OWENS, KEVIN J. v. the State of Texas

IN THE COURT OF CRIMINAL APPEALS
OF TEXAS

NO. PD-0075-24

KEVIN J. OWENS, Appellant

v.

THE STATE OF TEXAS

ON DISCRETIONARY REVIEW
FROM THE SEVENTH COURT OF APPEALS
BEXAR COUNTY

KEEL, J., delivered the opinion of the Court in which RICHARDSON,
NEWELL, WALKER, and MCCLURE, JJ., joined. PARKER, J., filed a concurring and
dissenting opinion in which SCHENCK, P.J., joined. YEARY, J., filed a dissenting
opinion. FINLEY, J., dissented.

OPINION

Appellant makes an as-applied challenge to the electronic communications

harassment statute. He argues that he was prosecuted for the content of the messages he

sent, and his messages did not fall into a historically unprotected category of speech.

The State counters that he was not prosecuted for the content of his messages, and if he
Owens—Page 2

was, it has a compelling interest in protecting people from abuse. We agree with

Appellant that he was prosecuted and punished for the content of his speech in violation

of the First Amendment. We reverse the judgments below and remand the case to the

trial court for dismissal of the charging instrument.

I. As-Applied Challenge

An as-applied challenge asserts that a statute has been unconstitutionally applied

to an individual. Bynum v. State, 767 S.W.2d 769, 774 (Tex. Crim. App. 1989). A

statute’s facial validity does not foreclose such a challenge. See Ex parte Ingram, 533

S.W.3d 887, 900 (Tex. Crim. App. 2017). Even a rule that does not address speech may

be used in a way that violates the First Amendment, and such uses can be remedied via an

as-applied challenge. Virginia v. Hicks, 539 U.S. 113, 124 (2003). The merits of such a

challenge depend on the evidence. State ex rel. Lykos v. Fine, 330 S.W.3d 904, 910

(Tex. Crim. App. 2011). The challenger must show that the statute was

unconstitutionally applied to him. Schlittler v. State, 488 S.W.3d 306, 313 (Tex. Crim.

App. 2016).

The First Amendment generally prohibits the government from prohibiting speech

or expressive conduct. R.A.V. v. St. Paul, 505 U.S. 377, 382 (1992). It forbids

government retaliation against individuals for speaking. Hartman v. Moore, 547 U.S.

250, 256 (2006). The freedom to speak without risking arrest is “one of the principal

characteristics by which we distinguish a free nation.” Houston v. Hill, 482 U.S. 451,

463 (1987). The government cannot restrict expression because of its message, ideas,
Owens—Page 3

subject matter or content. Brown v. Entm’t Merchs. Ass’n, 564 U.S. 786, 790-91 (2011)

(quoting Ashcroft v. ACLU, 535 U.S. 564 (2002)). Content-based regulations are

generally invalid but are permissible for a few, limited, traditional categories of speech.

R.A.V., 505 U.S. at 382-83. Those categories include obscenity, defamation, fraud,

incitement, and speech integral to criminal conduct. United States v. Stevens, 559 U.S.

460, 468 (2010). Although additional categories of historically unprotected speech may

yet be identified, there is no “freewheeling authority to declare new categories of speech

outside the scope of the First Amendment.” Id. at 472.

II. Content-Based Laws

Content-based laws target speech based on its communicative content. Reed v.

Town of Gilbert, 576 U.S. 155, 163 (2015). Generally, laws that distinguish favored

speech from disfavored speech based on the ideas or views expressed are content based.

Turner Broad. Sys. v. FCC, 512 U.S. 622, 643 (1994) (Turner I). Even a facially neutral

statute may be content based if its manifest purpose is to regulate speech based on its

message. Id. at 645. If it is necessary to look at the content of the speech to decide if

the speaker violated the law, the regulation is content based. Ex parte Nuncio, 662

S.W.3d 903, 917 (Tex. Crim. App. 2022) (citing Ex parte Thompson, 442 S.W.3d 325,

345 (Tex. Crim. App. 2014), and Ex parte Lo, 424 S.W.3d 10, 15 n.12 (Tex. Crim. App.

2013)).

