional free-speech protections so long as they first passed a law criminalizing the speech at hand.

To preserve the efficacy of constitutional free-speech protections, we agree that the speech-integral-to-criminal-conduct exception "cannot be triggered just by speech itself being a violation of a law, even a law that bans conduct as well as speech." Volokh, The "Speech Integral to Criminal Conduct" Exception, 101 Cornell L Rev 981, 1052 (2016). Instead, for the exception to apply, the speech must be integral to some conduct or scheme that is illegal in nature and independent of the speech that might be used to facilitate or accomplish the conduct or scheme. The prosecutor has not alleged that the creation of the robocall or the act of disseminating a robocall in Michigan is illegal per se. The prosecutor instead alleges that the content of the robocall coupled with the defendants' intent to discourage voting is illegal, and an otherwise lawful robocall was the mechanism to broadcast defendants' message. Thus, what the government seeks to punish is defendants' unsavory speech. Were this Court to apply the speech-integral-to-criminal-conduct exception to defendants' conduct in this case, we would be excluding speech from constitutional free-speech protections solely because the law prohibits the content of that speech. Here, the allegedly unlawful conduct was not "in part initiated, evidenced, or carried out by means of language," Giboney, 336 U.S. at 502 (emphasis added); instead, the allegedly unlawful conduct consisted solely of the speech at issue.¹⁷ We decline to apply the speech-integral-to-criminal-conduct exception under these circumstances, and leave for another day the appropriate opportunity to further define the boundaries of this controversial exception to constitutional free-speech protections.

17 Because it is possible to violate MCL 168.932(a) by means of either conduct or conduct plus speech, we do not hold that an offense prosecuted under MCL 168.932(a) is always exempt from the speech-integral-to-criminal-conduct exception.

B. WHETHER MCL 168.932(a) IS VOID FOR VAGUENESS OR OVERBROAD

Having concluded that neither of the prosecutor's two proposed exceptions to the constitutional protections of free speech applies to defendants' conduct, we must next consider defendants' remaining constitutional arguments, including that MCL 168.932(a) is unconstitutionally vague and overbroad on its face.

"It is a basic principle of due process that an enactment is void for vagueness if its prohibitions are not clearly defined." Grayned, 408 U.S. at 108. The vagueness doctrine "incorporates notions of fair notice or warning" and "requires legislatures to set reasonably clear guidelines for law enforcement officials and triers of fact in order to prevent arbitrary and discriminatory enforcement." Smith v Goguen, 415 U.S. 566, 572-573; 94 S.Ct. 1242; 39 L.Ed.2d 605 (1974) (quotation marks and citation omitted). Accordingly, a statute may be considered unconstitutionally vague if it "fail[s] to provide fair notice of the conduct proscribed" or "encourage[s] arbitrary and discriminatory enforcement." People v Harris, 495 Mich. 120, 133, 135; 845 N.W.2d 477 (2014); see also Plymouth Twp v Hancock, 236 Mich.App. 197, 200; 600 N.W.2d 380 (1999). "A statute provides fair notice when it give[s] a person of ordinary intelligence a reasonable opportunity to know what is prohibited," and such knowledge may be acquired "by referring to judicial interpretations, common law, dictionaries, treatises, or the common meaning of words." People v Miller, 326 Mich.App. 719, 738; 929 N.W.2d 821 (2019) (quotation marks and citations omitted; alteration in original). As applied here, we hold that defendants have not demonstrated that MCL 168.932(a) is unconstitutionally vague because the plain-language interpretations of the statutory terms, as detailed earlier, provide fair notice of the conduct being proscribed and prevent arbitrary and discriminatory enforcement of the law.

However, "a clear and precise enactment may nevertheless be 'overbroad' if in its reach it prohibits constitutionally protected conduct." Grayned, 408 U.S. at 114. See also Harris, 495 Mich. at 133-134. The overbreadth doctrine exists to prevent the chilling of speech, but this doctrine may not be "casually employed" and has been considered," 'manifestly, strong medicine.'" In re Chmura, 461 Mich. 517, 531; 608 N.W.2d 31 (2000), quoting Broadrick v Oklahoma, 413 U.S. 601, 613; 93 S.Ct. 2908; 37 L.Ed.2d 830 (1973). "[T]he mere fact that one can conceive of some impermissible applications of a statute is not sufficient to render it susceptible to an overbreadth challenge"; instead, a challenger must prove "a realistic danger that the statute itself will significantly compromise recognized First Amendment protections of parties not before the Court ...." Members of City Council of Los Angeles v Taxpayers for Vincent, 466 U.S. 789, 800-801; 104 S.Ct. 2118; 80 L.Ed.2d 772 (1984).¹⁸ Stated differently, where "conduct and not merely speech is involved," Broadrick, 413 U.S. at 615, a court must first determine "whether the law 'reaches a substantial amount of constitutionally protected conduct,'" People v Rapp, 492 Mich. 67, 73; 821 N.W.2d 452 (2012), quoting Village of Hoffman Estates v The Flipside, Hoffman Estates, Inc, 455 U.S. 489, 494; 102 S.Ct. 1186; 71 L.Ed.2d 362 (1982). "[The] statute's overbreadth must 'be substantial, not only in an absolute sense, but also relative to the statute's plainly legitimate sweep." Rapp, 492 Mich. at 73, quoting United States v Williams, 553 U.S. 285, 292; 128 S.Ct. 1830; 170 L.Ed.2d 650 (2008).

