1
STATE OF MINNESOTA

IN SUPREME COURT

A20-0231

Court of Appeals Chutich, J.
Dissenting, Thissen, J.
State of Minnesota,

Respondent,

vs. Filed: March 9, 2022
Office of Appellate Courts
Chris A. Mrozinski,

Appellant.

________________________
Keith Ellison, Attorney General, Saint Paul, Minnesota; and

Kimberly J. Maki, St. Louis County Attorney , Tyler J. Kenefick, Assistant County
Attorney, Hibbing, Minnesota, for respondent.

Cathryn Middlebrook, Chief Appellate Public Defender, Roy G. Spurbeck, Assistant State
Public Defender, Saint Paul, Minnesota, for appellant.

Cicely R. Miltich, Assistant Attorney General, Saint Paul, Minnesota, for amicus curiae
Minnesota Attorney General Keith Ellison.
________________________
S Y L L A B U S
1. A threat of violence is made “in a reckless disregard of the risk of causing
such terror,” in violation of the Minnesota threats of violence statute, Minnesota Statutes
section 609.713, subdivision 1 (2020), when the defendant makes the violent threat whil e
2
consciously disregarding a substantial and unjustifiable risk that the defendant’s words or
actions will cause terror.
2. True threats, a category of speech unprotected by the First Amendment, do
not require specific intent to threaten a victim and can encompass violent communications
that are made recklessly; Minnesota’s threats of violence statute, Minnesota Statutes
section 609.713, subdivision 1, punishes only true threats.
3. Minnesota’s threats of violence statute, Minnesota Statutes section 609.713 ,
subdivision 1, is not substantially overbroad under the First Amendment.
Affirmed.
O P I N I O N
CHUTICH, Justice.
This case involves a facial challenge to a part of the Minnesota threats of violence
statute, Minnesota Statutes section 609.713, subdivis ion 1 (2020), under the First
Amendment to the United States Constitution. The State charged appellant Chris A.
Mrozinski with four counts of threats of violence under section 609.713, subdivision 1.
Mrozinski moved to dismiss, claiming that the part of section 609.713, subdivision 1 that
applies to threats of violence made “in a reckless disregard of the risk of causing such
terror” was unconstitutionally overbroad. The district court denied her motion and found
her guilty of all four counts in a stipulated-facts court trial. The court of appeals affirmed.
Because we conclude that Minnesota Statutes section 609.713, subdivision 1, does not
violate the First Amendment, we affirm.
3
FACTS
In January 2017, Mrozinski slid an envelope under the door of the In itial
Intervention Unit of St. Louis County Children’s Protection Services. She had handwritten
“MISS ME YET?” on the outside of the envelope, which contained a letter and four toe
tags, like ones used to identify bodies in a morgue. The letter read:
JUST A LITTLE NOTE FOR ALL MY FRIENDS @
CPS . . .

DID YOU REALLY THINK YOUR LAME-ASS THREATS
TO CONTINUE YOUR ILLEGAL COURTHOUSE BULLSHIT
WOULD SCARE ME, YOU FUCKING CUNTS?!

MY CHILDREN WILL BE 16 SOMEDAY, AND YOU
WON’T BE ABLE TO DO SHIT ABOUT IT.
CHILDHOOD IS NOT FOREVER.
DEATH, ON THE OTHER HAND, IS.

SLEEP TIGHT, BITCHES!

(PERHAPS I SHOULD SAY . . . SLEEP WITH ONE EYE OPEN?)
Each of the four toe tags contained the handwritten name of a different person
associated with Mrozinski’s child protecti on matter that resulted in Mrozinski losing
custody of her children. In the space on the tags labeled “case number” were handwritten
numbers starting with “#1 of 9” going up to “#4 of 9.” Each toe tag also included a
handwritten address for “place of death,” a date for “date of birth,” “TBD” for “date of
death,” and insulting names in the other spaces. Some of the birthdays and addresses were
accurate, but some were not. All four people named on the toe tags reported making
changes to their daily routines and taking safety precautions accordingly. Three of the
4
persons named said that they believed that Mrozinski was capab le of carrying out her
threats.
Mrozinski was charged with four counts of threats of violence, one for each person
named in the toe tags, under Minnesota Statutes section 609.713, subdivision 1 (the
Statute). She moved to dismiss, arguing that the prohibi tion in the Statute of threats of
violence made “in a reckless disregard of the risk of causing such terror” was, among other
claims, facially invalid under the First Amendment. The district court denied the motion,
finding that the Statute did not facially violate the First Amendment. In a stipulated -facts
court trial, the district court found Mrozinski guilty on all counts, finding that Mrozinski
made true threats in “reckless disregard” of the terror they might cause. The district court
stayed execution of a 1 -year jail sentence and placed her on unsupervised probation for
1 year.
The court of appeals affirmed in a nonprecedential opinion. State v. Mrozinski ,
No. A20-0231, 2021 WL 416739 (Minn. App. Feb. 8, 2021). The court concluded that
even if t he Statute prohibited some protected speech, it did not sweep in a substantial
amount of protected speech and was therefore not facially overbroad under the First
Amendment. Id. at *5–11. We granted Mrozinski’s petition for review.1

1 At the court of appeals, Mrozinski renewed her arguments from the district court,
which challenged the Statute’s constitutionality on its face and as applied on First
Amendment and due process grounds. We granted review only on the issue of facial
overbreadth under the First Amendment.
5
ANALYSIS
Mrozinski asserts a facial challenge to the constitutionality of part of the Minnesota
threats of violence statute on First Amendment overbreadth grounds.2 We review questions
of law, including the constitutionality of statutes, de novo. State v. Jorgenson, 946 N.W.2d
596, 601 (Minn. 2020). We presume that statutes are constitutional and strike them down
“only if absolutely necessary.” Id. In the context of First Amendment challenges,
however, the “State bears the burden of showing that a content-based restriction on speech”
is constitutional. State v. Melchert-Dinkel, 844 N.W.2d 13, 18 (Minn. 2014). Defendants
have standing to challenge a statute as unconstitutionally overbroad even when it is not
unconstitutional as applied in their case “because prior restra int of free speech poses a
greater harm to society than does the possibility that some unprotected speech will go
unpunished.” Jorgenson, 946 N.W.2d at 601. Accordingly, Mrozinski may bring an
overbreadth challenge regardless of whether her conduct could have been constitutionally
punished.
The First Amendment prohibits the government from “abridging the freedom of
speech.” U.S. Const. amend. I; see also Melchert-Dinkel, 844 N.W.2d at 18 (noting that