A content-neutral restriction on the time, place, and manner of speech in a public

forum is acceptable. McCullen v. Coakley, 573 U.S. 464, 477 (2014). But it would be
Owens—Page 4

content based if authorities had to examine the content of the message to determine if a

violation occurred. Id. at 479. A law concerned with undesirable effects that arise from

“the direct impact of speech on its audience” or “[l]isteners’ reactions to speech” would

not be content neutral. Id. at 481 (citing Boos v. Barry, 485 U.S. 312, 321 (1988))

(stating that a clause regulating speech because of the potential emotive impact on its

listeners must be considered content-based). A law restricting speech because it offends

or discomfits the audience is not content neutral. Id. Put another way, “Listeners’

reaction to speech is not a content-neutral basis for regulation.” Forsyth County v.

Nationalist Movement, 505 U.S. 123, 134 (1992). If a person can avoid or mitigate the

effects of a regulation by altering his speech, the regulation is content based. See TikTok

Inc. v. Garland, 145 S.Ct. 57, 67 (2025) (“the fact that petitioners ‘cannot avoid or

mitigate’ the effects of the Act by altering their speech” confirms that the challenged

provisions do not impose a restriction or penalty based on content) (quoting Turner I, 512

U.S. at 644).

Offensiveness does not deprive communications of constitutional protection. Hill

v. Colorado, 530 U.S. 703, 715 (2000). “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 (1989). The First Amendment’s protections belong even to

those whose motives others may find misinformed or offensive. 303 Creative LLC v.

Elenis, 600 U.S. 570, 595 (2023). It protects the right to criticize and to express
Owens—Page 5

irritation or viewpoints that may upset others. Mahanoy Area Sch. Dist. v. B.L., 594

U.S. 180, 192-93 (2021). And it does so “regardless of whether the government

considers his speech sensible and well intentioned or deeply ‘misguided[]’ . . . and likely

to cause ‘anguish’ or incalculable grief[.]” 303 Creative, 600 U.S. at 586 (quoting

Hurley, 515 U.S. at 574, and Snyder v. Phelps, 562 U.S. 443, 456 (2011)). “Suffice it to

say that if protecting people from unwelcome communications . . . is a compelling state

interest, the First Amendment is a dead letter.” Hill, 530 U.S. at 748-49 (Scalia, J.,

dissenting).

Prosecution based on the content of a message is permitted only in very limited

circumstances. If a statute is content based, it must meet strict scrutiny; the statute is

presumptively unconstitutional and may be justified only if the government proves that it

is narrowly tailored to serve compelling state interests. TikTok, 145 S.Ct. at 67 (quoting

Reed, 576 U.S. at 163). Content-neutral statutes are subject to a less rigorous analysis

because they do not pose the same inherent risks to free expression. Turner Broad.

Systems, Inc. v. FCC, 520 U.S. 180, 213 (1997) (Turner II)). If a statute is content

neutral, then intermediate scrutiny applies, and the statute is upheld if it advances

important governmental interests unrelated to the suppression of free speech and does not

burden substantially more speech than necessary to further those interests. TikTok, 145

S.Ct. at 67 (quoting Turner II, 520 U.S. at 189).

A breach-of-the-peace prosecution violated the First Amendment because it was

based on the content of the defendant’s message as expressed by his jacket—“F--- the
Owens—Page 6

draft.” Cohen v. California, 403 U.S. 15, 18 (1971) (“The only ‘conduct’ which the

State sought to punish is the fact of communication. Thus, we deal here with a

conviction resting solely upon ‘speech,’ . . .”). The State proffered no compelling reason

for making the display of the offensive word a criminal offense. Id. at 26. See also

Johnson v. State, 755 S.W.2d 92, 95, 97 (Tex. Crim. App. 1988) (holding statute

unconstitutional as applied because act of flag burning was clearly “speech”

contemplated by the First Amendment), aff’d, Johnson, 491 U.S. at 420. But a

prosecution for destroying a draft card did not violate the First Amendment. United

States v. O’Brien, 391 U.S. 367, 376 (1968). Even if O’Brien spoke symbolically by

burning his draft card, doing so was not necessarily constitutionally protected. Id.

“[W]hen ‘speech’ and ‘nonspeech’ elements are combined in the same course of conduct,

a sufficiently important governmental interest in regulating the nonspeech element can

justify incidental limitations on First Amendment freedoms.” Id. The statute

criminalizing destruction of a draft card survived intermediate scrutiny because, among

other things, such destruction disrupted the government’s constitutional power to raise an

army. Id. at 377-380.