18 This rule is an exception for the First Amendment context from the general overbreadth principle that a litigant may only vindicate a violation of their own constitutional rights. New York v Ferber, 458 U.S. 747, 768-769; 102 S.Ct. 3348; 73 L.Ed.2d 1113 (1982). The courts have drawn this allowance for constitutional free-speech protections because "persons whose expression is constitutionally protected may well refrain from exercising their rights for fear of criminal sanctions by a statute susceptible of application to protected expression." Id. at 768 (quotation marks and citation omitted). In this case, the Court of Appeals erred by applying the standard for other contexts that a facial challenger must" 'establish that no set of circumstances exist[s] under which the [a]ct would be valid.'" Burkman, 341 Mich.App. at 760, quoting People v Johnson, 336 Mich.App. 688, 692; 971 N.W.2d 692 (2021) (second alteration in original).

Here, the statutory provision at issue, MCL 168.932(a), prohibits a person from attempting "by means of bribery, menace, or other corrupt means or device" to influence, deter, or interrupt an elector's vote. As discussed earlier, "other corrupt means or device" includes any other depraved or immoral method or scheme by which a person attempts to influence, deter, or interrupt an elector's vote.

We hold that the statute's catchall "or other corrupt means or device" is unconstitutionally overbroad because it poses a "realistic danger" of infringing constitutional free-speech protections. See id. More specifically, the catchall in MCL 168.932(a) poses a substantial risk of chilling political speech. Political speech is "an essential mechanism of democracy," because it provides "the means to hold officials accountable to the people" and for the people "to make informed choices among candidates for office ...." Citizens United v Fed Election Comm, 558 U.S. 310, 339; 130 S.Ct. 876; 175 L.Ed.2d 753 (2010) (quotation marks and citation omitted). For these reasons, political speech has been historically protected under the First Amendment and laws that burden it are subject to strict scrutiny. Id.; Susan B Anthony List v Driehaus, 814 F.3d 466, 473 (CA 6, 2016) ("Political speech is at the core of First Amendment protections.").

The broad sweep of the catchall language in MCL 168.932(a) conceivably prohibits several forms of purely political speech, including statements made via campaign speeches, rallies, door-to-door campaigning, flyers, and buttons. These political materials are often designed to influence an elector's vote, whether it be to affirmatively vote for a candidate or proposal, not to vote for a candidate or proposal, or not to vote at all, and so satisfy MCL 168.932(a)'s provision that a person be attempting "to influence an elector in giving his or her vote, or to deter the elector from, or interrupt the elector in giving his or her vote at any election held in this state." Although the term "corrupt" in the catchall provision limits MCL 168.932(a)'s scope, it remains likely that political speech is encompassed by "any other depraved or immoral method or scheme" in influencing or deterring votes. For example, one may consider a person posting false information online about a candidate in an effort to influence electors not to cast their votes for the candidate an immoral scheme, thus fulfilling the catchall phrase of MCL 168.932(a). See United States v Alvarez, 567 U.S. 709, 718; 132 S.Ct. 2537; 183 L.Ed.2d 574 (2012) (providing that false statements are generally within constitutional protection). Although the state has an undeniable interest in protecting the electors' franchise, see Mich Alliance for Retired Americans v Secretary of State, 334 Mich.App. 238, 257; 964 N.W.2d 816 (2020), and in "preserving the integrity of their election processes," In re Request for Advisory Opinion, 479 Mich. 1, 19; 740 N.W.2d 444 (2007), that right is not absolute. Laws enacted to preserve these interests must still be narrowly drawn to avoid chilling more speech than is necessary, and the catchall provision in MCL 168.932(a) is not. We conclude that the statute regulates substantially more political speech than its plainly legitimate sweep allows.

Having so concluded, this Court must address next the appropriate remedy. As noted, the invalidation of a statute on the basis of overbreadth "is 'strong medicine' and [we] have employed it with hesitation, and then 'only as a last resort.'" New York v Ferber, 458 U.S. 747, 769; 102 S.Ct. 3348; 73 L.Ed.2d 1113 (1982), quoting Broadrick, 413 U.S. at 613. However, "[a] statute may be saved from being found to be facially invalid on overbreadth grounds where it has been or could be afforded a narrow and limiting construction by state courts ...." People v Douglas, 295 Mich.App. 129, 144; 813 N.W.2d 337 (2011) (quotation marks and citation omitted). See also In re Chmura, 461 Mich. at 544; Broadrick, 413 U.S. at 613 ("Facial overbreadth has not been invoked when a limiting construction has been or could be placed on the challenged statute."). In fact, "[w]e are duty bound under the Michigan Constitution to preserve the laws of this state and to that end to construe them if we can so that they conform to [f]ederal and state constitutional requirements." People v Bricker, 389 Mich. 524, 528; 208 N.W.2d 172 (1973).