2 Mrozinski summarily argues that the Statute fails strict scrutiny, but she does not
engage in any analysis to support her argument. An issue raised without “argument or
citation to legal authority in [its] support” is forfeited. State v. Krosch, 642 N.W.2d 713,
719 (Minn. 2002); see also State v. Ture, 632 N.W.2d 621, 632 (Minn. 2001) (holding that
when a party “fails to provide any authority or argument to support [a] claim,” the claim is
forfeited). Accordingly, because Mrozinski forfeited her argument that s trict scrutiny
should apply to the Statute, we do not apply the analytical approach recently laid out in
State v. Casillas, 952 N.W.2d 629, 645 –46 (Minn. 2020) (stating that a scrutiny analysis
should be conducted first when a statute is challenged on both scrutiny and overbreadth
grounds). Here, we need only address the overbreadth challenge.
6
the First Amendment “applies to the states through the Fourteenth Amendment”). The
First Amendment establishes that the government “has no power to restrict expression
because of its message, its ideas, its subject matter, or its content.” Police Dep’t of Chicago
v. Mosley, 408 U.S. 92, 95 (1972). Yet cert ain types of speech have traditionally been
excluded from protection under the First Amendment; Congress and the states can
constitutionally prohibit these categories of speech . In re Welfare of A.J.B. , 929 N.W.2d
840, 846 (Minn. 2019). “Exceptions to Fi rst Amendment protections generally fall into
several delineated categories,” which include obscenity, defamation, fighting words, child
pornography, and true threats. Id.; United States v. Alvarez, 567 U.S. 709, 717 (2012).
Statutes are overbroad when they prohibit more than just unprotected speech and
sweep in a substantial proportion of protected speech and expressive conduct in
comparison with unprotected speech and conduct. A.J.B., 929 N.W.2d at 847. A
challenger asserting overbreadth must establish that “a substantial number of [a statute’s]
applications are unconstitutional, judged in relation to the statute’s plainly legitimate
sweep.” United States v. Stevens , 559 U.S. 460, 473 (2010) (internal quotation marks
omitted).
We use a four -step process to analyze questions of overbreadth under the First
Amendment. A.J.B., 929 N.W.2d at 847 –48. Because it is not possible to know whether
a statute sweeps in too much protected speech without first knowing what it covers, our
first step is to construe the statute. Id. Construing the statute includes noting in what ways
the statute covers broad or narrow conduct and any limiting factors. See id. at 849–50
(discussing language and elements that broaden or limit the reach of the statute).
7
In the second step, we determine whether the statute’s reach is “limited to
unprotected categories of speech or expressive conduct.” Id. at 847; State v. Hensel, 901
N.W.2d 166, 171 (Minn. 2017). When necessary, we clarify any confusion i n the
traditional categories of unprotected speech so that the scope of each category is clear
enough to address the parties’ arguments. See A.J.B., 929 N.W.2d at 852 (explaining the
scope of the “speech integral to criminal conduct” exception). When the statute covers
only unprotected speech, we typically end our analysis and uphold the statute. Id. at 847.
When our analysis shows that the statute covers more than just unprotected speech,
however, we move to the third step and determine whether the statute prohibits a substantial
amount of protected speech in comparison with unprotected speech. Id. (“If we conclude
that the statute is not limited to unprotected speech or expressive conduct, we turn to the
core overbreadth inquiry . . . .”). This step involves revisiting the language of the statute
and identifying whether it sweeps in only a little or a substantial amount of protected
speech. Id. at 853 (returning to the language of the statute and discussing its wide breadth).
We address the fourth and last step only when the statute prohibits a substantial
amount of protected speech. Id. at 848. At this stage, we determine whether we can save
the statute by narrowly construing or severing language to avoid the constitutional problem.
Id. When we cannot do so, we strike the statute down. Id.
I.
We begin with the Statute. The Minnesota threats of violence statute makes it a
crime for a person to “threaten[], directly or indirectly, to commit any crime of violence
with purpose to terrorize another . . . or in a reckless disregard of the risk of causing such
8
terror.” Minn. Stat. § 609.713, subd. 1 (emphasis added). Section 609.713, subdivision 1,
defines “crime of violence” by adopting the definition found in another statute, Minn. Stat.
§ 609.1095, subd. 1(d) (2020).
Mrozinski does not challenge the part of the Statute that prohibits threats of violence
made with the purpose to terrorize another. This is because, as we explain more fully
below, the Supreme Court of the United States has held that a threat of violence made with
the intent to intimidate is a true threat that may be prohibited under the First Amendment.
See Virginia v. Black, 538 U.S. 343, 360 (2003). Rather, Mrozinski challenges that part of
the Statute under which she was convicted, which punishes threats of v iolence made in
reckless disregard of the risk of causing terror.
We have previously interpreted many of the terms in the Statute. “A threat is a
declaration of an intention to injure another or [her] property by some unlawful act.” State
v. Schweppe, 237 N.W.2d 609, 613 (Minn. 1975). We must view words or phrases in their
context to determine whether they are harmless or threatening. Id. A communication is
threatening when, considered in context, it creates reasonable apprehension that the
defendant will carry through with or act on the threat. Id. In addition, the Statute only
prohibits a threat to specifically commit a crime of violence. Minn. Stat. § 609.713,
subd. 1. Notably, the statutory definition of “crime of violence” encompasses only serious
violent crimes, including felony assault and murder. 3 Minn. Stat. § 609.1095, subd. 1(d).

3 Minnesota Statutes section 609.1095, subd ivision 1(d), defines violent crime by
cross-listing various statutes, including those punishing murder, manslaughter, kidnapping,
felony assault, certain drug crimes causing injury, criminal sexual conduct, arson, and
more.
9
Lastly, the defendant must act “in a reckless disregard of the risk of causing such terror.”
Minn. Stat. § 609.713, subd. 1. To cause terror means “to caus e extreme fear by use of
violence or threats.” Schweppe, 237 N.W.2d at 614.
We have not yet defined what acting recklessly means under this Statute. In a
different First Amendment context, however, we concluded that recklessness requires
deliberate act ion in disregard of a known, substantial, and unjustifiable risk. State v.
Mauer, 741 N.W.2d 107, 115 (Minn. 2007) (addressing a First Amendment challenge to a
child pornography statute). Under this definition, a person acts recklessly when she
“ ‘consciously disregards a substantial and unjustifiable risk that the element of an offense
exists.’ ” Id. (quoting State v. Zupetz, 322 N.W.2d 730, 733 (Minn. 1982)). We conclude
that this same definition should apply here because it aligns with the common
understanding of the term “reckless” and is consistent with our First Amendment
precedent. See State v. Engle , 743 N.W.2d 592, 594 (Minn. 2008) (explaining that the
recklessness definition adopted in Zupetz and Mauer “comports with the most common
usage of the term”). In the context of this crime, to act “in a reckless disregard of the risk
of causing such terror,” a defendant must be aware of a substantial and unjustifiable risk
that her words or actions will cause terror in another, and she must act in con scious
disregard of that risk.
In sum, a person recklessly makes threats of violence, in violation of
section 609.713, subdivision 1, when (1) through words or actions, she communicates an
intention to injure another or their property; (2) the threat is to commit a statutorily defined
crime of violence; (3) in context, those words or conduct create a reasonable apprehension
10
that she will follow through with or act on the threat; and (4) she makes the violent threat
in conscious disregard of a substantial and unjustifiable risk that her words or conduct will
cause extreme fear.
II.
In the second step of our traditional overbreadth analysis, we consider whether a
statute’s reach is “limited to unprotected categories of speech or expressive conduct.”
A.J.B., 929 N.W.2d at 847. In other words, we ask here whether the Statute’s reach falls
outside the scope of the First Amendment entirely because it prohibits only true threats, a
traditionally unprotected category of speech. Id. at 846. But because the mental state
required for speech to be considered an unprotected true threat is unsettled, we must first
define the scope of the true threats exclusion from the First Amendment’s protection, see
Part II.A., before determining the Statute’s reach in Part II.B. below.
A.
The parties disagree on the requirements for a communication to qualify as a true
threat—and consequently whether the Statute punishes any protected speech. Mrozinski
contends that the Supreme Court’s decision in Virginia v. Black , 538 U.S. 343 (2003),
created a specific intent requirement for words or conduct to qualify as a true threat and
thereby fall outside of the protection of the First Amendment. The crux of Mrozinski’s
argument is that a mental state of recklessness for true threats is incompatible with the First
Amendment and potentially sweeps up a wide range of protected speech.
The State maintains that neither Black, nor any other Supreme Court case, created
or implied a specific intent requirement for t rue threats. Additionally, the State compares
11
true threats with other categories of unprotected speech that do not require specific intent,
such as child pornography, obscenity, or defamation, and argues that the same standard
should apply for true threat s. We discuss these arguments below after considering the
background of the true threats doctrine. Ultimately, we agree with the State that true threats
do not require specific intent.
1.
The Supreme Court first recognized the exclusion of true threats f rom First
Amendment protection in Watts v. United States , 394 U.S. 705 (1969). Watts, a young
man who had just received his draft card, spoke out against the draft at a political rally in
the late 1960s. Id. at 706. He was convicted of “knowingly and willfully” threatening the
President because he stated, “If they ever make me carry a rifle the first man I want to get
in my sights is L.B.J.” Id. at 705 –06. Watts made this statement after swearing that he
would never join the armed forces, and the crowd’s response was to laugh. Id. at 706–07.
The Court held that although the statute forbidding threats against the President was
certainly constitutional on its face, it was not constitutional as applied to Watts. Id. at
707−08. Based on the context, his statements only amounted to political hyperbole; “a
kind of very crude offensive method of stating a political opposition to the President.” Id.
at 708. The Court acknowledged that true threats fall outside of the First Amendment and
that the government can constitutionally prohibit them. Id. at 707 (“What is a threat must
be distinguished from what is constitutionally protected speech.”). The Court did not,
however, provide a definition or require a certain mental state for true threats; it merely
concluded that Watts’s political hyperbole did not amount to a true threat. See id. at 708.
12
The Supreme Court dealt with true threats again in Virginia v. Black, 538 U.S. 343
(2003), a case involving Virginia’s cross -burning statute. That statute made it “u nlawful
for any person or persons, with the intent of intimidating any person or group of persons ,
to burn, or cause to be burned, a cross on the property of another, a highway or other public
place.” Va. Code Ann. § 18.2 -423 (2021) (emphasis added). A m ajority of the Court
explained that 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,” regardless of w hether the speaker actually
intends to carry out the threat. Black, 538 U.S. at 359–60.
The Court then stated that “[i]ntimidation in the constitutionally proscribable sense
of the word is a type of true threat, where a speaker directs a threat to a person or group of
persons with the intent of placing the victim in fear of bodily harm or death.” Id. at 360.
The Court also elaborated on the sound reasoning behind excluding true threats from
protected speech: “[A] prohibition on true threats ‘protect[s] i ndividuals from the fear of
violence’ and ‘from the disruption that fear engenders,’ in addition to protecting people
‘from the possibility that the threatened violence will occur.’ ” Id. at 360 (second alteration
in original) (quoting R.A.V. v. St. Paul , 505 U.S. 377, 388 (1992)). The Virginia statute
passed constitutional muster because cross burning done with the statutorily required intent
to intimidate was a true threat, and the intent to intimidate distinguished this type of cross
burning from other types of cross burning that were protected speech. Id. at 363.
13
A plurality of the Court did, however, invalidate a part of the statute that contained
a prima facie evidence provision. 4 Id. at 367. The plurality concluded that the provision,
as interpreted in a jury instruction, allowed evidence of the cross burning, by itself, to be
sufficient for a jury to find an intent to intimidate if the defendant presented no evidence.
Id. at 365. I t held that this provision essentially created strict liability for cross burning
because it “permit[ted] the Commonwealth to arrest, prosecute, and convict a person based
solely on the fact of cross burning itself.” Id. Consequently, the plurality determined that
this provision was unconstitutional on its face “at this point,” but left open the possibility
that the Virginia Supreme Court could interpret the provision differently or conclude that
it was severable on remand.5 Id. at 367.
2.
After Black, federal circuit and state courts split on the issue of what level of intent
true threats require. The Court had an opportunity to resolve the split in Elonis v. United
States, a case that challenged a federal statute that punished the transmission in inte rstate
commerce of “any communication containing any threat . . . to injure the person of
another,” but did not contain a mental state. 575 U.S. 723, 732, 740 (2015) (quoting 18
U.S.C. § 875(c) (2019)). The Court chose to decide the case on purely statut ory grounds.