If a content-neutral statute is applied in a manner that regulates speech based on its

content, it must meet strict scrutiny to survive a First Amendment challenge. In Holder

v. Humanitarian Law Project, 561 U.S. 1, 25 (2010), the Supreme Court considered an

as-applied challenge to a statute forbidding material support to designated terrorist

organizations. The statute generally functioned as a regulation of conduct, but as applied
Owens—Page 7

to the plaintiffs, it regulated their speech based on its content; they were prohibited from

speech that “imparts a ‘specific skill’ or communicates advice derived from ‘specialized

knowledge’ . . . .” Id. at 27-28. Because it regulated the content of the speech, a more

demanding standard of scrutiny was applied. Id. at 28. Still, it passed constitutional

muster because of the government’s interest in preventing terrorism. Id. at 36-37.

The government may prohibit the intrusion into the home of unwelcome views and

ideas that cannot be banned from public dialog. Cohen, 403 U.S. at 21 (citing Rowan v.

Post Off. Dept., 397 U.S. 728, 738 (1970)) (“That we are often ‘captives’ outside the

sanctuary of the home and subject to objectionable speech and other sound does not mean

we must be captives everywhere.”). The government’s ability to regulate speech

depends on a showing that the speech invades substantial privacy interests in an

essentially intolerable manner. Id. Selective restriction of offensive speech has been

upheld when there was a captive audience or an invasion of unwanted ideas into the

home. Erznoznik v. Jacksonville, 422 U.S. 205, 209 (1975). “The right to avoid

unwelcome speech has special force in the privacy of the home[.]” Hill, 530 U.S. at 717.

We “have repeatedly recognized the interests of unwilling listeners in situations where

‘the degree of captivity makes it impractical for the unwilling viewer or auditor to avoid

exposure.’” Id. at 718 (citations omitted). The right of a person to be left alone must be

weighed against the right of others to communicate. Rowan, 397 U.S. at 736. Pitting

the First Amendment rights of speakers against the privacy rights of those who may be

unwilling listeners demands delicate balancing. Hill, 530 U.S. at 718 (citing Erznoznik,
Owens—Page 8

422 U.S. at 208). “Nothing in the constitution compels us to listen to or view any

unwanted communication” in the privacy of our own homes; we are permitted to bar

solicitors, block senders of mail, or turn off a radio or television to prevent offensive

communications from entering the home. Rowan, 397 U.S. at 737.

III. The Harassment Statute

Texas Penal Code section 42.07(a)(7) has two parts; intent and repeated

communications sent in a certain manner. It reads as follows:

(a) A person commits an offense if, with intent to harass, annoy, alarm,
abuse, torment, or embarrass another, the person:
(7) sends repeated electronic communications in a manner
reasonably likely to harass, annoy, alarm, abuse, torment, embarrass, or
offend another;

Tex. Penal Code § 42.07(a)(7).

We have upheld its facial constitutionality on grounds that it regulates conduct and

not speech. Ex parte Barton, 662 S.W.3d 876, 884 (Tex. Crim. App. 2022), and Ex

parte Sanders, 663 S.W.3d 197, 216 (Tex. Crim. App. 2022), held that the statute

prohibits non-speech conduct. Its gravamen “is the sending of repeated electronic

communications in a manner reasonably likely to harass, annoy, alarm, abuse, torment,

embarrass, or offend another[.]” Sanders, 663 S.W.3d at 215. No speech is required to

commit the crime. Id. “The statute is equally violated by the repeated sending of

communications containing expressive speech as it is by the repeated sending of

communications containing no speech at all.” Id. at 215-16. It could be violated by

repeatedly sending messages consisting of a single letter or indecipherable machine
Owens—Page 9

language. Id. at 216. The statute “does not regulate expressive conduct. Instead, it

focuses upon conduct that is not inherently expressive. If there was an intent to be

expressive, the actor would have to provide separate speech accompanying and

explaining the conduct.” Id. at 216.

Because section 42.07(a)(7) regulates non-speech conduct, Barton applied a

rational-basis test to decide whether the statute served a legitimate governmental interest.

Barton, 662 S.W.3d at 884. The statute’s interest is to punish and deter those who

invade the substantial privacy interests of another in an essentially intolerable manner.

