In the Court of Criminal Appeals of Texas ════════════ No. PD-0310-23 ════════════
EX PARTE JOHN MORGAN STAFFORD, Appellant
═══════════════════════════════════════ On State’s Petition for Discretionary Review From the Fifth Court of Appeals Collin County ═══════════════════════════════════════
YEARY, J., filed a dissenting opinion.
The Court has an obligation, when confronted with a claim that a statute is unconstitutional on its face, to construe that statute in such a way as to avoid the unconstitutionality so long as the statutory text can reasonably bear such a construction. Ex parte Perry, 483 S.W.3d 884, 903 (Tex. Crim. App. 2016). But today the Court’s opinion seems to construe the statute in such a way as to ensure its unconstitutionality, concluding that it reaches conduct that it plainly does not, or which it STAFFORD – 2
could readily at least be read not to reach, and then finding the scope of the statute, as thus construed, to be too unfettered to satisfy strict scrutiny. See Majority Opinion at 17 (claiming that the statute “criminalizes anonymous, unsigned, and factually accurate political communications”). Along the way, the Court’s opinion completely skips the first step of a proper First Amendment free speech analysis when it fails—again, presumably as a product of its unnecessarily broad construction of the statute—to ask whether the target of the statute should even be regarded as protected speech in the first place. For these reasons and more, I must respectfully dissent. I. THE STATUTE Subsections 255.004(b) and (c) of the Texas Election Code make it a Class A misdemeanor if a “person . . .[,] with intent to injure a candidate or influence the result of an election, . . . knowingly represents in a campaign communication that the communication emanates from a source other than its true source.” TEX. ELECTION CODE §§ 255.004(b), (c). “Campaign communication” is defined as “a written or oral communication relating to a campaign for nomination or election to public office or office of a political party or to a campaign on a measure.” TEX. ELECTION CODE § 251.001(17). Nowhere in the statutory scheme is there to be found a definition of the phrase “a source other than its true source.” That being the case, the Court has wide latitude, and indeed an obligation, to give Section 255.004(b) a narrowing construction—should that even prove necessary to preserve its constitutional integrity. The State argues that all the statute does is prohibit a person from knowingly making a false attribution within a campaign STAFFORD – 3
communication as to the source of that communication when that false attribution is specifically intended to injure a candidate or influence the result of an election. I agree. It simply does not, as the Court insists, make anonymous campaign communications a crime. 1 Moreover, the Court could easily have understood the phrase “a source other than its true source” to refer to an actual person or entity (as it plainly does), and not to include a fictitious one. To the extent that the Court might have thought that the phrase “a source other than its true source,” even as so construed, makes satirical portrayals of actual persons or entities a crime, such portrayals are only illegal under the statute if they are made “in” a campaign communication itself—not, for instance, in the pages of “The Onion.” Similarly, no reasonable prosecutor would think a Saturday Night Live spoof would constitute a “campaign communication,” even under the broad definition of Section
1 Nothing in Section 255.004(b) makes it an offense to fail to attribute
any source at all within a campaign communication. That the Court today insists that it does is bemusing. If the statute did proscribe anonymous campaign communications, then it may well have some unconstitutional applications—at least to that extent—under McIntyre v. Ohio Elections Commission, 514 U.S. 334 (1995). That is because the Supreme Court has said that, “[u]nder our Constitution, anonymous pamphleteering is not a pernicious, fraudulent practice, but an honorable tradition of advocacy and of dissent.” Id. at 357. The statute at issue in McIntyre, however, is very, very different from the one at issue in this case. The one at issue in McIntyre plainly and expressly required that the source of a campaign communication identify himself. Id. at 338 n.3 (quoting OHIO REV. CODE ANN. § 3599.09(A)). There is no comparable requirement anywhere in the text of Section 255.004(b). At the same time, in McIntyre, the Supreme Court expressly acknowledged that “[t]he State may . . . punish fraud directly.” Id. (emphasis added). That is precisely and purely what Section 255.004(b) does by criminalizing fraudulent misattribution of the source of a campaign communication—as long as it is not misconstrued to proscribe anonymous campaign communications as well, as the Court does today. STAFFORD – 4