4 The prima facie evidence provision reads: “Any such burning of a cross shall be
prima facie evidence of an intent to intimidate a person or group of persons.” Va. Code
Ann. § 18.2-423, held unconstitutional by Elliott v. Commonwealth, 593 S.E.2d 263, 267
(Va. 2004).

5 On remand, the Virginia Supreme Court concluded that the jury instruction properly
interpreted the provision, determined it was unconstitutional on its face, and severed it from
the statute. Elliott, 593 S.E.2d at 267–68.
14
The Court explained that, as a general rule, when a federal statute omits a mental state, it
must read a mental state into the statute that requires some awareness of wrongdoing and
more than a negligence standard. Id. at 734–37. The Court specifically declined to address
what mental state above negligence was appropriate as a matter of statutory interpretation,
leaving the question open of whether recklessness would suffice. Id. at 740. The Court
also declined to address the First Amendm ent question, finding it unnecessary given its
disposition of the case. Id.
Consistent with the State’s position, most state and federal courts presented with the
issue have decided that true threats do not require specific intent to threaten. These courts,
as well as Justices Thomas and Alito in their separate writings in Elonis—which urged
addressing the First Amendment issue that the majority declined to reach —reject
interpreting Black as creating a specific intent requirement for true threats. 6

6 See Elonis, 575 U.S. at 765 (Thomas, J. dissenting) (stating that the Court in Black
had no reason to decide whether true threats required specific intent because the statute at
issue explicitly required an intent to intimidate); id. at 746–48 (Alito, J., concurring in part
and dissenting in part) (contending that, consistent with Black, “ [r]equiring proof of
recklessness is . . . sufficient” under the First Amendment); United States v. Jeffries , 692
F.3d 473, 479 (6th Cir. 2012) (“[Black] says nothing about imposing a subjective standard
on other threat-prohibiting statutes, and indeed had no occasion to do so: the Virginia law
itself required subjective ‘intent.’ ”); State v. Trey M. , 383 P.3d 474, 481 (Wash. 2016)
(“[T]he ‘intent to intimidate’ element was . . . part of the given context of the Supreme
Court’s First Amendment inquiry in Black, but nothing in Black imposes in all cases an
‘intent to intimidate’ requirement”); see also People v. Ashley , 162 N.E.3d 200, 215 (Ill.
2020) (“[W]e construe the phrase ‘means to communicate’ [from Black] as requiring that
the accused be consciously aware of the threatening nature of [her] speech . . . .”); Major
v. State , 800 S.E.2d 348, 352 (Ga. 2017) (holding that recklessness “fits within the
definition of a true threat” because it “requires a knowing act”). The minority of courts
that do require true threats to be made with specific intent read Black to so dictate. See
United States v. Heineman , 767 F.3d 970, 978 (10th Cir. 2014) (“We read Black as

15
We agree. The Supreme Court in Black did not resolve this issue; 7 it had no
occasion to decide whether true threats require specific intent because the Virginia statute
at issue required a specific intent to intimidate. See 538 U.S. at 362. Moreover, that the
Court in Elonis specifically declined to consider the First Amendment question suggests
that it concluded that Black was not determinative. See Elonis, 575 U.S. at 740; see also
People in Interest of R.D., 464 P.3d 717, 729 (Colo. 2020) (“Thus, after Elonis, the proper
test for true threats remains an unsolved doctrinal puzzle.”).
Accordingly, we join the majority of courts to have considered the issue and agree
with the State that a specific intent to threaten the victim is not required for violent speech
or expressive conduct to be a true threat. The Supreme Court’s statement that true threats
“encompass those statements where the speaker means to communicate a serious
expression of an intent to commit an act of unlawful violence” does not clarify what
minimal mental state is required for speech to be a true threat. See Black, 538 U.S. at 359.
Additionally, the Court’s explanation of intimidation simply referred to the mental state set
forth in the Virginia statute, stating that intimidation was “a type of true threat.” Id. at 360
(emphasis added); see also United States v. Jeffries , 692 F.3d 473, 480 (6th Cir. 2012)

establishing that a defendant can be constitutionally convicted of making a true threat only
if the defendant intended the recipient of the threat to feel threatened.”); State v. Boettger,
450 P.3d 805, 812 (Kan. 2019) (asserting that “[ Black] explains the intent necessary to
have a true threat prosecuted without violating the First Amendment’s protections,”
although it “did not directly address” a conviction made without intent to cause fear).