Because invalidation should be avoided where possible, we offer a limiting construction of MCL 168.932(a)'s catchall "other corrupt means or device" language. Specifically, we hold that when the charged conduct is solely speech and does not fall under any exceptions to constitutional free-speech protections, MCL 168.932(a)'s catchall phrase operates to proscribe that speech only if it is intentionally false speech that is related to voting requirements or procedures and is made in an attempt to deter or influence an elector's vote. This limiting construction cures the serious and realistic danger that MCL 168.932(a)'s catchall provision infringes constitutional free-speech protections by limiting the statute's reach to areas where government regulation is constitutionally provided or has been historically upheld. See U.S. Const, art 1, § 4, cl 1 (imbuing the states with the authority to regulate the time, place, and manner of congressional elections); Const 1963, art 2, § 4(2) (giving the Michigan Legislature the same authority for state elections and also providing the power "to preserve the purity of elections" and "to guard against abuses of the elective franchise"); Minnesota Voters Alliance v Mansky, 585 U.S. 1, 19 n 4; 138 S.Ct. 1876; 201 L.Ed.2d 201 (2018) ("We do not doubt that the State may prohibit messages intended to mislead voters about voting requirements and procedures."). Intentionally false speech about voting requirements or procedures serves no purpose other than defrauding electors with respect to their franchise. Compare Citizens United, 558 U.S. at 339-340 (discussing the purpose of constitutionally protected political speech).

We reverse the Court of Appeals insofar as it concluded that MCL 168.932(a) was not overbroad, and we offer a limiting construction of the statute's catchall phrase. We remand to the Court of Appeals so that Court may decide whether defendants' conduct falls within the limiting construction of MCL 168.932(a) offered here and, if so, resolve defendants' remaining constitutional arguments.

IV. CONCLUSION

For the reasons expressed in this opinion, we affirm in part and reverse in part the Court of Appeals judgment. Regarding defendants' statutory arguments, we reverse the Court of Appeals' determination that defendants' conduct falls within the statutory term "menace," but affirm its determination that defendants' conduct falls within the statutory catchall term "other corrupt means or device." Regarding defendants' constitutional arguments, we affirm the Court of Appeals' determination that defendants' conduct is not excluded from constitutional free-speech protections under the true-threat exception, but reverse its determination that defendants' conduct is excluded from constitutional free-speech protections under the speech-integral-to-criminal-conduct exception. In sum, we conclude that defendants' conduct falls within the text of MCL 168.932(a) and that defendants' conduct is subject to constitutional free-speech protections. However, because MCL 168.932(a) poses a serious and realistic danger that it will encompass protected speech, we adopt a limiting construction of the statute's catchall provision and remand to the Court of Appeals for consideration of whether defendants' conduct falls within that limiting construction, and, if so, for consideration of defendants' remaining constitutional arguments.

Elizabeth T. Clement, Richard H. Bernstein, Megan K. Cavanagh, Elizabeth M. Welch, Kyra H. Bolden

ZAHRA, J. (concurring in part and dissenting in part).

Defendants, John ("Jack") Burkman and Jacob Wohl, concede that shortly before the 2020 presidential election they attempted to discourage voting by mail by placing robocalls targeted at predominantly African American communities in various states, including Michigan. The message stated:

Hi, this is Tamika Taylor from Project 1599, a civil rights organization founded by Jack Burkman and Jacob Wohl. Mail-in voting sounds great, but did you know that if you vote by mail your personal information will be part of a public database that will be used by police departments to track down old warrants and be used by credit card companies to collect outstanding debts? The CDC [Centers for Disease Control and Prevention] is even

pushing to use records from mail-in voting to track people for mandatory vaccines. Don't be finessed into giving your private information to the man. Stay safe and beware of vote by mail.

After the election, the prosecution charged defendants with violating MCL 168.932(a) by attempting by "menace and/or . . . corrupt means . . . either directly or indirectly, to deter . . . [an elector] in giving his or her vote at any election held in this state." Following a preliminary examination, the district court found that there was probable cause to believe that defendants committed the charged offenses and bound defendants over for trial in the circuit court. In the circuit court, defendants moved to quash the bindover, which was denied.

The Court of Appeals granted defendants' application for leave to appeal, and a majority of that court affirmed the trial court's bindover decisions.¹ We granted defendants' application for leave to appeal.

1 Judge REDFORD concurred in part and dissented in part, arguing that there was not probable cause to believe that defendants' conduct was a "menace" in violation of MCL 168.932(a). He concluded: "In this case, defendants did not express to recipients of the robocalls that defendants would do the acts threatened.... Accordingly, defendants' expression in the robocalls does not fit the jurisprudential concept of a menace or menacing, and therefore, the robocall was not menacing." People v Burkman, 341 Mich.App. 734, 768; 992 N.W.2d 341 (2022) (REDFORD, J., concurring in part and dissenting in part).

I agree with the majority opinion that defendants' conduct was not a "menace" within the meaning of MCL 168.932(a). I dissent, however, from the remainder of the majority opinion, which travels down an uncharted and curious path of interpretation to arrive at a limiting construction of MCL 168.932(a) that narrows the phrase "other corrupt means" to proscribe speech only if it is:

(1) intentionally false,

(2) related to voting requirements or procedures, and

(3) made in an attempt to deter or influence an elector's vote.