251.001(17). And, even if a misattribution in the form of a satirical portrayal were to appear “in” a campaign communication itself, in order to be an offense, it would have to have been perpetrated with the requisite intent. Finally, to the extent that the Court frets about the lack of an explicit exception “for social media reposts[,]” Majority Opinion at 17, once again, the statute could have been understood to only be directed at the “person” who originated the “campaign communication,” along with its false attribution—not those who may later have reposted it with no knowledge of the original misattribution. After all, the language of “source other than its true source” first appeared in the statutory scheme in 1975, well before the advent of social media. Acts 1975, 64th Leg., ch. 711, § 12, p. 2269, eff. September 7, 1975. If the Court simply construed the statute in this reasonable and plausibly narrow fashion, it could not then have concluded that the Legislature failed to adequately tailor its proscription to the specific evil it apparently intended to address. Indeed, the Court’s overbroad reading of the statute causes it to misidentify the specific evil the statute manifestly targets and to conclude that the statute is therefore too radically off-the-rack to withstand constitutional scrutiny. As construed by the Court today, Section 255.004(b) “is not directed at dishonest conduct and, in fact, criminalizes truthful political messages.” Majority Opinion at 17. Hardly! The statute does not criminalize the message contained in a campaign communication beyond the dishonest attribution made STAFFORD – 5
therein to “a source other than the true source.” It is, then, in short, a statute that is designed to do nothing more than to prohibit a specific kind of fraud. 2 It is otherwise completely agnostic with respect to the truthfulness of the campaign communication’s content. Misattribution of the actual source of that communication may be harmful to a candidate or political cause regardless of the veracity of its content. The statute is aimed at anyone who misrepresents the source of the message the campaign communication contains—whether that message be true, false, arguable, or indifferent—to the extent that the misrepresentation is intended to fool potential voters who would evaluate the force of the message, at least in part, according to who seems to be propagating it. That is the specific form of fraud the Court should be focusing on in gauging the constitutionality of this law. II. IS THIS SPEECH EVEN PROTECTED? And with that in mind, the Court should be further asking itself whether the “speech” that Section 255.004(b) targets is even “protected” at all according to United States Supreme Court precedents. Certain types of “speech” are subject to governmental regulation
2 In McIntyre, the Supreme Court recognized the constitutional validity
of election code provisions that are aimed at fraud, and even cited by way of contrast to an Ohio statute making it an offense, as does our Election Code Section 255.004(b), to “[f]alsely identify the source of a statement[.]” 514 U.S. at 349−50 n.12 (quoting OHIO REV. CODE ANN. §§ 3599.09.1(B)(8), 3599.09.2(B)(1)). While denouncing Ohio’s separate statutory prohibition of anonymous political speech, the Supreme Court in McIntyre made quite evident its attitude that Ohio’s statutes that directly address fraudulent political speech—as does Section 255.004(b), at least properly construed— would almost certainly pass any level of First Amendment scrutiny. Only by inappropriately glomming anonymous speech onto Section 255.004(b)’s prohibition against false attribution does the Court today render it constitutionally suspect. STAFFORD – 6
notwithstanding ordinary limits imposed by the First Amendment. “Obscenity,” “child pornography,” “defamation,” “speech integral to criminal conduct”—all fall within this category of speech that is unprotected. United States v. Stevens, 559 U.S. 460, 468−69 (2010); United States v. Alvarez, 567 U.S. 709, 717 (2012). Among those categories of unprotected speech is also “fraud.” Stevens, 559 U.S. at 468; Alvarez, 567 U.S. at 717. 3 It is at least arguable that misattributing a political message to a false source is just the sort of “fraud” that the First Amendment simply does not insulate from governmental regulation—and that would be all she wrote. 4 There would then be no need to resort to any other mode of
3Fraud is defined as “[a] knowing misrepresentation or knowing concealment of a material fact made to induce another to act to his or her detriment.” BLACK’S LAW DICTIONARY (10th ed. 2019) at 802. In Alvarez, the Supreme Court observed:
Where false claims are made to effect a fraud or secure moneys or other valuable considerations, say offers of employment, it is well established that the Government may restrict speech without affronting the First Amendment. See, e.g., Virginia Bd. of Pharmacy [v. Virginia Citizens Consumer Council, Inc.], 425 U.S. [748], at 771 (noting that fraudulent speech generally falls outside the protections of the First Amendment).