7 This opinion and the dissent are joined on this point—that neither Black nor other
Supreme Court precedent compels a specific intent requirement for a true threat .
See D-4–D-5.
16
(interpreting Black’s statements about intimidation as merely describing one type of true
threat that was specifically penalized in the statute, not all true threats).
Moreover, the plurality’s invalidation of the prima facie evidence provision in Black
does not show that specific intent to intimidate is essential to separating true threats from
protected speech. The plurality in Black explained that “the provision as s o interpreted
would create an unacceptable risk of the suppression of ideas” because it allowed
conviction “based solely on the fact of cross burning itself,” without regard to context. 538
U.S. at 365 (internal quotation marks omitted). This explanation suggests not that specific
intent to intimidate is required in every case to distinguish protected speech from true
threats, but rather that “[s]tates must prove more than the mere utterance of threatening
words—some level of intent is required.” Perez v. Florida, 580 U.S. __, 137 S. Ct. 853,
855 (2017) (Sotomayor, J., concurring in denial of petition for writ of certiorari).
Accordingly, Black does not definitively answer the question of what minimal mental state
true threats require, but the plurality suggests that some level of intent is necessary as part
of a contextual analysis.
We also conclude that the reasoning behind the exclusion of true threats from First
Amendment protection supports rejecting an exclusive specific intent mental state and
permitting a reckless one to suffice. Excluding true threats “protect[s] individuals from the
fear of violence” and “from the disruption that fear engenders.” R.A.V., 505 U.S. at 388.
Additionally, it protects people “from the possibility that the threaten ed violence will
occur” because actual violence does often follow threats of violence. Id. These purposes
are best served by including within the definition of true threats those communications that,
17
although made without specific intent, were communicated in a reckless fashion. Whether
the person making the communication to commit a violent act specifically intends to
threaten the victim, a violent statement can still elicit fear and cause harm; it can still cause
a person to feel intimidated. See Elonis, 575 U.S. at 746 (Alito, J., concurring in part and
dissenting in part) (“But whether or not the person making a threat intends to cause harm,
the damage is the same.”).
Limiting true threats only to those violent threats made with the specific intent to
threaten the victim fails to protect victims from the fear of violence and instead protects
speech the defendant knows has a substantial and unjustifiable risk of causing that fear. A
defendant who recklessly communicates threats of violence “necessar ily grasps that [she]
is not engaged in innocent conduct”; instead, she recognizes the substantial likelihood that
someone could “regard [her] statements as a threat, but [she] delivers them anyway,”
without justification. Id. at 745–46. Indeed, protecting this type of speech has a corrosive
effect on society because it allows bullies who espouse violence to intimidate others,
potentially stifling public discourse.
Finally, and perhaps most importantly, this result is consistent with Supreme Court
precedent, as well as our precedent, concerning other categories of unprotected speech.
We have recognized that “the recklessness standard represents the lowest level of scienter
that has thus far been affirmatively approved by the Supreme Court under the First
Amendment.” Mauer, 741 N.W.2d at 114. For example, a legislature can criminalize
fighting words likely to provoke an ordinary person “without proof of an intent to provoke
a violent reaction”; a state can convict a defendant “of mailing obscenity . . . without proof
18
that [she] knew the materials were legally obscene”; and a person can be subject to tort
liability “for false statements about private persons on matters of private concern even if
the speaker acted negligently with respect to the falsity of those statements.” Elonis, 575
U.S. at 766−67 (Thomas, J., dissenting) (citing Cohen v. California , 403 U.S. 15, 20
(1971); Hamling v. United States, 418 U.S. 87, 120–24 (1974); Philadelphia Newspapers,
Inc. v. Hepps, 475 U.S. 767, 770, 773–775 (1986)).
Concluding that true threats require specific intent would make true threats of
violence more protected than several other categories of unprotected speech. Considering
the serious and violent nature of true threats, the harm and intimidation true threats cause,
and the corrosive effect true threats have on society even when the violent threats are made
recklessly, requiring a higher level of specific intent for speech in this unprotected
category—and not for other areas of unprotected speech —makes little sense. See Elonis,
575 U.S. at 746 (Alito, J. concurring in part and dissenting in part) (“True threats inflict
great harm and have little if any social value.”). If in the context of other First Amendment
categories, a reckless standard “provides adequa te breathing space” for protected speech,
requiring proof of recklessness in the context of true threats similarly provides sufficient
protection of speech. Id. at 748.
Accordingly, we reject a specific intent requirement for true threats and hold that a
reckless state of mind is sufficient for a defendant’s violent communication to be a true
threat excluded from the protection of the First Amendment.
19
B.
Having clarified that specific intent is not required to make a communication a true
threat, we consider whether Minnesota’s threats of violence statute is limited to prohibiting
only true threats. Because the Statute has various safeguards embedded into its text and
the caselaw interpreting it, we conclude that it punishes only reckless speech that is a t rue
threat.
First, the Statute prohibits only a very narrow type of threat to commit an act of
serious violence. In Schweppe we defined a threat of violence as “a declaration of an
intention to injure another or [her] property by some unlawful act.” 23 7 N.W.2d at 613.
The Statute prohibits only threats to commit crimes of violence, statutorily defined as
serious felony-level offenses involving violence. Minn. Stat. § 609.713, subd. 1. These
requirements comport with Black’s statement that a true thre at expresses “an intent to
commit an act of unlawful violence.” 538 U.S. at 359. The Statute is also consistent with
Watts, which upheld a statute prohibiting threats to kill or injure the President, because it
punishes only threats of serious, unlawful violence or conduct. See 394 U.S. at 705, 707.
Second, context drives our analysis of violent threats under the Statute and narrows
its reach. Our court interprets a communication in context to determine whether it would
reasonably cause a person to apprehend that the defendant would carry out or act on the
threat. Schweppe, 237 N.W.2d at 613. Requiring courts to view communications in their
context avoids sweeping in those communications that may be threatening on their face but
that a listener could easily understand as a joke or crude political commentary, just as
Watts’s commentary was not a threat but rather political hyperbole. Watts, 394 U.S. at
20
707–08. Consideration of context not only limits the scope of the Statute, but it also
satisfies a purpose of the true threats exclusion, to protect against the fear of violence. See
R.A.V., 505 U.S. at 388.
Third, that the communication must be made in reckless disregard of causing terror
limits the scope of the Statute and satisfies another purpose of the true threats exclusion.
We defined recklessness in the context of the Statute as consciously disregarding a
substantial and unjustifiable ri sk of causing extreme fear by use of violent threats. This
requirement protects victims against the “disruption that fear [of violence] engenders”
because it prohibits communications that the defendant is aware are highly likely to cause
extreme fear. See id. And, as held above, communications made with a reckless mental
state can be true threats. Accordingly, given the safeguards that narrow the Statute
substantially, the reckless disregard part of the threats of violence statute is not overbroad,
because it punishes only reckless speech that is a true threat.
III.
Even if we were to assume that the Statute may reach a recklessly made threat that,
in its context, could be considered protected speech, Mrozinski has not met her burden of
showing that “a substantial number of [the Statute’s] applications are unconstitutional,
judged in relation to [its] plainly legitimate sweep.” 8 Washington State Grange v.
Washington State Republican Party, 552 U.S. 442, 449, n.6 (2008).