Relying on traditional and long-accepted tools of statutory construction, I take a straightforward path and conclude MCL 168.932(a) cannot reasonably be interpreted to prohibit defendants' action in this case. Properly employing the ejusdem generis contextual canon of construction, I conclude MCL 168.932(a) uses "corrupt means" to proscribe:

(1) conduct that is dishonest or without integrity,

(2) that promises or threatens to take or not take action that would benefit or harm an individual,

(3) for the purpose of causing that individual to take an action.

Simply put, the content of the robocall does not threaten or promise to take any action against or in support of its intended recipients. Thus, dissemination of defendants' message was neither a menace (as concluded in the majority opinion) nor accomplished by corrupt means under MCL 168.932(a). The judgment of the Court of Appeals should be reversed in total and this matter remanded to the trial court for the purpose of quashing the information with respect to both defendants. Because defendants' conduct is outside the scope of MCL 168.932(a), I would not consider defendants' weighty and substantial constitutional arguments. For these reasons, as discussed in detail below, I concur only in the conclusion that defendants' conduct was not a menace under MCL 168.932(a) and dissent in all other respects.

I. ANALYSIS²

2 The basic facts and procedural history are not in dispute.

Under MCL 168.932(a), "[a] person shall not attempt, by means of bribery, menace, or other corrupt means or device, either directly or indirectly, to influence an elector in giving his or her vote, or to deter the elector from, or interrupt the elector in giving his or her vote at any election held in this state." The prosecution alleges that, among other things, defendants used corrupt means to deter voting. The Legislature did not statutorily define the phrase "corrupt means." Thus, this Court must use traditional tools of statutory construction to provide meaning to this phrase.

The majority opinion purports to use the ejusdem generis contextual canon of construction to conclude the phrase" 'other corrupt means or device' [means] any other depraved or immoral method or scheme of deterring or preventing someone from voting."³This exceedingly broad construction is the product of a misapplication of the ejusdem generis canon. Properly employing the ejusdem generis canon of construction, I conclude that the placement of the complained-of robocall does not constitute "corrupt means" as that term is used in MCL 168.932(a).

3 Defendants were not charged with alleged wrongful conduct achieved by a "corrupt device." The information regarding each defendant only alleges wrongful conduct by menace or corrupt means. Nonetheless, the majority opinion focuses its statutory construction around the phrase "other corrupt means or device." The majority opinion does not explain why an interpretation of "corrupt device" is relevant.

A. EJUSDEM GENERIS

The contextual canon of construction ejusdem generis provides: "Where general words follow an enumeration of two or more things, they apply to persons or things of the same general kind or class specifically mentioned."⁴ MCL 168.932(a)'s structure and language invoke the ejusdem generis canon because the general phrase "corrupt means" follows the specific terms "bribery" and "menace." Accordingly, this Court must give "corrupt means" the same general kind or class of meaning as "bribery" and "menace."

4 Scalia & Garner, Reading Law: The Interpretation of Legal Texts (St. Paul: Thomson/West, 2012), p 199 ("The ejusdem generis canon applies when a drafter has tacked on a catchall phrase at the end of an enumeration of specifics, as in dogs, cats, horses, cattle, and other animals."). See also People v Smith, 393 Mich. 432, 436; 225 N.W.2d 165 (1975) ("[C]ourts are guided by a rule of construction known as 'ejusdem generis'. This is a rule whereby in a statute in which general words follow a designation of particular subjects, the meaning of the general words will ordinarily be presumed to be and construed as restricted by the particular designation and as including only things of the same kind, class, character[,] or nature as those specifically enumerated."), citing 73 Am Jur 2d, Statutes, § 214, pp 407-408; Norfolk & W R Co v American Train Dispatchers Ass'n, 499 U.S. 117, 129; 111 S.Ct. 1156; '113 L.Ed.2d 95 (1991) ("Under the principle of ejusdem generis, when a general term follows a specific one, the general term should be understood as a reference to subjects akin to the one with specific enumeration."), citing Arcadia v Ohio Power Co, 498 U.S. 73, 84-85; 111 S.Ct. 415; 112 L.Ed.2d 374 (1990).

The ejusdem generis canon does not invite courts to review the specific terms of a statute and give the general term a meaning that is so broad that it covers all aspects of each specific term. For example, if a civil statute regulated the sale of "birds, bats, moths, butterflies, and other creatures," application of the ejusdem generis canon of construction would preclude the phrase "other creatures" from being construed to include creatures that do not have features common among "birds, bats, moths, and butterflies," such as horses, fish, and elephants. Ejusdem generis is a contextual canon; it is a rule to provide meaning through context. Thus, applying this contextual canon of construction, we may reasonably conclude that the phrase "other creatures" is not to be understood to include all other creatures on earth that share a feature with any one of the listed creatures.

Ejusdem generis is not a canon of expansion; in fact, it is precisely the opposite. Ultimately, ejusdem generis narrows the meaning of a general term because it "implies the addition of similar after the word other."⁵ At its core, ejusdem generis requires courts to contextualize a general term before giving it a construction. A court faithfully applying ejusdem generis should determine the potentially valid meanings of the general term, examine the specific terms of the statute to identify commonalities among them, and select a definition of the general term that is in line with and shares commonalities of the specific terms. In the above example, the phrase "other creatures" would be put into context to limit the meaning to something akin to "other flying creatures of a certain size and character."

