Vidal v Elster; Trump too small; and Thomas too small-minded

Trump too small, t-shirtsIn Vidal v Elster 602 US __ (2024) (pdf), the US Supreme Court yesterday upheld the refusal of the US Patent and Trademark Office to register “Trump too small” as a trademark. In his opinion for the Court, Thomas J proved himself small-minded both in his approach to First Amendment analysis in general and in his approach to the fraught inter-relationship of trademark restrictions and the First Amendment in particular.

In this case, the plaintiff, Steve Elster, sought trademark protection for t-shirts featuring the slogan “Trump too small” (pictured above left). The slogan refers to a debate in the 2016 presidential primaries when Senator Marco Rubio teased Donald Trump about the size of his hands (with implications about other features). The plaintiff wanted to use it to criticize President Trump in general, and, specifically, to convey “that some features of President Trump and his policies are diminutive”. Section 1052(c) of the Lanham Act (Trademark Act, 1946) (15 USC § 1052(c)) precludes registration of a trademark that (emphasis added):

consists of or comprises a name, portrait, or signature identifying a particular living individual except by his written consent, …

As a consequence of this “names” clause, an examiner from the US Patent and Trademark Office (USPTO) refused the plaintiff’s application to register “Trump too small”, and the Trademark Trial and Appeal Board (TTAB) affirmed. However, the Court of Appeals for the Federal Circuit (26 F 4th 1328 (Fed, 2023)) viewed the plaintiff’s slogan as political speech protected by the First Amendment, and held that the names clause was unconstitutional as applied to the slogan (see, eg, Annick Runyon “Obtaining Trademark Registration for Marks Containing Political Commentary: A Look into Vidal v Elster78(4) University of Miami Law Review 1158 (2024); Michael Grynberg “The Trademark Problem of ‘TRUMP TOO SMALL'” 46 Columbia Journal of Law & Arts 47 (2022)). The Supreme Court granted cert on 5 June 2023 (pdf); the case was argued on 1 November 2023 (pdf); and audio of the argument is here. In their assessments of the oral argument, Lisa Ramsey, Jennifer Rothman, and Ronald Mann, all predicted that the Court was likely to hold that the section is viewpoint-neutral and constitutional. They were right: Thomas J for the Court held that the names clause does not violate the First Amendment. Ramsey went further, arguing that the Court should have clarified the applicable standard of scrutiny. On this hope, she will be disappointed: Thomas J took a frankly bonkers approach (though Barrett and Sotomayor JJ, concurring in the result, but sharply critical of Thomas J for differing reasons, do articulate similar potential applicable standards of scrutiny).

This was the third time in recent years that trademark restrictions upon speech had been before the US Supreme Court. Another clause in the Lanham Act (15 USC § 1052(a)) provided that no trademark “shall be refused registration on the principal register on account of its nature unless it … consists of or comprises immoral, deceptive, or scandalous matter; or matter which may disparage or falsely suggest a connection with persons, living or dead, institutions, beliefs, or national symbols, or bring them into contempt, or disrepute; …” (emphasis added). In Matal v Tam 582 US 218 (2017) (concerning The Slants‘ attempts to trademark their “disparag[ing]” name), the Court struck down the prohibition on registering disparaging matter (the “disparagement” clause). And, in Iancu v Brunetti 588 US 388 (2019) (concerning Erik Brunetti‘s attempts to trademark his “immoral … or scandalous” clothes brand FUCT), the Court struck down the prohibition on registering immoral or scandalous matter (the “immorality” and “scandal” clauses). Both of these trademark restrictions amounted to discrimination based on viewpoint and thus infringed the First Amendment. Here, however, the names clause does not discriminate based on viewpoint, and so does not suffer from the same infirmity as the disparagement, immorality or scandal clauses.

On the other hand, the restriction in the names clause relates to the topic being discussed or to the idea or message being expressed, and thus discriminates based (not on the viewpoint but) on the content of the speech. It is usually the case that content-based restritions on speech must satisfy strict scrutiny (see, eg, Thomas J’s opinion for the Court in Reed v Town of Gilbert 576 US 155 (2015), where he held that an ordinance against signs advertising religious services was not narrowly tailored to serve a compelling government interest). However, the Court in both Matal v Tam and Iancu v Brunetti had been slow to confirm that strict scrutiny was the appropriate standard of scrutiny for trademark restrictions on speech. In Matal v Tam, Alito J for the Court declined to decide whether the special intermediate standard of scrutiny for commercial speech outlined in Central Hudson Gas & Electric Corporation v Public Service Commission of New York 447 US 557 (1980) or a higher level of scrutiny such as strict scrutiny applied, on the grounds that the disparagement clause could not survive even Central Hudson review. Kennedy J, concurring, held that, because the restriction amounted to viewpoint discrimation, the strict scrutiny test applied, and it was not necessary to consider Central Hudson at all. In Iancu v Brunetti, Kagan J for the Court (in an opinion in which Thomas J joined) simply held that the immorality and scandal clauses constituted unconstitutional viewpoint discrimination, and she did not expressly consider the standard of scrutiny at all. Breyer J, concurring, expressed skepticism as to the utility of category-based tiers of scrutiny, preferring instead to see the Court’s speech-related categories not as outcome-determinative rules, but as rules of thumb (he had made a similar point in Reed (above)). 