8 As a result, we have no need to reach the fourth step as to whether we can save the
statute by narrowly construing or severing language to avoid the constitutional problem.
21
Here, Mrozinski cites three ex amples to try to show that the Statute applies to
protected speech and sweeps too broadly. None of them are persuasive. Mrozinski uses
the facts of Watts as her first example of protected speech she claims the Statute prohibits.
Although Watts made a st atement that on its face may be a threat —that “[i]f they ever
make me carry a rifle the first man I want to get in my sights is L.B.J.” —when taken in
context, as required by the threats of violence statute, this commentary is not a true threat.
See Watts, 394 U.S. at 706 –08. Given that Watts was at a political rally, the audience
reacted with laughter, and the threat was conditioned on an event (enlistment) that Watts
swore would never happen—all things noted by the Supreme Court—his statement would
not cause reasonable apprehension that he would act on the threat. See id.
Mrozinski next poses the example of a person who, out of anger, makes a throat -
slashing gesture to a friend about a third party but does not make the gesture in the third
party’s presence. Mrozinski does not provide any context that clarifies whether the gesture
would cause reasonable apprehension that the actor would carry out the threat on the third
party or why there might be a substantial and unjustifiable risk of causing extreme fe ar in
the third party. Merely making a throat -slashing gesture could convey many different
meanings, and without context —which the Statute requires courts to consider —it is
difficult to conclude that the gesture was made with reckless disregard of causing terror.
The last example Mrozinski mentions is a Minnesota Twins fan yelling “kill the
ump” after a bad call. The State persuasively rebuts this example by comparing it to
Watts’s statement; the fan’s statement would not likely create a reasonable appreh ension
that the fan would act on the statement. Additionally, it would be difficult to prove that
22
the fan acted recklessly because the State likely could not show that the fan was aware of
a substantial and unjustifiable risk that the statement would cause the umpire to feel terror.
Even if some examples of reckless but protected speech covered by the Statute exist,
courts cannot conclude that a statute is substantially overbroad simply because “one can
conceive of some impermissible applications.” United States v. Williams , 553 U.S. 285,
302 (2008) (internal quotation marks omitted). Given the statutory safeguards listed above
that narrow the Statute’s reach —that the threat be considered in context, that it refer to a
violent crime, that the defendant be conscious of a substantial and unjustifiable risk of
causing extreme fear and disregard the risk —few situations of reckless but protected
threats would be swept up in criminal prosecutions. Accordingly, the Statute is not facially
overbroad.
CONCLUSION
For the foregoing reasons, we affirm the decision of the court of appeals.
Affirmed.

D-1
D I S S E N T
THISSEN, Justice (dissenting).
“True threats” is a category of speech that is not protected by the First Amendment.
But the boundaries of the category are ill -defined. In this case, our fundamental task is to
define the contours of the true threats category. Specifically, we must det ermine whether
the unprotected true threats category includes threatening speech or expressive conduct
where the speaker does not intend to cause extreme fear in the person to whom a threat is
directed. I conclude that the answer to that question is “No.” Accordingly, I dissent.
Appellant Chris Mrozinski’s conviction for violating Minn. Stat. § 609.713, subd. 1 (2020),
for making threat s in reckless disregard of the risk of causing terror, but without any
determination that she acted with a specific intent to cause terror, must be reversed.
A.
Section 609.713, subdivision 1, criminalizes two distinct acts. First, the statute
criminalizes a person’s declaration of an intention to injure another or their property by
(1) committing a statutorily defined cr ime of violence, where (2) the declaration, in
context, would have a reasonable tendency to create apprehension that the declarant will
carry through with or act on the threat, and (3) the declarant has the purpose of causing
extreme fear in the other person.1 Minn. Stat. § 609.713, subd. 1. This is a specific intent
crime. The State must prove the speaker had the intent to cause the target of the speech

1 The statute also criminalizes a threat made with the purpose of (or in reckless
disregard of the risk of) causing the “evacuation of a building, place of assembly, vehicle
or facility of public transportation or otherwise to cause serious public inconvenie nce.”
Minn. Stat. § 609.713, subd. 1. That provision of the statute is not before us.
D-2
extreme fear. 9 Henry W. McCarr & Jack Nordby, Minnesota Practice—Criminal Law
and Procedure § 44:3 (4th ed.) (“Language in a statute that a prohibited act be done ‘for
the purpose of’ a particular result creates a specific intent crime.”); State v. Wilson , 830
N.W.2d 849, 853–54 (Minn. 2013) (statutory language that a prohibited act be done “for
the purpose of” a particular result creates a specific intent crime). Under United States
Supreme Court precedent, criminalizing this conduct does not violate the First
Amendment: a threat to use criminal violence to injure another when made with the
purpose or intent of causing extreme fear in the person to whom the speech is directed is a
constitutionally unprotected “true threat.” See Virginia v. Black, 538 U.S. 343, 360 (2003).
Second, the statute criminalizes a person’s declaration of an intention to injure
another or their property by (1) committing a statutorily defined crime of violence, where
(2) the declaration, in context, would have a reasonable tendency to create apprehension
that the declarant will carry through with or act on the threat, and (3) the de clarant acts in
conscious disregard of a substantial and unjustifiable risk (“reckless disregard”) that her
words or expressive act will cause extreme fear in the other person. Minn. Stat. § 609.713,
subd. 1. This is not a specific intent crime. A perso n can be convicted of the crime even
when she does not have any purpose to cause extreme fear in the target of the speech.
Indeed, because the statute separately criminalizes speech made with precisely such a
purpose, this second crime applies only if the person does not have a specific intent to cause
extreme fear in the target of the speech. Neither the specific intent crime nor the reckless
disregard crime requires the target of the threat to experience extreme fear. State v.
Schweppe, 237 N.W.2d 609, 614 (Minn. 1975).
D-3
Further, to be convicted of the “reckless disregard” crime, the speaker must
subjectively recognize or believe that her words have the potential —a substantial and
unjustifiable potential, but still just a potential—to cause a reasonable person extreme fear.
The subjective portion of this standard is an inquiry into the speaker’s awareness of a
substantial and unjustified risk. But because the risk concerns how a person other than the
speaker will react, the question of what constitutes a substantial and unjustifiable risk must
be judged against an objective standard: would a reasonable observer consider it likely that
the threat would cause extreme fear in a reasonable person to whom the threat is directed.2

2 There are multiple reasonable persons milling around in section 609.713,
subdivision 1First, there is the reasonable third-party observer who would recognize that a
substantial and unreasonable potential exists that the threat would cause extreme fear in the
person at whom the threat is directed . Second there is the reasonable person at whom a
threat is directed. That reasonable person is used to determine whether a threat was made
in the first place. True threats are, of course, a subset of all threats. The statute applies
only to communications that, in context, have “a reasonable tendency to create
apprehension that its originator will act according to its tenor.” Schweppe, 237 N.W.2d at
613 (citation omitted) (internal quotation marks omitted). Communications that do not fall
into that category are not threats under the statute. That requirement narrows the scope of
the statute.
As a bit of foreshadowing, the distinction between whether a reasonable person
would feel apprehensive because of the communication and whether the speaker intended
to make the person feel that way is important. In the context of the central question in this
case—defining the boundari es of the true threat s categorical exemption from First
Amendment protection —the question of reasonable tendency to create apprehension is
relevant to the existence of harm. As we shall see, however, the existence of harm is merely
the precondition that may justify classifying a type of speech —threatening speech —as
unprotected speech. If speech is not threatening, by definition, it cannot be an unprotected
true threat. But not all speech that is harmful because it is threatening is unprotected by
the First Amendment. The question of whether speech that is harmful because it threatens
criminal violence falls in the generally protected category of unprotected true threats turns
on the speaker’s purpose in making the threat and not on how the individual at whom the
threat is directed perceives the threat. See Megan R. Murphy, Comment, Context, Content,
Intent: Social Media’s Role in True Threat Prosecutions , 168 U. Pa. L. Rev. 733, 744 –46

D-4
In other words, under the s tatute, a speaker may be criminally convicted for threatening
speech or other expressive conduct even when the speaker not only lacked any purpose to
(specific intention of) causing extreme fear in the listener , but did not even think that her
speech had the potential to cause the listener to experience extreme fear, as long as she
perceived that some other fictive person—an objective observer—would perceive a
substantial possibility that the speech or expressive conduct would cause the target of the
threat to experience extreme fear. This is an attenuated mess of a standard for criminal
liability in that it is confusing to apply and seemingly straddles an odd line between
recklessness (requiring proof of a substantial and unjustifiable risk) and negli gence
(ultimately judged from the perspective of an objective observer).
The Supreme Court of the United States has never said that this type of speech —a
threat to use criminal violence to injure another made without the specific purpose or intent
of causing extreme fear in the person to whom the speech is directed—falls within the true
threats category of unprotected speech. Cf. Perez v Florida , 137 S. Ct. 853, 855 (2017)
(Sotomayor, J., concurring in denial of petition for writ of certiorari) (stating that Supreme
Court precedent “strongly suggest[s] that it is not enough that a reasonable person might
have understood the words as a threat—a jury must find that the speaker actually intended
to convey a threat”). On the other hand, the Supreme Court has a lso never said that this
type of speech does not fall within the true threat s category of unprotected speech. I
disagree with those courts that have concluded that Black or other Supreme Court precedent