567 U.S. at 723.
4 It is true that even speech that is categorically subject to governmental
regulation, such as fraud, may still not be regulated selectively, depending upon a viewpoint unrelated to that speech. See R.A.V. v. City of St. Paul, 505 U.S. 377, 383−84 (1992) (“[T]hese areas of speech can, consistently with the First Amendment, be regulated because of their constitutionally proscribable content (obscenity, defamation, etc.)—[but that does] not [mean] that they are categories of speech entirely invisible to the Constitution, so that they may be made the vehicles for content discrimination unrelated to their distinctively STAFFORD – 7
analysis, including “strict” or “exacting” scrutiny, before declaring Section 255.004(b) to be constitutionally sound. And, because Appellant has not invoked the so-called overbreadth doctrine, there is also no need to inquire whether the statute exceeds the scope of an otherwise permissible regulation. See Majority Opinion at 4 n.1 (“The instant case is a traditional facial challenge, not an overbreadth challenge.”). 5 III. STRICT SCRUTINY If the fraudulent speech that Section 255.004(b) prohibits is not, for whatever reason, deemed to be unprotected speech, and thus categorically subject to governmental regulation, then even so, the statute in my view passes “exacting,” or even “strict” scrutiny—at least as properly construed. 6 The Court acknowledges that the State has a
proscribable content.”). Section 255.004(b) does not present an instance of such unrelated content discrimination—it does not take sides. It proscribes fraudulent source attribution without reference to the speaker’s party affiliation or the nature of, or viewpoint expressed by, the message he conveys.
5 Presiding Judge Keller joins the Court’s opinion today, but the approach she takes in her separate concurring opinion sounds to me like an overbreadth analysis. Because the majority opinion today does not purport to resolve the case on overbreadth grounds, however, I will say no more about that here. I do note, however, that in concluding that Section 255.004(b) is not sufficiently tailored to satisfy strict scrutiny, the Court takes a divide-and- conquer approach to construing it. Majority Opinion at 10−13. It analyzes each part of the statute in isolation, identifying each independent clause’s possible scope, instead of reading those provisions together and in context, in an attempt to reach a construction of the statute as a whole that precisely serves its evident anti-fraud purpose. The Court should more appropriately only consider each clause’s scope as limited by the other clauses in the statute.
6 In McIntyre, the Supreme Court held that, “[w]hen a law burdens core
political speech, we apply ‘exacting scrutiny,’ and we uphold the restriction STAFFORD – 8
compelling interest “in preventing certain dishonest conduct that is harmful to the democratic process.” Majority Opinion at 8, 10. Indeed, “[a] State indisputably has a compelling interest in preserving the integrity of its election process.” Eu v. San Francisco County Democratic Central Committee, 489 U.S. 214, 231 (1989); see also Burson v. Freeman, 504 U.S. 191, 199 (1992) (observing that the State’s interest “in an election conducted with integrity and reliability” is “obviously . . . compelling”); Doe v. State, 112 S.W.3d 532, 535 (Tex. Crim. App. 2003) (“It has been consistently recognized that there must be substantial regulation of elections if they are to be fair, honest, and orderly.”). The fatal defect in the statute, the Court believes, is that it fails the second prong of the strict scrutiny analysis: it does not employ “the least restrictive means” in attacking that dishonest conduct. See Majority Opinion at 8 (“Although the State has a compelling interest in fair and honest elections and preventing fraud on the electorate, the statute is not narrowly tailored to achieve that goal.”). But the Court is
only if it is narrowly tailored to serve an overriding state interest.” 514 U.S. at 347. The Court today applies so-called “strict scrutiny” instead. Majority Opinion at 6. For its part, the court of appeals in this case applied a “strict scrutiny” standard while at the same time noting that “exacting scrutiny” has had “varying and inconsistent applications” over the years. Ex parte Stafford, 667 S.W.3d 517, 224 n.2 (Tex. App.—Dallas 2023). The court of appeals found it unnecessary to resolve the inconsistency it detected because “application of the exacting scrutiny would not alter our conclusion that the statute is not narrowly tailored.” Id. The difference may be, the court of appeals thought, that strict scrutiny, unlike exacting scrutiny, seems to involve the so-called “least restrictive means” standard in determining whether a statute employs a sufficient degree of narrow tailoring. Id. The Court today does not address this distinction. Because I agree with the court of appeals that it ultimately does not matter to the bottom-line which standard is used, however, I will not “parse the differences” either. See Stafford, 667 S.W.3d at 224 n.2. (quoting McCutcheon v. FEC, 572 U.S. 185, 199 (2014)). STAFFORD – 9
only able to conclude that the statute fails to employ the least restrictive means because it 1) misconstrues the scope of the statute, and 2) focuses on the content of the underlying message itself rather than on the fraudulent attribution of source. I will address these mistakes in turn. A. Misconstruing the Scope of the Statute The Court construes the statute to reach anonymous campaign communications as well as “imitation, impersonation, sarcasm, parody, surrogacy, or pen names[.]” E.g., Majority Opinion at 17. 7 Because this reading of the statute means that many more campaign communications will be banned than if the statute were read to simply prohibit (as I think it does) only false attributions of source, and nothing more, the Court deems it insufficiently narrow. The Court insists that the State’s construction of the statute (with which I agree) “is not accurate.” Majority Opinion at 16. But what is “not accurate”—indeed, what is inexplicably inaccurate—is the Court’s own understanding of the statute. On its face, Section 255.004(b) simply makes it an offense to positively misidentify the source of a campaign communication—to positively “represent[] in” the campaign communication itself that it emanates from “a[n actual and real] source other than its true source.”