5 Reading Law, p 199.

Applying the ejusdem generis canon to MCL 168.932(a), we must first interpret the word "corrupt" as used in MCL 168.932(a) so that we will know what qualifies as corrupt means.⁶ Next, we must define the words "bribery" and "menace" with an eye toward identifying the commonalties between the two words. Last, we must contextualize the valid meaning of "corrupt" with the commonalities in "bribery" and "menace." In this way, we can interpret and give meaning to the word "corrupt."

6 As used in MCL 168.932(a), "means" refers to the action taken by a defendant that was intended to accomplish something corrupt. See Webster's New Twentieth Century Dictionary (1956) (defining "means" as "[t]hat which is used to attain an end; the medium through which anything is done or carried out; a measure or measures employed to effect a purpose; agency; medium; instrumentality"); The American College Dictionary (1963) (defining "means" as "an agency, instrumentality, method, etc., used to attain an end"). Here, defendants concede that they took "action" when they caused the robocall to be circulated. Accordingly, this case turns on whether dissemination of the robocall to deter voting by mail was "corrupt" as that word is used in MCL 168.932(a).

Our Legislature has also aided this Court in the endeavor of interpreting statutes. Under MCL 8.3a, "[a]ll words and phrases shall be construed and understood according to the common and approved usage of the language[.]" Therefore, this Court usually presumes that a word should be given its most common meaning, as long as the most common meaning is not plainly inapplicable and does not violate an applicable canon of construction.⁷ "If a statute does not define a word, it is appropriate to consult dictionary definitions to determine the plain and ordinary meaning of the word."⁸ "Each word and phrase in a statute must be assigned such meanings as are in harmony with the whole of the statute, construed in light of history and common sense."⁹

7 Potter v McLeary, 484 Mich. 397, 411; 774 N.W.2d 1 (2009) ("Individual words and phrases, while important, should be read in the context of the entire legislative scheme. In defining particular words in statutes, we must consider both the plain meaning of the critical word or phrase as well as its placement and purpose in the statutory scheme.").

8 People v Feeley, 499 Mich. 429, 437; 885 N.W.2d 223 (2016) (cleaned up).

9 Honigman Miller Schwartz & Cohn LLP v Detroit, 505 Mich. 284, 295; 952 N.W.2d 358 (2020) (quotation marks and citation omitted). See Hecht v Nat'l Heritage Academies, Inc, 499 Mich. 586, 621 n 62; 886 N.W.2d 135 (2016) (" 'An undefined statutory term must be accorded its plain and ordinary meaning. A lay dictionary may be consulted to define a common word or phrase that lacks a unique legal meaning.' ") (citation omitted).

In ascertaining the common meaning of the term "corrupt," we must look to dictionaries printed relatively close to MCL 168.932(a)'s enactment date of 1954.¹⁰

10 The majority opinion compounds its erroneous application of the ejusdem generis canon by consulting dictionaries that were not issued contemporaneously with the legislative enactment of MCL 168.932(a) in 1954. The majority opinion asserts that its definition of "menace" "remains true even if one chooses to consult only historical dictionaries in light of the mid-1800s enactment of the statute's predecessor, 1846 RS, ch 19, § 2." I disagree with the majority opinion's assessment that 1846 RS, ch 19, § 2 is a predecessor to MCL 168.932(a). First, MCL 168.932(a) stands on its own; it is not an amendment of 1846 RS, ch 19, § 2, and it does not adopt, modify, or revise any part of that statute. Second, no evidence that I can find shows that the Legislature of 1954 intended to transplant the phrase "other corrupt means" with its meaning at the time from 1846 RS, ch 19, § 2. I conclude that 1954-not 1846 or 1999-is the relevant time frame within which this statute should be interpreted. Accordingly, I do not rely on any of the dictionaries used in the majority opinion. But even if I were to conclude that 1846 RS, ch 19, § 2 is a predecessor to MCL 168.932(a), that would likely lead to the conclusion that "other corrupt means" is a term of art. And, as discussed below, historical legal dictionaries defined "corrupt" more narrowly than the majority does, such that defendants' conduct likely would not fall within the definition.

Because the meanings of words change over time, a period dictionary is a more reliable indicator of a word's meaning at that time than a dictionary printed well before or after the statute's enactment.¹¹

11 In re Certified Question from the United States Court of Appeals for the Ninth Circuit, 499 Mich. 477, 484; 885 N.W.2d 628 (2016) (explaining that the appropriate dictionary to examine for statutory interpretation is one "from the era in which the legislation was enacted"); Tanzin v Tanvir, 592 U.S. 43, 48; 141 S.Ct. 486; 208 L.Ed.2d 295 (2020) ("Without a statutory definition, we turn to the phrase's plain meaning at the time of enactment."); Reading Law, p 78 ("Words must be given the meaning they had when the text was adopted.") (boldface omitted).