(2020) (drawing distinction between role of objective observer and subjective intent of
speaker in threat statutes for purposes of the true threats doctrine).
D-5
requires specific intent to cause fear in the person to whom the speech is directed.
Consequently, I turn to first principles.
B.
Generally, speech and expressive conduct is protected by the First Amendment and
cannot be prohibited or criminalized. In re Welfare of A.J.B., 929 N.W.2d 840, 846 (Minn.
2019). However, certain categories of speech that are “of such slight social value as a step
to truth that any benefit that may be derived from them is clearly outweighed by the social
interest in order and morality” are not protected. Black, 538 U.S. at 358 –59 (citations
omitted) (internal quotation marks omitted).
The Supreme Court defined true threats as “encompass[ing] those statements where
the speaker means to communicate a serious expression of an intent to commit an act of
unlawful violence to a par ticular individual or group of individuals.” Id. at 359. “The
speaker need not actually intend to carry out the threat.” Id. Rather, the focus is on the
speaker’s purpose in creating fear in the target of the threat. The true threats categorical
carve-out from First Amendment protection is justified by the fact that speech that threatens
violence may undermine the social interest in order and morality. As the Black court noted,
the fear engendered by threats of violence disrupts the social order. Id. Moreover, when a
person threatens violence, actual violence often follows. That also undermines the social
interest in order. R.A.V. v. City of St. Paul, 505 U.S. 377, 388 (1992) (noting that allowing
criminalization of true threats p rotects people “from the possibility that the threatened
violence will occur”).
D-6
Certainly, not all statements threatening violence disrupt the social order to the same
extent. The seriousness of the risk to the social order turns on whether, under the
circumstances, the persons to whom the threat is directed believe that the threat will be
carried out. We recognized this point when we held that section 609.713, subdivision 1,
applies only to communications that, in context, have “a reasonable tendency to create
apprehension that its originator will act according to its tenor.” Schweppe, 237 N.W.2d at
613 (citation omitted) (internal quotation marks omitted); see, supra, n.2.
The existence of harm and disruption of the social order does not end the ana lysis.
If it did, then any expression meaningfully (in the Schweppe sense) threatening violence
would be unprotected by the First Amendment. But that is clearly not the case. The same
argument can be made about a statement meaningfully threatening violence made without
due care for whether it will elicit fear or, indeed, a statement made without any knowledge
or awareness whatsoever that it will elicit fear. But no one contends that a negligence or
strict liability standar d properly defines the contours of unprotected true threats.
Accordingly, it is not enough to say that because a violent statement made without specific
intent can still elicit fear and cause harm it passes constitutional muster.
We must also consider th e extent to which the speech has value as a step toward
truth. This part of the analysis reflects our basic understanding that the First Amendment
protects speech from government interference because a “free trade in ideas” is a necessary
precondition of a functioning democratic republic. See Abrams v. United States, 250 U.S.
616, 630 (1919) (Holmes, J., dissenting). Allowing a person to test her ideas and attempt
to persuade fellow citizens to care about an issue or enact change through vigorous
D-7
discussion is how a government of the people makes policy that reflects our collective
wisdom and holds those in power accountable. That process may be messy. It requires us
to tolerate the expression of ideas that we find distasteful or discomforting. Texas v .
Johnson, 491 U.S. 397, 414 (1989); Abrams, 250 U.S. at 630 (Holmes, J., dissenting). And
these ideas are often expressed in crude, inexact, and offensive ways. Watts, 394 U.S. at
708. Further, the First Amendment ’s protections extend beyond expression s “touching
upon a matter of public concern.” Connick v. Myers, 461 U.S. 138, 147 (1983).
That said, First Amendment jurisprudence deems some forms of persuasion to be
illegitimate because they do not persuade in a way that leads to the truth. That is certainly
true for some expressions threatening violence. For instance, if I voice agreement with a
speaker—and give up my contrary idea—solely because I am afraid the speaker might beat
me up, our discussion has moved us no closer to the truth. On the oth er hand, I may be
persuaded by language threatening violence because such language captures the speaker’s
passion about and commitment to her idea. Such heightened language can crystalize the
stakes at issue in a particular controversy so as to change hea rts and minds by breaking
through the cacophony of competing voices in the public square.
Indeed, the tension reflected in those two situations underlies the Supreme Court’s
decision in Watts. Watts spoke at a 1966 political rally during the Vietnam War. 394 U.S.
at 706. As part of a larger discussion, the 18-year-old Watts told the group that he received
his draft card and had to report for his physical the following Monday. Id. He said: “I am
not going. If they ever make me carry a rifle the fir st man I want to get in my sights is
L.B.J.” Id. He was charged with and convicted of the felony of “knowingly and
D-8
willingly . . . [making] any threat to take the life of or to inflict bodily harm upon the
President of the United States.” Id. at 705 (alteration in original).
The Supreme Court held that the conviction violated the First Amendment. The
Court did not strike down the statute as unconstitutional. Id. at 707 (“Certainly the statute
under which [Watts] was convicted is constitutional on its face.”). The Court
acknowledged that certain threats to kill or harm the President could be criminalized
consistent with the First Amendment because of the social order interest in “protecting the
safety of [the President] and in allowing him to perform his duties without interf erence
from threats of physical violence.” Id. Rather, employing a constitutional avoidance
analysis, the Court concluded that the word “threat” in the statute did not mean all threats
against the President. Id. at 707–08. The Court reasoned:
[A] statute such as this one, which makes criminal a form of pure speech,
must be interpreted with the commands of the First Amendment clearly in
mind. What is a threat must be distinguished from what is constitutionally
protected speech. . . .
We do not believe that the kind of political hyperbole indulged in by
[Watts] fits within that statutory term. For we must interpret the language
Congress chose against the background of a profound national commitment
to the principle that debate on public issues should be uninhibited, robust,
and wide -open, and that it may well include vehement, caustic, and
sometimes unpleasantly sharp attacks on government and public officials.
The language of the political arena, like the language used in [the private
setting of] labor disputes is often vituperative, abusive, and inexact. We
agree with [Watts] that his only offense here was “a kind of very crude
offensive method of stating a political opposition to the President.” Taken
in context, and regarding the expressly conditio nal nature of the statement
and the reaction of the listeners, we do not see how it could be interpreted
otherwise.