Id. at 885. We concluded that the statute was facially constitutional. Id.; Sanders, 663

S.W.3d at 216.

The question before us now is narrower: was the statute unconstitutional as it was

applied to Appellant?

IV. Background

Appellant was convicted of harassment for sending about three dozen electronic

messages to his former therapist during a 15-week period; most were email, some were

text, and one was on Facebook. All the messages were sent to her professional, not

personal, accounts. Some messages were outside the date range alleged in the charging

instrument, and some had content only in the subject line, but they were all admitted.

Appellant objected to their admission as constitutionally protected speech and sought a

defensive instruction based on the First Amendment, but the trial court overruled him on
Owens—Page 10

both points. 1

In 2016, Dr. Lindsay Bira was a newly licensed psychologist, and Appellant

became one of her clients, meeting her weekly for therapy. He was soon dissatisfied

with the sessions, and Bira was uncomfortable with him. Bira tried to refer him to

another psychologist, but Appellant refused the referral, and after eleven sessions, he

stopped seeing Bira, cancelled his remaining sessions, and emailed Bira telling her to

never contact him again. She had no contact with him until about two years later.

On May 13, 2018, Appellant sent Bira an email that said “some very concerning,

upsetting things.” It said:

My life is just as hopeless as ever
Maybe if I had the genes that would allow me to consider a modeling career
then my life would be better, but I didn’t.
You exploited, abused, and then abandoned me. I will never give you any
more money, but if you wanted to talk to me then that would be possible.
I’m sure you have better things to do though.

Bira testified, “This was sickening. It was highly concerning, especially what I already

knew about him. And I was scared.” She did not reply to the email but forwarded it to

the San Antonio Police Department (SAPD). They told her not to block Appellant’s

1
His requested instruction read as follows:

It is a defense to the crime of harassment if a person’s speech falls into one of the
recognized categories of historically protected speech. As a general matter, the First
Amendment means the government has no power to restrict expression, because of the
message, its ideas, the subject matter, or its content and as a result, the United States
Constitution demands that content-based restrictions on speech be presumed invalid and
that the government bear the burden of showing their constitutionality. Therefore, if you
find that the speech of Kevin Owens falls into a recognized category of historically
protected speech, you must find Kevin Owens not guilty.
Owens—Page 11

email address so officers could document the messages and see if they escalated.

In Appellant’s second email, sent June 14, 2018, he rambled on about various

aspects of Bira’s personal and professional life that he found on the internet and social

media. He commented about her family, childhood, friends, boyfriends, and career. He

researched a traumatic event she experienced in college. He indicated that he knew her

personal phone number and home address. He mentioned photos of her that he found on

social media; one from her “modeling days in a see-through top,” and others related to

her dating relationships. He referred to her as “eye candy” and said he would not be

surprised if she were a prostitute. Bira said the email was “horrific and concerning.”

She felt terrified that someone would make accusations about her, would want her to

suffer, and would want her to know that he knew personal information about her and her

friends and family. She did not respond to the email. Later that day, Appellant sent

another email that said, “You have nothing to say? I’m surprised, I thought you were a

powerful woman.” Bira testified that even though her personal Facebook and Instagram

accounts were private, police advised her to block anyone with Appellant’s name on

social media.

In the next email, sent July 1, 2018, Appellant purported to revoke his agreement

with Bira’s policies related to her practice, privacy, and consent for psychological

services. Bira testified that these were standard legal intake forms that Appellant had

signed when he began therapy, and there was nothing to revoke since treatment had

already ended.
Owens—Page 12

An email sent July 3, 2018, included a comment about one of her previous

relationships along with a photo of Bira that had been posted on her boyfriend’s

Instagram page. She did not know how Appellant found out who she was dating, and

she found it “extremely concerning and terrifying.” Thirty minutes later, Appellant sent

another email calling Bira “a shitty therapist and an even worse psychologist” and said

she would “always be a terrible person.” He indicated that he was monitoring her social

media and her client website and surmised that she was more active on social media

when she could not fill her client schedule. Bira testified that her professional social

media was public, but she became uncomfortable, uneasy, and anxious about posting

anything to social media knowing that Appellant was monitoring her posts. Appellant

sent two more emails that day saying Bira had abused him, raped him, and exploited him.

Bira submitted these to SAPD as escalated contact and harassment and was again advised

not to respond.