The American College Dictionary (1963) lists the most common definition of "corrupt" as "dishonest; without integrity; guilty of dishonesty, esp[ecially] involving bribery[.]"¹² The most applicable definition in Webster's New Twentieth Century Dictionary (1956) similarly defines "corrupt" as "[t]o pervert or vitiate, as integrity: to bribe; as, to corrupt a judge." Accordingly, the plain meaning of "corrupt," as it was understood near the time of MCL 168.932(a)'s enactment, is conduct that is dishonest or without integrity.

12 The American College Dictionary attempts to "measure[] the occurrences of various meanings [of words] in the general vocabulary . . . [and to use that count] to determine with some certainty which are the common meanings [of words] and to put them first." The American College Dictionary (1963), p vii. This dictionary thus attempts to organize competing definitions of words in descending order, with the most commonly used definition placed first.

The American College Dictionary defines "bribery" as the "act or practice of giving or accepting bribes."¹³ A "bribe" is "any valuable consideration given or promised for corrupt behavior in the performance of official or public duty."¹⁴ Likewise, Webster's New Twentieth Century Dictionary defines "bribery" as "[t]he act or practice of giving or taking rewards for corrupt practices; the act of paying or receiving a reward for a false judgment or testimony, or for the performance of that which is known to be illegal or unjust." The American College Dictionary defines "menace" as "something that threatens to cause evil, harm, injury, etc.; a threat" or, when used as a verb, "to utter or direct a threat against; threaten."¹⁵ Meanwhile, Webster's defines "menace" as "[a] threat; a threatening; a declaration or indication of a disposition, intention, or determination to inflict punishment or other evil; indication of a probable evil to come." Menace and bribery have commonality because each describes the enticement (whether forceful or persuasive) to do something. Essentially, each word expresses an unethical manner of convincing a person to take or refrain from taking an action. Using these definitions in the application of the ejusdem generis canon therefore supports the conclusion that MCL 168.932(a) uses "corrupt" to mean:

13 The American College Dictionary (1963).

14 Id.

15 Id.

(1) conduct that is dishonest or without integrity,

(2) that promises or threatens to take or not take action that would benefit or harm an individual,

(3) for the purpose of causing that individual to take an action.¹⁶

16 If I were to conclude that "corrupt" is a term of art, then 1844 would be the most relevant time period within which to interpret the word "corrupt"; the resulting definition would be even narrower and would require that defendants sought some personal gain. Bouvier's 1839 dictionary did not define "corrupt," but it did define "corruption" as "an act done with an intent to give some advantage inconsistent with official duty and the rights of others. It includes bribery, but is more comprehensive; because an act may be corruptly done though the advantage to be derived from it be not offered by another." 1 Bouvier, A Law Dictionary (1839). The first edition of Black's Law Dictionary, from 1891, had a similar, albeit distinct, definition of "corruption": "Illegality; a vicious and fraudulent intention to evade the prohibitions of the law. The act of an official or fiduciary person who unlawfully and wrongfully uses his station or character to procure some benefit for himself or for another person, contrary to duty and the rights of others." Black, A Law Dictionary (1891). And in the context of a prohibition on governmental officers "corruptly" accepting gifts or gratuities or promises of such, we approved of a jury instruction stating the following:

"A corrupt act means an act characterized by an intent to gain advantage not consistent with one's official duty and the rights of others. It is also defined to be dishonest, without integrity, or guilty of dishonesty involving liberty, as when one accepts money with the intent and purpose of being influenced into doing something inconsistent with his official duty as a public officer." [People v Ewald, 302 Mich. 31, 45; 4 N.W.2d 456 (1942).]

In these cases, there is no evidence that defendants sent out the robocall to gain any advantage or benefit for themselves. Thus, their conduct would not fall under the definition of "corrupt" as that word has been historically used as a legal term of art.

B. APPLICATION

There are strong indications in the record that defendants' robocall message was a malicious and ill-intended attempt to prevent voting by mail. But to bind defendants over for trial under MCL 168.932(a) there must be probable cause to believe that defendants' speech was more than ill-intended, immoral, or depraved. Ill-intended, immoral, or depraved speech, particularly in the political arena, is precisely the kind of speech the First Amendment is designed to protect.¹⁷ Defendants' speech must rise to the level of a menace, bribe, or "corrupt means" that, like menacing and bribery, threatens or promises a harm or benefit in exchange for a change in an elector's vote. Under a proper construction of MCL 168.932(a), probable cause does not support the trial court's bindover decisions.

17 Bible Believers v Wayne Co, Mich, 805 F.3d 228, 243 (CA 6, 2015), citing Nat'l Socialist Party of America v Village of Skokie, 432 U.S. 43, 43-44; 97 S.Ct. 2205; 53 L.Ed.2d 96 (1977) ("The First Amendment offers sweeping protection that allows all manner of speech to enter the marketplace of ideas. This protection applies to loathsome and unpopular speech with the same force as it does to speech that is celebrated and widely accepted. The protection would be unnecessary if it only served to safeguard the majority views. In fact, it is the minority view, including expressive behavior that is deemed distasteful and highly offensive to the vast majority of people, that most often needs protection under the First Amendment.").