In Matal, Thomas J, concurring, held that, when when the government seeks to restrict truthful speech in order to suppress the ideas it conveys, strict scrutiny is always the appropriate standard of review, whether or not the speech in question may be characterized as “commercial”. In other words, he would have rejected the special Central Hudson standard of scrutiny for commercial speech (see also his concurrences in Lorillard Tobacco Co v Reilly 533 US 525 (2001) and 44 Liquormart Inc v Rhode Island 517 US 484 (1996)). It might, perhaps, have been expected, therefore, that Thomas J would have taken advantage of the opportunity presented in Vidal v Elster to establish that strict scrutiny was the appropriate standard of scrutiny for trademark restrictions on speech, and perhaps even to overrule Central Hudson and apply strict scrutiny to all commercial speech cases. But he did not do this. Instead, he took a different – and frankly irresponsible – approach that he had developed in his opinion for the Court in New York State Rifle and Pistol Association v Bruen 597 US 1 (2022) (a case that needed re-interpretation in its own terms; see United States v Rahimi 603 US __ (2024) (pdf)) in the contex of the Second Amendment, where he rejected the application of any standard of scrutiny such as strict or intermediate scrutiny, instead assessing the lawfulness of the restriction by scrutinizing whether it comported with history and tradition.

In a series of posts on this blog, I viewed with horror (here, here, and here) the prospect of the application of this nutty, faux originalist, approach to the context of the First Amendment. But that is precisely what Thomas J has now done in Vidal v Elster. Pointing out that the Court had not decided whether heightened scrutiny extends to a content-based trademark restriction, he observed that trademark rights have always coexisted with the First Amendment, and that the inherently content-based nature of trademark law has never been a cause for constitutional concern. I note, en passant, that this argument would have doomed the development of the standard in New York Times Co v Sullivan 376 US 254 (1964). This is, no doubt, intentional: Thomas J has recently been very critical of that development, and this opinion is all of a piece with his ongoing project of dismantling modern US defamation law. However, the other members of the Court had reaffirmed Sullivan in Counterman v Colorado 600 US 66 (2023), so it strikes me as odd (to say the least) that many of them joined Thomas J’s opinion here. In any event, he concluded that the history and tradition of restricting trademarks containing names is sufficient to conclude that the names clause is compatible with the First Amendment.

Kavangh J concurred, but observed that a viewpoint-neutral, content-based trademark restriction might well be constitutional even absent the historical pedigree relied on by Thomas J. Barrett J concurred in the result, but she was chary of relying upon history and tradition in this case. She held that Thomas J had neither explained why hunting for historical forebears on a restriction-by-restriction basis is the right way to analyze the constitutional question nor established an historical analogue for the names clause. Ouch! As Rothman shows, the history here is difficult to navigate. Indeed she is equally critical of Thomas J’s faux orginalism in Rahimi (above); and these critiques, from a far more substantial originalist than Thomas J, ought to be shattering blows to his approach, but I doubt they will be. Nevertheless, this difference between Thomas and Barrett JJ in Vidal tells us far more about the Court’s originalist jurisprudence than it does about trademark law. It is a pity that Barrett J did not discuss Rothman’s scholarship (though Thomas J cited it, briefly). However, the misguided appeal to history and tradition once displaced here, Barrett J would instead have applied a specialist standard of scrutiny, holding that content-based criteria for trademark registration do not abridge the right to free speech so long as they reasonably relate to the preservation of the trademark owner’s goodwill and the prevention of consumer confusion; and she concluded that the names clause passed muster under this test. Sotomayor also J concurred in the result, but she rejected any reliance at all upon speculative history and tradition. Instead she held that withholding benefits for content-based but viewpoint-neutral reasons does not violate the First Amendment when the applied criteria are reasonable and the scheme is necessarily content based, and concluded that this was the situation here. 

The tests proposed by Barrett and Sotomayor JJ are not dissimilar, and are not a million miles away from the three-pronged scrutiny test proposed by Lisa Ramsey (here, here, here, and here) to evaluate the constitutionality of trademark laws:

First, courts should identify the goals of this specific trademark law that are important government interests. Second, they must confirm this trademark law directly and materially furthers those trademark goals. Third, it is critical for courts to discuss whether and how the trademark law endangers free speech, and ensure the law does not suppress or chill protected expression more than necessary to further those goals.