On July 4, 2018, Appellant sent an email asking Bira for a refund of the money he

paid for the therapy sessions, saying that she did not help him, she tricked him, she

cheated him, and she owed him $1,785.

Five days later, Appellant sent an email with the subject line “You are

encouraging me to kill myself.” Bira testified that she knew Appellant was trying to get

her to respond, and she was not concerned on a clinical level, but she had to respond to

cover her bases and to make sure she was doing the right thing. She replied to

Appellant’s email from an office manager administrative account that was used for
Owens—Page 13

dealing with patients who might pose a risk. The “office manager” response to

Appellant’s email advised him to call 911, go to the emergency room, or call a suicide

hotline; it said his contact had been reported to SAPD and future contact from him would

be forwarded to police and legal personnel. Appellant replied that he was not

considering suicide and asked for the office manager’s name. Appellant sent Bira

another email saying that she abused him, she wanted him to be a slave, and he was

“raped every day.” Another message was sent to the “office manager” email address

again asking for her full legal name.

Appellant began text messaging Bira on July 10, 2018, asking for his money back

and saying she was abusive, trying to get revenge, was a terrible person, and had lied to

him and cheated him.

On July 12, 2018, Appellant emailed Bira asking when she was going to return his

money. He said, “You lied to me and you didn’t do your job. I want my money back.

You owe me. You didn’t earn it and you never deserved it.” He also used an alias to

send a Facebook message calling Bira “a terrible therapist and a shitty person” and

saying, “I want the money you owe me, and then we will be done.”

Appellant sent eight emails from July 17th to 18th in which he requested a refund,

referred to himself as a victim, and claimed Bira had abused and raped him. He

referenced talking about Bira to one of her colleagues and quoted information that had

been posted on her professional Instagram page. He included a quote about life being

stolen by fear and said, “You are the one who told me to not be afraid. How’s that
Owens—Page 14

working out for you?” He also referenced one of her favorite quotes that she often used

in public speaking, “People are disturbed not by a thing, but by their perception of a

thing.” Appellant said, “You’re just making yourself upset, so I’m not doing anything

wrong.”

From July 25, 2018, through August 25, 2018, Appellant sent seven emails

accusing Bira of being a con artist and of abusing him, tricking him, exploiting him,

sexually assaulting him, and touching him “in a sexual and inappropriate way during

therapy.” He said she violated confidentiality, and he requested his records and

referrals. He also asked Bira to find a girlfriend for him.

Bira testified that she forwarded all the messages to SAPD and that Appellant was

sent cease and desist letters from both SAPD and her attorney. She testified that the

messages made her feel scared, horrified, abused, harassed, and embarrassed, and she

was concerned for her safety. She said she had difficulty seeing patients during this time

frame because she was worried that Appellant was going to show up at her office and

harm her. She eventually stopped seeing patients in person, switched her practice to

video therapy sessions only, and moved out of state.

Defense counsel asked Bira if she felt harassed because Appellant sent her

messages or because of what the messages said. She replied, “It was repeated forced

contact from Kevin Owens, along with what he chose to say to me . . .The action of him

repeatedly e-mailing me even after two cease and desists, and me saying do not. That

act felt harassing, and also what he chose to say to me felt harassing.” She said that if
Owens—Page 15

Appellant had sent her an email initially that said, “hey, I really want to chat with you.

It’s been a while, but I have something that I want to discuss that’s lingering with me”

she would have said, “absolutely, let’s set up a call.” Instead, Appellant “chose an

illegal harassing way to approach” her to discuss his concern, so she did not reply to him.

The defense asked Bira again whether it was the fact of getting the emails that harassed

her or their content that made her feel harassed, and she replied that it was both. Bira

testified that she “felt abused from that very first email. Highly harassed.”

The defense argued that there was no evidence of Appellant’s intent to harass,

annoy, alarm, abuse, torment, or embarrass Bira; he was upset, and he wanted a refund;

he had no control over Bira’s reaction to his messages, and Bira could have blocked or

ignored the emails. The State argued that Appellant’s words showed his intent and that

freedom of speech is not a defense to harassment.

The jury found Appellant guilty of two counts of harassment. At the punishment

hearing, the defense argued that Appellant was being punished for his speech. The trial

judge responded, “Of course it’s punishment for speech . . . If he was saying good

morning, it’s not - - there has to be some content in his speech that is the problem here . .