The substance of the robocall does not threaten a harm or promise a benefit. The message defendants disseminated amounts to a deceitful, false statement that might make a recipient think twice before casting a vote by mail. Specifically, the robocall intimates that law enforcement, debt collectors, and government health officials will cross-reference voting records with arrest warrants, debtor judgments, and health records to potentially locate someone for whom they have been looking. These distasteful and speculative statements do not amount to a threat. As the majority opinion explains, for a statement to be a threat, "the victim [must] reasonably believe that the speaker will execute the threat." Because no reasonable person would believe that defendants were saying they would use the public database to track down outstanding warrants, collect debts, or track people for mandatory vaccinations, the statement was not a menace or other corrupt means in violation of the statute.¹⁸

18 Although not argued by any litigant, it is worth noting that MCL 168.932(a) bars bribery, menace, or other corrupt means that either "directly or indirectly" influences, deters, or interrupts the vote of an elector. I am not persuaded that defendants indirectly violated the statute. Defendants placed a robocall allegedly from Tamika Taylor, presumably a fictitious character, who conveyed privacy concerns allegedly related to mail-in voting. Such a message is simply not an indirect threat.

In sum, probable cause does not support a finding that there was probable cause to believe that defendants used corrupt means to dissuade voting by mail, and the judgment of the Court of Appeals should be reversed in total. This matter should be remanded to the trial court with instructions to quash the information with respect to both defendants.

II. WHERE THE MAJORITY OPINION WENT ASTRAY

A. EJUSDEM GENERIS

The majority opinion misdescribes and misapplies the ejusdem generis canon. The majority opinion reasons that" 'bribery' and 'menace'-the terms that precede the general catchall-both inform the definition of 'corrupt means or device[s],' because that term must also include 'bribery' and 'menace.'" The majority opinion goes astray from the traditional application of the ejusdem generis canon when it gives "corrupt means" a meaning inclusive of the most universal characteristics of both "bribery" and "menace"- two vastly disparate terms. The ejusdem generis canon requires courts to give a general term a meaning that incorporates the commonalities of the specific terms that precede it.¹⁹Ironically, while attempting to exhaustively include the base characteristics of "bribery" and "menace" into the general term "corrupt," the majority opinion has given that general phrase a meaning that is unlike the specific terms that precede it. "Menace" and "bribery" each describe a particular kind of action that threatens or promises a harm or benefit to a person unless they comply with the speaker's wishes. The majority opinion ignores the common aspects of these terms and broadly defines "corrupt means" as any ill intent. Consequently, the majority opinion's interpretation of "corrupt means" is overly broad and indiscriminate, encompassing all ill-intended actions. Such an interpretation is a misapplication of the ejusdem generis canon because it gives the general term a meaning unlike and not in line with the specific terms' meanings. "Corrupt" can mean anything depraved or immoral, but when put in context through the ejusdem generis canon it is reasonably understood as used in MCL 168.932(a) to mean action that is dishonest or without integrity by promising or threatening to take or not take action that would benefit or harm an individual to persuade that individual to take an action that the individual would otherwise not take, such as bribery and menace.²⁰ The phrase "corrupt means" is so broadly interpreted in the majority opinion that it arrives at the inescapable conclusion that MCL 168.932(a) is unconstitutionally overbroad because it encompasses too much innocent activity, rendering the statute unworkable.

19 See Rott v Rott, 508 Mich. 274, 309-310; 972 N.W.2d 789 (2021) (VIVIANO, J., concurring in part and dissenting in part). Under the ejusdem generis canon, the general term "bear[s] a restricted meaning"-one "including only things of the same kind, class, character or nature as those specifically enumerated." Id. at 309 (quotation marks and citations omitted).

20 The construction of the phrase "corrupt means or device" adopted in the majority opinion also violates the surplusage canon of construction, which directs that, where possible, every word and every provision used in a statute is to be given effect. Reading Law, p 174. If, as the majority opinion holds, "corrupt" means any action of ill intent, the words "menace" and "bribery" become meaningless surplusages within MCL 168.932(a). In footnotes 10 and 11 of its opinion, the majority defends itself by arguing:

[U]nder the dissent's narrow construction of the catchall term, it would not be a violation of MCL 168.932(a) for a person to engage in a robocall campaign that falsely informs voters that the polls will remain open three hours later because such action does not promise or threaten to take or not take action that would benefit or harm an individual.

The majority opinion effectively argues that a narrow construction is improper because under a narrow construction MCL 168.932(a) would not apply to every politically motivated high jinks imaginable. But Michigan has other statutes and common-law offenses that work in concert with MCL 168.932(a). For example, depending on the exact content of the message, falsely informing voters that polls will remain open three hours later might violate MCL 750.217c (falsely impersonating a public officer or employee) or it might be fraud in the inducement. See Hi-Way Motor Co v Int'l Harvester Co, 398 Mich. 330, 336; 247 N.W.2d 813 (1976). Indeed, the majority opinion's response nakedly states what I only suggested: that the majority's reading is intended to expand MCL 168.932(a) so that it empowers the government to prosecute broad swaths of constitutionally protected political speech.