I therefore look forward to what Ramsay has to say about Vidal.[*] It is a pity that neither Barrett J nor Sotomayor J cited to her scholarship. Had their approaches commanded a majority, they would at least have been consistent with existing First Amendment jurisprudence on tiers of scrutiny, even if they would have further fragmented it by adding yet another specialist tier. Of course, this would not have been the first time that the Court would have been content to apply a specialist standard of scrutiny in an intellectual property context. In San Franciso Arts & Athletics, Inc v United States Olympic Committee 483 US 522 (1987) Powell J held that statutory protection for the word “Olympic” and for various Olympic symbols were not broader than Congress reasonably could have determined to be necessary to promote the USOC’s activities.

The Ramsey/Barrett/Sotomayor tests are directed to the question of the compatibility of trademark regulations with the First Amendment as a general matter, and go to the question of whether the names clause is constitutional on its face. But there is in this case an additional important factor that animated the Court of Appeals for the Federal Circuit but which is entirely absent from the opinions in the Supreme Court, and it goes to the question of whether the names clause is constitutional as it was applied to Elster’s application for a trademark. A major purpose of the First Amendment is to protect the free discussion of governmental affairs (see, eg, Whitney v California 274 US 357 (1927); New York Times v Sullivan (above); National Rifle Association of America v Vullo 602 US __ (2024)). Hence, the protection of political speech is at the heart of the First Amendment, and laws that burden it are subject to strict scrutiny (Federal Election Commission v Wisconsin Right to Life, Inc 551 US 449 (2007); Citizens United v Federal Election Commission 558 US 310 (2010)). Whether the appropriate standard of review is strict scrutiny or Central Hudson, the Court of Appeals held that the refusal to register Elster’s “Trump Too Small” trademark could not be sustained because the government did not have a sufficient interest in restricting speech critical of government officials. On the other hand, on the basis of the lower standard of scrutiny in SFAA v USOC (above), Powell J noted that, whilst the SFAA’s use of the word “Olympics” was intended to convey a political statement, the prohibition on using it did not prevent the SFAA from conveying its message.

However, there is no trace of this discussion in the Supreme Court. Not only is this a pity in purely analytic terms, but it also reflects an increasing autocracy on the part of the Court, or at least of certain Justices, in failing to engage with what the Courts below have said, and in ignoring the cert questions presented by the appellant, in favour of using cases as a vehicle for other ends. Judging in good faith would have required an engagement with the question actually presented by the petitioner: “Whether the refusal to register a mark under Section 1052(c) violates the Free Speech Clause of the First Amendment when the mark contains criticism of a government official or public figure”. This is the question which the Court of Appeals answered. And it was the question upon which cert was granted. But it was neither addressed nor answered by the Supreme Court. Instead, Thomas J took advantage of the case to go off on a small-minded frolic of his own “into uncharted territory that neither party request[ed]” (as Sotomayor J put it in her judgment). It may be that the Court avoided this issue, as the utility of “as applied” versus facial challenges was subsequently considered in Moody v NetChoice 603 US __ (2024) (pdf plus revision)). Even so, the matter was raised here too, and should have been considered.

Here be dragons

When medieval cartographers reached the end of explored territory, they warned that, in the land beyond, here be dragons. By eschewing questions relating to standards of scrutiny, and instead embracing history and tradition, Thomas J has certainly reached the territory of the dragons, if not of the dinosaurs. His approach has the capacity to undermine the entire edifice upon which First Amendment analysis has been painstakingly constructed, at least since the Supreme Court began to articulate tiers of scrutiny in, eg, Abrams v United States 250 US 616 (1919) and Whitney v California (above)). Anticipating the horror with which First Amendment scholars are likely to greet this travesty of a decision, Thomas J concluded with the caveat that the decision is narrow is narrow in its scope:

It does not set forth a comprehensive framework for judging whether all content-based but viewpoint-neutral trademark restrictions are constitutional. Nor does it suggest that an equivalent history and tradition is required to uphold every content-based trademark restriction. The Court holds only that history and tradition establish that the particular restriction here, the names clause in §1052(c), does not violate the First Amendment.

Yeah, right. It is a beach-head for the history and tradition replacement of tiers of scrutiny in First Amendment analysis. It will not be the last time that Thomas J will seek to make such inroads. I will watch with mounting horror as he does so.

*Update: Lisa Ramsey has posted “Heightened Constitutional Scrutiny is Not Required for Content-Based Trademark Registration Laws That Are Viewpoint-Neutral—Vidal v. Elsteras a guest post on Eric Goldman’s blog. It is highly recommended.