. if you’re saying good morning in an e-mail, it’s not the same thing as calling someone a

name like a whore in an e-mail. That’s not the same thing.” The judge sentenced

Appellant to 180 days in jail and a $500 fine.

V. Court of Appeals

On Appeal, Appellant argued that he was punished due to the content of his
Owens—Page 16

messages because they caused Bira to feel harassed. Owens v. State, No. 07-23-00115-

CR, 2024 Tex. App. LEXIS 110, at *7 (Tex. App.—Amarillo Jan. 5, 2024) (not

designated for publication). The court of appeals disagreed. It said that while words

can be used to commit the offense, the conduct prohibited by the statute is distinct from

recognized categories of expressive conduct. Id. at *7-8 (citing Sanders, 663 S.W.3d at

213). Reasoning that the words used to harass Bira were integral to criminal conduct

and so outside the protections of the First Amendment, the court of appeals concluded

that Appellant failed to show that section 42.07(a)(7) operated unconstitutionally as

applied to his circumstances. Id. at *8.

The court of appeals overruled Appellant’s facial and as-applied challenges to the

constitutionality of section 42.07(a)(7).

VI. Analysis

Sending messages is an act, but the messages themselves are speech, and the

prosecution in this case was based on Appellant’s speech, not his action. It was the

content of the messages, not the manner of their sending, that drove the prosecution.

Bira called the police on receipt of the first message—not after the receipt of repeated

messages. She was disturbed by the content of the first and subsequent messages, not

merely the manner in which they were sent. She admitted that both the “repeated forced

contact” and the content of the communications were harassing. Appellant would not

have been prosecuted if his messages had expressed a different tone or message; he

would have avoided prosecution if he had said “good morning” instead of accusing Bira
Owens—Page 17

of raping him. Bira and the judge both said so; if Appellant’s first email had been

worded differently, if he had reached out politely asking to speak with her, she would

have obliged, and he would not have been prosecuted and punished. Instead, she

contacted SAPD because of the content of the first message. She “felt abused from that

very first email. Highly harassed.” As the trial judge pointed out when assessing

Appellant’s punishment, “Of course it’s punishment for speech . . . if you’re saying good

morning in an e-mail, it’s not the same thing as calling someone a name like a whore in

an e-mail.” The manner of the communications was immaterial; it was their content that

drove the prosecution.

We acknowledge Bira’s right to be free from unwelcome ideas invading her

substantial privacy rights in an essentially intolerable manner. Barton upheld the facial

constitutionality of section 42.07(a)(7) because it protects from conduct that causes such

invasions of privacy. But here section 42.07(a)(7) was used to regulate Appellant’s

speech, not his conduct. Appellant’s First Amendment right to communicate must be

delicately balanced with Bira’s privacy rights, and the scale is tipped in Appellant’s favor

in this case for three reasons.

First, there was no invasion into the home. The messages were not sent to Bira’s

home or her personal accounts, they were sent to her professional email and office phone

that she used for communicating with patients and to her professional social media

account that was public. Second, Bira was not a captive audience in this situation; she

was not powerless to avoid the messages. She could have deleted the messages without
Owens—Page 18

reading them or blocked Appellant’s email address, phone number, and social media

accounts, but she chose not to. Third, the government’s ability to regulate speech

depends on more than a simple invasion of privacy; it requires an invasion of substantial

privacy rights in an essentially intolerable manner. See Cohen, 403 U.S. at 21. Thirty-

four messages sent in a span of more than three months to publicly accessible,

commercial accounts controlled by a willing listener is no such invasion.

Because Appellant was prosecuted for the content of his messages, the statute’s

application is presumptively unconstitutional and may be justified only if the government

proves its application was narrowly tailored to serve compelling state interests. The

State makes no such showing here. Instead, the State relies on the O’Brien intermediate-

scrutiny standard and says that when a statute covers a combination of speech and non-

speech elements in the same conduct, incidental limitations on speech can be justified by

an important government interest. But Appellant’s case is distinguishable from O’Brien.

In O’Brien, the message did not matter to the prosecution. O’Brien would have been

prosecuted even if his message had been “I love the pretty colors created by burning my

draft card,” but Appellant would not have been prosecuted if his message had been “good

morning.”