The majority opinion also goes astray by trying to buttress its definition of "corrupt means" by looking at caselaw defining "corrupt" in the context of the common-law crime of misconduct in office, which has nothing to do with MCL 168.932(a). In the first case, People v Perkins,²¹ the word "corrupt" was not contained in any statute under review. The Court defined the phrase "corrupt intent" because the misconduct-in-office statute did not have a mens rea. In the context of a common-law crime that was adopted without the addition of a mens rea element, it was proper for this Court to give the phrase "corrupt intent" a broad construction to describe the mens rea required for misconduct in office, a general-intent offense. Perkins merely stands for the proposition that this Court will create a base mens rea for a criminal offense when the political branches codify a common-law criminal offense without specifying a mens rea for it. The second case cited in the majority opinion, People v Waterstone,²² is inapposite and unpersuasive for precisely the same reason. There too this Court interpreted the word "corruption" to have a broad meaning in the context of the common-law offense of misconduct in office in which the political branches had codified a common-law offense but not specified its mens rea.²³

21 People v Perkins, 468 Mich. 448; 662 N.W.2d 727 (2003).

22 People v Waterstone, 296 Mich.App. 121; 818 N.W.2d 432 (2012).

23 Id. at 138.

B. THE PRESUMPTION-OF-VALIDITY CANON

In addition to its flawed application of the ejusdem generis canon, the majority opinion's construction is also in conflict with the presumption-of-validity canon. The presumption of validity is a fundamental canon that "disfavors interpretations that would nullify [a] provision or [an] entire instrument [or] . . . cause a statute to be unconstitutional."²⁴ The Supreme Court of the United States concisely described this canon by observing that "when a statute is reasonably susceptible of two interpretations, by one of which it is unconstitutional and by the other valid, the court prefers the meaning that preserves to the meaning that destroys."²⁵ The presumption-of-validity canon, however, cannot be used to revive dead or doomed statutes that have only one possible, unconstitutional construction; it is not an invitation to misinterpret a statute.²⁶

24 Reading Law, p 66 (discussing the presumption of validity and constitutional doubt canon); see also Mich Dep't of Transp v Tomkins, 481 Mich. 184, 191; 749 N.W.2d 716 (2008) ("Statutes are presumed constitutional ....").

25 Panama Refining Co v Ryan, 293 U.S. 388, 439; 55 S.Ct. 241; 79 L.Ed. 446 (1935) (Cardozo, J., dissenting), citing United States v Delaware & Hudson Co, 213 U.S. 366, 407; 29 S.Ct. 527; 53 L.Ed. 836 (1909); Knights Templars' & Masons' Life Indemnity Co v Jarman, 187 U.S. 197, 205; 23 S.Ct. 108; 47 L.Ed. 139 (1902).

26 See Reading Law, pp 66, 68 (explaining that "the proper application of the presumption of validity" permits a construction of a statute only" 'if it may reasonably be done[.]' "), quoting United States v Cornell, 25 F Cas 646, 649 (CCD RI, 1819).

The canon is anchored to the idea that the Legislature would not have enacted a facially unconstitutional statute, so a word or phrase should not be interpreted to make the statute unconstitutional when an equally valid alternative construction is available that would make the statute constitutional. The canon also promotes a respect for the publicpolicy choices of the political branches by counseling courts to avoid choosing to vitiate democratically enacted statutes when another reasonable interpretation would save the statute.

Here, the majority opinion concludes the phrase "corrupt means" is overbroad and unconstitutional:

[T]he statutory provision at issue, MCL 168.932(a), prohibits a person from attempting "by means of bribery, menace, or other corrupt means or device" to influence, deter, or interrupt an elector's vote. As discussed earlier, "other corrupt means or device" includes any other depraved or immoral method or scheme by which a person attempts to influence, deter, or interrupt an elector's vote.

We hold that the statute's catchall "or other corrupt means or device" is unconstitutionally overbroad because it poses a "realistic danger" of infringing constitutional free-speech protections. [Emphasis added.]

After concluding that MCL 168.932(a) is unconstitutionally vague and overbroad, the majority opinion curiously resuscitates the statute it deemed unconstitutional, i.e., dead letter, by providing a "limiting construction of MCL 168.932(a)." Under a proper application of the presumption-of-validity canon, however, a statute would not be declared unconstitutional when it may fairly be interpreted in a manner that would render the statute constitutional.

I seriously question the method of construction embraced in the majority opinion to arrive at its "limiting construction," the entire sum and substance of which is as follows:

[W]e hold that when the charged conduct is solely speech and does not fall under any exceptions to constitutional free-speech protections, MCL 168.932(a)'s catchall phrase operates to proscribe that speech only if it is intentionally false speech that is related to voting requirements or procedures and is made in an attempt to deter or influence an elector's vote.

This construction is unsupported by the language of MCL 168.932(a), and no canon of construction or other traditional interpretive tool of construction is offered to support it. In short, it is a construction drawn from thin air. As a result, the majority opinion subjects defendants to criminal liability that the Legislature had not plainly expressed, raising additional concern about whether defendants had constitutional notice of the majority's newly imposed "limiting construction of MCL 168.932(a)" when they committed the alleged offense.

III. CONCLUSION

The evidence does not support a finding that defendants used menace or corrupt means to dissuade voting. Accordingly, this Court ought to reverse the Court of Appeals' opinion affirming the trial court's decision to bind defendants over and remand this case to the trial court with instructions that the trial court quash the charges brought under MCL 168.932(a).