7 It seems to me that it is the Court itself that makes the statute susceptible to targeting these forms of speech by affirmatively misconstruing it, not the statute itself. To be sure, these forms of speech are no more fraudulent than the Supreme Court found anonymous speech to be in McIntyre. See 514 U.S. at 357 (“[A]nonymous pamphleteering is not a . . . fraudulent practice[.]”). But an interpretation of Section 255.004(b) that does not reach these forms of speech, and only targets the specific act of positively misattributing the source of a campaign communication with the requisite intent (that is to say, fraudulently), is more than plausible and would serve to preserve the constitutionality of the statute. STAFFORD – 10
Here is what the Court should say about Section 255.004(b) if it were concerned (as it should be) with construing it correctly, and with an eye towards preserving its constitutionality: Anonymity: The Court could explain that simply failing to identify any source whatsoever of a campaign communication does not amount to positively representing a false source, as the statute plainly (and plainly only) proscribes. It does not violate the First Amendment, under McIntyre v. Ohio Election Commission, 514 U.S. 334 (1995), on that account.
Pen Names: The Court could also acknowledge that a too-literal focus on the terms in the statute might lead some to misconstrue its terms to include even the use of a pen name employed to shield the “true identity” of the source. But if we are to construe Section 255.004(b) with a mind to preserving its constitutionality rather than (as the Court seems to do) to undermine it, then we could easily understand the phrase “a source other than its true source” more narrowly than the Court does, in keeping with the reasonably evident (and not unconstitutional) meaning of the statute, to refer only to a false attribution to an actually existing person or entity, and not to an obviously fictitious pseudonym.
Parodies: And the Court could explain that any satirical representation that pretends to be a “source” other than the satirist himself, before it may be deemed criminal under Section 255.004(b), must occur within the confines of a campaign communication itself, and must, moreover, be perpetrated with the requisite fraudulent intent, as opposed to one merely included for comic effect, ridicule, or mockery. 8
8 “Parody” means “[a] literary or artistic work that broadly mimics an author’s characteristic style and holds it up to ridicule.” THE AMERICAN HERITAGE DICTIONARY OF THE ENGLISH LANGUAGE 954 (1975). STAFFORD – 11
An interpretation of the statute that thus confines its reach within the confines of our Constitution would serve to suppress far fewer campaign communications than the Court’s wildly overbroad construction, and it should not be deemed violative of the First Amendment because it is insufficiently restrictive when an appropriately restrictive construction is readily available. B. Focusing on the Underlying Speech Instead of the Fraud Time after time, the Court’s opinion claims that Section 255.004(b) “criminalizes truthful political messages.” E.g., Majority Opinion at 16. 9 It is certainly true to say, I think, that a campaign communication that carries a fraudulent attribution of source is criminalized, even though its underlying political message may be true. It is not criminalized because of the incorporated political message (whether true or not), however, but because of the fraudulent attribution. Section 255.004(b) is unconcerned with the veracity of the political message itself. In that sense, at least, it is content-neutral: It criminalizes the fraud regardless of the truth of the attendant political message. The fraud it seeks to suppress comes from the fact that the actor would have the electorate believe that a political opponent, or other recognizable public figure or association, advocates for the incorporated political message when that is not the case. Properly construed, the
9 See also Majority Opinion at 14 (claiming that Section 255.004(b) “criminalizes protected anonymous and truthful political speech”); id. at 17 (“It criminalizes anonymous, unsigned, and factually accurate political communications.”) (all emphases added). STAFFORD – 12
statute will only suppress the accompanying political message—true, false, arguable, or indifferent—when the campaign communication containing that political message falsely attributes it to another actual source, with the specific intent to misinform the electorate. It makes no sense for the Court to focus on the truthfulness of the attendant political message as part of what makes Section 255.004(b) insufficiently narrowly tailored while refusing to acknowledge the obvious and legitimate anti-fraud purpose of the prohibition against the false attribution of the source. C. Least Restrictive Means In McIntyre, while concluding that Ohio could not constitutionally restrict anonymous political speech in the name of protecting the electorate from fraud, the Supreme Court contrasted Ohio statutes that were not challenged in that case and that were more specifically aimed at prohibiting fraudulent political speech. 514 U.S. at 349−50 & n.12. Those statutes made it an offense, during the course of political campaigns, to “[f]alsely identify the source of a statement[.]” Id. Although the Supreme Court refrained from directly evaluating the First Amendment acceptability of those statutes, id., its clear implication was that such straightforward anti-fraud provisions, such as the statute at issue here, are constitutionally sound because they are appropriately tied to Ohio’s compelling interest in avoiding fraudulent political speech. Section 255.004(b) may not be quite as succinct or direct in its prohibition against fraudulent misattribution of the source of a campaign communication as the language in the Ohio misattribution- STAFFORD – 13
of-source statutes. 10 But it is readily susceptible to a statutory construction (eminently more natural than the one applied by the Court today) that limits it to the same narrow range of conduct as is found in those statutes. And as so construed, it is hard to imagine how Section 255.004(b) could more restrictively serve the Legislature’s compelling anti-fraud interest. IV. THE COURT INSULATES ITSELF FROM REVIEW AND PUTS OUR STATE AT RISK
The way I see it, the Court has taken what otherwise was a perfectly constitutional statute and, on its own strength, construed the statute so broadly as to implicate serious First Amendment questions.