Id. (internal citations omitted) (internal quotation marks omitted).
D-9
What is critical about Watts for purposes of this case is that the Sup reme Court’s
assessment of whether Watts’ s threatening speech could be constitutionally criminalized
turned on the speaker’s purpose in using words that threatened criminal violence. In
conducting its First Amendment analysis, the Watts Court gave no consideration as to
whether the speech in fact interfered with the social order by interfering with the
President’s performance of his duties. The Court did not mention the impact on the
President.
The Black Court proceeded on a similar path. After acknowledging the social order
justification for ex cluding certain speech threatening criminal violence from the
presumptive protection of speech under the First Amendment, Black, 538 U.S. at 359–60,
the Court held that a state could constitutionally criminalize the burning of a cross with an
intent to intimidate any person or groups of persons, id. at 363. The Court reviewed the
history of the Ku Klux Klan’s use of cross burning, concluding that burning a cross is often
meant to intimidate or “coerce the victim to comply with the Klan’s [or some other cro ss
burner’s] wishes unless the victim is willing to risk the wrath of the Klan [in the form of
violence].” Id. at 357. Intimidating someone into compliance with one’s wishes through
threats of violence is, once again, an impermissible form of persuasion and so is not
protected by the First Amendment.
Significantly, however, a plurality of the Black Court also held that a provision of
the cross burning statute, which allowed a jury to infer intent to intimidate solely from the
act of burning of the cross, rendered the cross burning statute facially invalid because the
provision “strip[ped] away the very reason why a State may ban cross burning with the
D-10
intent to intimidate.” Id. at 365, 367. The Court noted that “a burning cross is not always
intended to intimidate. Rather, sometimes the cross burning is a statement of ideology, a
symbol of group solidarity. . . . Indeed, occasionally a person who burns a cross does not
intend to express either a statement of ideology or intimidation.” Id. at 365 –66.
Accordingly, it is apparent that the provision to allow an inference of intent to intimidate
solely from the act of burning the cross would create an unacceptable risk of the
suppression of ideas. Id. at 365 (internal quotation marks omitted) (internal cita tion
omitted). Although “[t]he act of burning a cross may mean that a person is engaging in
constitutionally proscribable intimidation, . . . that same act may mean only that the person
is engaged in core political speech.” Id. The Court concluded that the “First Amendment
does not permit such a shortcut.” Id. at 367.
As in Watts, the Black Court’s true threats analysis turned on the speaker’s purpose
in making the threatening statement. The Court’s First Amendment analysis did not turn
on the adverse impacts to the social order from threats of violence . Rather, such adverse
impacts are the characteristic that brings otherwise protected speech within the contours of
what may constitute unprotected true threat s. Indeed, the Court acknowledged that even
when the harm is real, and the social order is endangered by the arousal of anger or hatred
associated with such an awful threat of violence as a cross burning, the First Amendment
will protect speech that is not meant to persuade by way of intimidation or coercion through
violence. Id. at 366–67.
The challenge, then, is to figure out where and how to draw the line between speech
that tries to persuade in an impermissible way through intimidation or coercion based on
D-11
instilling a fear of violence and speech that tries to persuade in a permissible, if crude, way
through the expression of ideas. Further, we must approach this challenge with a concern
about the potential that any rule prohibiting or criminalizing speech may chill the
legitimate, permissible expression of ideas and the requirement that the harm to social order
caused by the speech “clearly outweigh[]” the value of allowing the speech. See id. at 358–
59 (internal quotation marks omitted).
In short, the point of the true threats categorical exemption from First Amendment
protection is our collective understanding that trying to coerce or intimidate someone to
adopt your position or demand by means of threatened criminal violence (and the
associated harm to the social order) is an illegitimate means of persuasion . It makes no
progress toward truth. Whether a communication is a true threat turns on the speaker’s
purpose and not on the listener’s reaction (although the listener’s reaction may b e
circumstantial evidence of the speaker’s purpose).
C.
With this context in mind, I return to the question before us: Does speech threatening
unlawful violence fall within the true threat s category of unprotected speech when the
speaker does not intend that the person to whom the speech is directed will believe that the
speaker will follow through on the threat?
The reason that intent is important in the true threats analysis is that it helps us draw
the line between permissible and impermissible types of persuasion; to determine whether
the speech is meant to persuade by coercion or intimidation or is meant to persuade by
D-12
expressing ideas. 3 A speaker’s intent to engender fear in listeners is a strong —indeed
decisive—signal that the speaker is trying to persuade by impermissible, unacceptable

3 The fact that the definition of true threats turns on the reason the speaker is uttering
the words makes comparisons to other types of categorically unprotected speech unhelpful
in def ining the contours of true threats. For example, defamatory statements are
unprotected speech because the speech is false and hurts another’s reputation. Whether
speech is false and hurts another’s reputation does not turn on the speaker’s reason for
making the speech. The constitutional limitations set forth in cases like Philadelphia
Newspapers, Inc. v. Hepps, 475 U.S. 767 (1986), and New York Times v. Sullivan, 376 U.S.
254 (1964), were put in place not to define the category—at common law, all false speech
that tends to injure a reputation could subject the speaker to damages liability —but rather
to put up a protective bubble around speech about public figures and public issues.
Philadelphia Newspapers, 475 U.S. at 772-73. The mens rea requirement limited the scope
of government restrictions on speech. Obscene materials are “utterly without redeeming
social importance.” Roth v. United States , 354 U.S. 476, 484 (1957). The category is
defined by considering whether the material, taken as whole, (1) appeals to the prurient
interests (exciting lustful or lascivious thoughts) as judged by an average person applying
contemporary community standards, (2) “depicts or describes, in a patently offensive way,
sexual conduct specifically defined by the applicable state law,” and (3) “lack s serious
literary, artistic, political, or scientific value.” Miller v. California, 413 U.S. 15, 24 (1973).
Like defamation, the Supreme Court’s discussion of mens rea in Hamling v. United States, 418 U.S. 87, 123 –24 (1974)—holding that the distribution of obscene materials may be
criminalized based solely on the defendant’s knowledge of the contents of the materials,
without additional proof that the defendant knew that the materials were legally obscene—
is consequently beside the point. Indeed, Hamling assumes that the purpose for making
the speech is entirely irrelevant to the question of whether the material falls into the
obscenity category of unprotected speech. 418 U.S. at 123 (“It is constitutionally sufficient
that the prosecution show that a defendant had knowledge of the contents of the materials
he distributed, and that he knew the character and nature of the materials.”). Where, as
here, the contours of the unprotected category turn on the purpose for, or intent with, which
the speech was made, mens rea plays a different , and much more important, line-drawing
role.
The incitement to violence categorical exception is perhaps more analogous to the
true threats categorical exception in that both turn on the reaction of some third person to
the speaker’s words. In the incitement context, speech likely to incite or produce the
imminent use of force by others may be criminalized only where the speaker intends to
incite or produce such action by third persons. Brandenberg v. Ohio, 395 U.S. 444, 447-48
(1969) (advocacy of force or violence may be criminalized only if “directed to inciting or
producing” imminent use of force); see United States v. Alvarez, 567 U.S. 709, 717 (2012)

D-13
means. That is the speaker’s purpose . Conversely, when a speaker does not intend the
threat of violence to engender fear in listeners, it is less clear that the speaker is trying to
persuade through impermissible means of intimidation and coercion by threat of violence;
the signal is far fuzzier.
For instance, consider a “reckless disregard” mental state such as that required under
the portion of section 609.713, subdivision 1. Such a standard criminalizes speec h when
the speaker does not intend to instill extreme fear of violence in the target of the threat.
Instead, it imposes criminal liability based on whether the speaker has subjective
knowledge that a n objective observer would perceive a substantial and unreasonable
potential that the threat would instill extreme fear of violence in the target of the threat.
The fact that a substantial possibility of instilling extreme fear is enough to establish
criminal liability creat es space for a communication where the speaker is attempting
protected speech —that is, persuasion through ideas, even if crudely and offensively
articulated—under circumstances where a fictive objective observer would perceive the
speaker to be attempting to persuade through the illegitimate means of intimidation or
coercion through threat of criminal violence. This sends too murky a signal for a speaker
to figure out in advance whether the speaker is doing the thing that is not protected by the
First Amendment (persuasion through intimidation or coercion based in criminal violence),
rather than doing the thing that is protected by the First Amendment (persuasion through
ideas, even if those ideas are crude, offensive, distasteful or discomforting, and have a

(stating that Brandenberg categorically exempts from First Amendment protection
“advocacy intended, and likely, to incite imminent lawless action”).
D-14
substantial potential to create extreme fear in the listener). Cf. Black, 538 U.S. at 366–67
(threatening speech in the form of cross burning may be protected by the First Amendment
even if it engenders a reaction adverse to the social order by “arous[i ng] a sense of anger
or hatred among the vast majority of citizens”). That conclusion is especially true when
one factors in the increased possibility that such a murky line will unnecessarily chill
legitimate speech.4