Because the State focused on the intermediate-scrutiny standard, it failed to make

a showing that the harassment statute survived strict scrutiny as applied to Appellant in

this case.

VII. Other Arguments
Owens—Page 19

The State concedes that the court of appeals went too far in declaring that the

content of the messages was speech integral to criminal conduct but says the court of

appeals still reached the correct result. It argues that if Appellant’s conviction is invalid,

then the statute would be, too, because admitting the content of the messages was

necessary to prove what it contends are two elements of the crime: Appellant’s intent to

cause negative feelings and the reasonable likelihood that his messages would cause such

feelings. The State is mistaken for two reasons. First, we do not hold that the messages

were inadmissible for any purpose, and Appellant concedes that they were admissible to

show intent. But they were admitted without limitation over Appellant’s objection, so

the prosecution was based on their content, not merely the manner of their sending.

Second, the State has misread the statute. A conviction under it does not require a

showing that the messages were reasonably likely to cause negative feelings; it requires a

showing that the manner in which they were sent was reasonably likely to do so. Their

content is irrelevant to their manner of sending.

The State argues that “it is not speech at all when the actor sends repeated

electronic messages with an intent to harass.” But it supports this assertion with an

inapplicable quote from Sanders quoting Scott about the telephone harassment statute,

not this one. 663 S.W.3d at 203 (“[P]ersons whose conduct violates § 42.07(a)(4) will

not have an intent to engage in the legitimate communication of ideas, opinions, or

information; they will have only the intent to inflict emotional distress for its own sake.”)

(quoting Scott v. State, 322 S.W.3d 662, 670 (Tex. Crim. App. 2010)) (overruled on other
Owens—Page 20

grounds, Wilson v. State, 448 S.W.3d 418, 422-23 (Tex. Crim. App. 2014)). Scott

upheld the telephone harassment statute because making repeated phone calls was

noncommunicative conduct that did not implicate First Amendment free speech

protections. Scott, 322 S.W.3d 670. In contrast, Appellant’s messages intended to

communicate ideas, opinions, and information, and his prosecution was based on their

content.

The State argues that electronic harassment is non-speech conduct even if it

includes the use of words, and it is outside First Amendment protection because it does

not necessarily contain an expressive message; the speech explaining the conduct is

separate. While Sanders held that the statute restricts non-speech conduct, that was not

how it was applied here. In this case the State used the statute to punish Appellant’s

speech based on its content. Appellant was not punished for merely sending messages;

he was punished for the things he said.

Judge Parker’s concurring and dissenting opinion confounds as-applied

challenges with facial ones and seems to say that part of the statute is facially

unconstitutional for allowing conviction when the speech’s content may annoy, alarm,

embarrass, or offend. She mistakes our holding for a sufficiency of the evidence

analysis asking whether the evidence supports a conviction on a constitutional basis.

But our holding answers the question of whether Appellant was prosecuted for the

conduct of sending messages or for the content of the messages.

Judge Parker suggests that we remand for a harm analysis, but when the State has
Owens—Page 21

chosen to apply a statute in a manner that violates a defendant’s First Amendment rights,

the proper remedy is dismissal. See, e.g., Johnson, 755 S.W.2d at 97-98 (holding the

flag burning statute unconstitutional as-applied, reversing the judgments of the lower

courts, and remanding to the trial court for dismissal of the information), aff’d, Johnson,

491 U.S. at 420; Flores v. State, 245 S.W.3d 432, 443 (Tex. Crim. App. 2008)

(Cochran, J., concurring) (“If the defendant prevails on his ‘as applied’ constitutional

claim, there will be no new trial. There is only one remedy for either the trial or

appellate court: dismiss the indictment and enter an acquittal because the defendant was

convicted under an unconstitutional application of an otherwise valid penal statute.”).

The State does not get to try again to convict just because the statute could have been

applied in a different, and constitutional, manner.

VIII. Conclusion

The court of appeals erred in categorizing Appellant’s messages as speech integral

to criminal conduct outside the protections of the First Amendment. As applied to

Appellant’s case, the electronic harassment statute was a content-based restriction on his

speech, and the State failed to show that the statute was narrowly tailored to serve a

compelling State interest. The statute’s application to Appellant’s case was

unconstitutional. We reverse the judgment of the court of appeals and remand the case

to the trial court for dismissal of the indictment.

Delivered: June 4, 2025
Owens—Page 22

Publish