10 The Court deems Section 255.004(b) insufficiently narrow, in part,
because it perceives another statute, Section 255.005(a), to more efficaciously accomplish the same end. Majority Opinion at 13−14 (quoting TEX. ELECTION CODE § 255.005(a)). But I agree with the State, once again, that these distinct provisions target somewhat different “dirty tricks”: The former (Sec. 255.004(b)) makes it an offense to fraudulently misattribute the source of the campaign communication while the latter (Sec. 255.005(a)) makes it an offense for the speaker to misidentify himself in that communication (while not necessarily misidentifying the source of the communication itself), with fraudulent intent. Suppose Section 255.005(a) had been challenged as unconstitutionally broad in a case that had come to us before this one. Would the Court have held that it was in fact unconstitutional because of the existence of Section 255.004(b)? There could potentially be occasions where a particular speaker violates both provisions at once, but that does not necessarily make one provision wholly redundant or overbroad. Indeed, it is not unusual to find overlapping provisions in statutes proscribing fraudulent conduct. Clinton v. State, 354 S.W.3d 795, 802 (Tex. Crim. App. 2011). The Ohio statutes cited approvingly in McIntyre likewise make it an offense both to “[f]alsely identify the source of a statement,” and to “issue statements under the name of another person without authorization[.]” 514 U.S. at 349 n.12 (quoting OHIO REV. CODE ANN. §§ 3599.09.1(B)(8), 3599.09.2(B)(1)). There was no suggestion in McIntyre that either of these provisions was constitutionally suspect because of the co-existence of the other. STAFFORD – 14
In assessing the First Amendment validity of a state statute, however, the United States Supreme Court will ordinarily deem itself bound by the authoritative construction of a statute from that state’s highest court. E.g., R.A.V. v. City of St. Paul, 505 U.S. 377, 381 (1992). It will evaluate the constitutional validity of such a statute under the assumption that it means no more and no less than what the state’s high court says it means. This means that, in the view of the Supreme Court, the statute has the meaning that we assign to it, and none other. Today, however, the Court unnecessarily construes Election Code Section 255.004(b) so expansively that the Supreme Court would almost have to agree it is unconstitutional. Unfortunately, under the Court’s overly broad construction, it is much less likely that the Supreme Court would grant any petition by the State of Texas seeking review of this Court’s determination of Section 255.004(b)’s First Amendment validity. The Court applies its overly broad understanding of the statute even though a narrower (and likely constitutional) understanding is readily available—the one I have here proposed—and in doing so it also seems to immunize itself from higher court scrutiny. And that is why the Court’s decision is so objectionable to me. The power to determine the constitutionality of our statutes is big. It is the power to check an overreaching government’s exercise of authority that has not been granted by the sovereign people of our State. But that power wielded uncarefully can also cause a lot of damage. The Court’s decision in this case, with its expansive and overly broad construction of this law (and decision to strike it down), could cause a lot of damage. And because it is accomplished by over-construing the STAFFORD – 15
reach and meaning of this statute at the state high-court level, it also has the potential effect of insulating the Court’s decision from further review. Most concerning, however, is that, until the Legislature can meet to cure the problem, our State is now placed at risk from the very dangers and campaign “dirty tricks” that this otherwise perfectly good law sought to remedy. V. CONCLUSION I would not declare the statute at issue here to be facially unconstitutional. I would instead conclude that the statute is not unconstitutional in violation of the First Amendment, and I would reverse the judgment of the court of appeals. I would then remand the case to that court for its resolution of Appellant’s remaining issues on appeal. Because the Court does not, I respectfully dissent.
FILED: September 4, 2024 PUBLISH