4 The process of drawing lines around the categorical exceptions to First Amendment
protection shares features with the First Amendment overbreadth doctrine. Under the
overbreadth doctrine, we allow litigants to challenge statutes as unconstitutionally
overbroad even when their own conduct could, consistent with constitutional requirements,
be punished under a more narrowly drawn statute because we fear the chilling effect on
legitimate speech. Broadrick v. Oklahoma, 413 U.S. 601, 612 (1973); see Paul T. Crane,
Note, “True Threats” and the Issue of Intent , 92 Va. L. Rev. 1225, 1276 & n.226 (2006).
I conclude that the more direct path of properly defining the scope of the true threat s
categorical exception to First Amendment protections is the better one. F irst, it is an
analytically clearer route. It is impossible to apply the four-factor overbreadth test without
defining the relevant category of unprotected speech. See A.J.B., 929 N.W.2d at 847
(noting that we must ask whether the challenged statute’s re ach is limited to unprotected
categories of speech or expressive conduct). For instance, I read the court’s opinion to hold
that section 609.713, subdivision 1, fully nestles within the court’s definition of the scope
of the true threats categorical exception. Accordingly, once the court defined true threats
as it did, there is no need to consider the additional steps in our traditional overbreadth
analysis. Second, defining the scope of the exception is more in line with the categorical
nature of those exceptions to First Amendment protection. Such exceptions are not
analyzed “as an ad hoc calculus of [whether] cost s and benefits tilt[] in a statute’s favor,”
but rather because the category is historically recognized. United States v. Stevens , 559
U.S. 460, 471 (2010). Finally, the structure of section 609.713, subdivision 1, supports a
focus on the scope of the true threats exception since it makes a sharp distinction between—
essentially creates distinct crimes addressing –two types of behavior: speech with the
purpose or specific intent of causing extreme fear and speech made without any purpose or
intent to cause terror, but rather in reckless disregard of the substantial possibility of
causing extreme fear. It makes little sense to apply the third factor of the overbreadth test
(does the statute sweep in too much protected speech along with unprotected speech) when
the categories in the statute itself are mutually exclusive. The better course is severance of
the category of protected speech.
D-15
In short, the category of unprotect ed speech called true threat s is limited to speech
that threatens criminal violence where the speaker has the specific intent or purpose to
place the person to whom the threat is directed in fear of physical violence. It is persuasion
by illegitimate means of violent intimidation and coercion rather than by the expression of
ideas, no matter how crude. Speech made without proof of such a specific intent or purpose
is not a true threat and so constitutes protected speech. Accordingly, I would hold that the
“reckless disregard” portion of 609.713, subdivision 1 —criminalizing a person’s
declaration of an intention to injure another or their property by committing a statutorily
defined crime of violence, where the declaration, in context, would have a reasona ble
tendency to create apprehension that the declarant will carry through with the threat, and
the declarant acts in conscious disregard of a substantial and unjustifiable risk that her
words or expressive act will cause extreme fear in the other person—is unconstitutional on
its face.5

5 The portion of section 609.713, subdivision 1, that criminalizes speech made “in a
reckless disregard of the risk of causing . . . terror or inconvenience” is re adily severable
from the statute. We presume that unconstitutional language is severable, “unless the valid
and invalid provisions of the statute are so essentially and inseparably connected that the
Legislature would not have enacted the valid provisions without the void language, or
where (after severance) the remaining valid language would be incomplete and incapable
of being executed.” A.J.B., 929 N.W.2d at 863 (internal citation omitted) (internal
quotation marks omitted).
Neither exception applies here. First, the distinct mental states in the statute —
specific intent and reckless disregard—are listed in the disjunctive; the requirement that a
person act with “purpose to terrorize another” is not so essentially and inseparably
connected with, and so dependent upon, the alternative mental state (“reckless disregard of
the risk of causing such terror”) that the Legislature would not have enacted the valid
provision without enacting the unconstitutional provision. Stated another way, I conclude
that the Legislature would have enacted a statute that criminalized only threats made with

D-16
I do not believe that this conclusion will seriously undermine the reach of section
609.713, subdivision 1. There are likely to be very few cases where evidence that a speaker
had a subjective awareness that there existed a substantial and unjustified potential that a
threat might cause an objective observer to experience extreme fear will not also support a
conclusion that a speaker intended or had the purpose to cause extreme fear in a person.
Both claims can be proved only by circumstantial evidence and assessment of the
credibility of the speaker. This case is a perfect example. The stipulated evidence presents
a strong case that Mrozinski had a specific intent or purpose to instill extreme fear in the
government employees to whom she sent the toe tags. But it is precisely those rare cases
where a difference may exist and create indeterminacy of the legality of speech that First
Amendment concerns about criminalizing and chilling political speech are most serious.
D.
Having concluded that the true threats category of unprotected speech is limited to
threats of violence made with a specific intent to cause fear in the target of the speech, I
must now address whether the reckless disregard portion of section 609 .713, subdivision
1, is nonetheless a permissible restriction of protected speech. Criminal prohibitions that
target speech based on its communicative content are permissible in narrow circumstances.

a purpose to cause terror if it had recognized the constitutional flaws inherent in the broader
language it enacted.
Second, a statute that reads “whoever threatens, directly or indirectly, to commit
any crime of violence with purpose to terrorize another . . . may be sentenced to
imprisonment for not more than five years or to payment of a fine of not more than $10,000,
or both,” is not incomplete and is pe rfectly capable of being executed in accordance with
legislative intent. Consequently, the portion of the statute criminalizing threats of criminal
violence made with a purpose to cause terror remains valid and enforceable.
D-17
State v. Casillas , 952 N.W.2d 629, 640 n.6 (Minn. 2020) (stating that “content -based
restrictions are not prohibited per se and . . . governmental regulation based on subject
matter has been approved in narrow circumstances” (internal quotation marks omitted)).
But “content -based restriction s are presumptively unconstitutional and may be justified
only if the government proves that they are narrowly tailored to serve compelling state
interests. Under a strict scrutiny analysis, narrow tailoring means that the statute must be
the least restrictive means for addressing the government ’s interest.” Id. at 640 (citations
omitted) (internal quotation marks omitted).
The State has an interest protecting people from experiencing extreme fear of
criminal violence from another. I conclude, however, that the part of section 609.71 3,
subdivision 1, that criminalizes threats made with reckless disregard of whether the threat
will cause extreme fear in the target of the threat is not sufficiently narrowly tailored to
survive strict scr utiny. First, the crime does not require any proof that the target of the
threat actually experience d extreme fear. Schweppe, 237 N.W.2d at 614. Such a
requirement would narrow the reach of the statute to align the crime more closely with the
harm it is intended to prevent and still achieve its purpose of harm prevention. Second, the
language of the statute is not limited to private speech and does not include specific
exceptions for categories of core public speech. Cf. Casillas, 952 N.W.2d at 643
(concluding that revenge -porn statute was sufficiently narrowly tailored because it was
limited by its terms to private speech and included several clear enumerated exceptions).
The case-by-case contextual carve-out that we grafted onto section 609.713, subdivision 1,
is insufficiently narrowing. For instance, I am not as comforted as my colleagues that a n
D-18
objective observer (filtered through the imagination of a jury or factfinding judge) would
inevitably conclude that the politically tinged speech made in Watts was merely bluster that
would not cause apprehension that the speaker would act on the threat. Finally, the portion
of section 609.71 3, subdivision 1, that constitutionally prohibits threats made with the
purpose or specific intent to cause fear in the target of the threat provides substantial
protection against the harms the statute is designed to prevent. The additional reckless
disregard crime will not add much to justify infringement upon protected speech. Cf.
Casillas, 952 N.W.2d at 643 (concluding that revenge -porn statute was sufficiently
narrowly tailored because it required specific intent and did not proscribe reckless
conduct). Indeed, those rare circumstances where a factfinder determines that a speaker is
not trying to persuade by illegitimate means, but where an objective observer may perceive
the speaker as trying to persuade by illegitimate means, are precisely the circumstances
where the government prohibition is most dangerous.
E.
I now turn to whether M rozinski’s conviction following a court trial is
constitutional. The district court found Mrozinski guilty under section 609.71 3,
subdivision 1, because she “made the threat [at issue here] in reckless disregard of the risk
of causing . . . terror.” The district court never determined that Mrozinski had a specific
intent or purpose to cause extreme fear and that she would carry through on the threat.
Consequently, the conviction must be reversed and the case remanded for a trial under the
proper standard.