Evaluating the Constitutionality of Viewpoint-Neutral Trademark Registration Laws That Do Not Restrict Speech—Vidal v. Elster (Guest Blog Post)

By guest blogger Lisa Ramsey, Professor of Law, University of San Diego School of Law

The Supreme Court will likely hold in Elster that Section 2(c) is consistent with the First Amendment, but will it clarify how to balance trademark and free speech rights?

Vidal v. Elster is a trademark dispute at the Supreme Court focused on the issue of whether Steve Elster has a free speech right to register a political message (“Trump Too Small”) as a trademark for shirts. Elster claims these words invoke an exchange between former President Trump and Senator Marco Rubio from the 2016 presidential primary debate where Rubio made a crude joke about the implications of Trump having small hands after Trump called him “little Marco”. The United States Patent and Trademark Office (USPTO) refused registration of “Trump Too Small” under Section 2(c) of the Lanham Act (15 USC 1052(c)) because the phrase includes a living individual’s name without his written consent. This law protects the rights of publicity and privacy of people like politicians and celebrities who don’t want others to register their name or image as a trademark for goods or services.

On February 24, 2022, the Federal Circuit Court of Appeals held that Section 2(c) was an unconstitutional regulation of expression as applied to Elster’s proposed mark because “the government does not have a privacy or publicity interest in restricting speech critical of government officials or public figures in the trademark context—at least absent actual malice, which is not alleged here.” After the Supreme Court granted cert in Vidal v. Elster and held oral argument on November 1, 2023, most commentators believe—like me—that the Court will reverse the Federal Circuit and hold that Section 2(c) is consistent with the Free Speech Clause of the First Amendment on its face and as applied to Elster’s trademark application. Examples of trademark experts with this view include professors Rebecca Tushnet and Jennifer Rothman; attorneys Megan Bannigan, David Bernstein, Timothy Cuffman, and Jon Jekel; and news reporters from the New York Times (Adam Liptak) and Bloomberg (Greg Stohr).

Before the Elster oral argument, many trademark professors and attorneys told me they thought the Supreme Court would follow its approach in Matal v. Tam (2017) and Iancu v. Brunetti (2019) and strike down this trademark registration law as an unconstitutional regulation of speech protected by the First Amendment. I disagreed, and continue to think the Court will uphold the constitutionality of Section 2(c), but the question is what free speech doctrine(s) the Justices will use to make this determination and whether they will provide additional guidance on evaluating First Amendment challenges to trademark laws. Below I explain how the Supreme Court could apply its free speech jurisprudence in Elster in a different way than in Tam and Brunetti, and propose a First Amendment test it can use in Elster and other trademark disputes to resolve conflicts between trademark and free speech rights in a way that is consistent with the Court’s approach in the Jack Daniel’s v. VIP Products (2023) opinion and its other trademark cases.

Section 2(c) is a viewpoint-neutral trademark law. The first reason the Supreme Court is likely to use a different approach to its First Amendment analysis in Elster in that the challenged trademark registration law does not discriminate based on viewpoint. This fact makes Section 2(c) different than the viewpoint-discriminatory provisions in Section 2(a) of the Lanham Act at issue in Tam and Brunetti that banned registration of subject matter that may be disparaging, immoral, or scandalous. The text of Section 2(c) is viewpoint-neutral. It prohibits registration of a trademark that “[c]onsists of or comprises a name, portrait, or signature identifying a particular living individual except by his written consent, or the name, signature, or portrait of a deceased President of the United States during the life of his widow, if any, except by the written consent of the widow.” Moreover, at oral argument, Elster’s attorney was unable to point to any evidence that USPTO trademark examining attorneys apply the law in a manner that discriminates based on the viewpoint of the applicant or the proposed mark.

The USPTO has invoked Section 2(c) to refuse registration of phrases like “Obama Pajama” for pajamas, and not just marks that contain criticism of government officials or public figures. It is true that politicians, celebrities, and other people may be unlikely to consent to the registration of marks like “Trump Too Small” that are critical of the named individual. Still, this fact does not make Section 2(c) viewpoint discriminatory on its face or as applied, as explained in the government’s brief and the amicus brief of the International Trademark Association (INTA). INTA contends that Section 2(c) “passes Constitutional review” because this “viewpoint-neutral” law “does not create any significant or undue restriction on speech”, “Congress has a substantial basis to regulate registration of trademarks that appropriate and trade on names and associated publicity rights of recognized individuals”, and “Section 2(c) refusals permit more speech, not less.” (Note INTA argued that Section 2(a)’s ban on registration of disparaging, immoral, or scandalous matter was inconsistent with the First Amendment in the amicus briefs it filed in Tam and Brunetti.)

Section 2(c) does not restrict speech. Another reason the Court will likely uphold the constitutionality of Section 2(c) is that this trademark registration law does not restrict Elster’s political message. Questions asked by several Justices during oral argument strongly suggest they thought this fact was important. A trademark registration is not required to display a political, social, or religious message, hate speech, swear words, or other common phrases or symbols on clothing, jewelry, mugs, posters, stickers, buttons, flags, tote bags, home décor, or other types of expressive merchandise. (Moreover, this type of expression usually fails to function as a source-identifying trademark for these products, and is often refused registration by the USPTO on this ground.) Elster could apply to register another phrase or name as a trademark for shirts, such as “Too Small” or “Steve Elster”, and continue to display “Trump Too Small” prominently on the front or back of the shirts he sells under his “Too Small” or “Steve Elster” brand.

The Federal Circuit and Elster are correct that Donald Trump cannot invoke state laws protecting the rights of publicity and privacy to prevent Elster from displaying the critical political message “Trump Too Small” on clothing. Courts would find this expressive use of Trump’s name is permitted by defensive doctrines in those states’ laws, or is otherwise allowed by the First Amendment. (See the amicus brief discussing free speech limits on the rights of publicity and privacy filed by the Motion Picture Association in Elster.) The problem with Elster’s argument, however, is that Section 2(c) does not prohibit any expression. To the contrary, this trademark registration law leaves political expression and other messages that include a person’s name or image in the public domain for everyone to use—including Elster—unless another law prohibits use of the expression in this way (such as copyright law, which protects copyright in photographs, drawings, or paintings of people). If trademark applicants like Elster can register other people’s names or images as marks for clothing without their consent, the trademark registrant can invoke trademark law to chill or suppress use of this expression in the same manner by that person and competitors in situations where the named individual (here Trump) could not sue for violations of their rights of publicity and privacy.

Of course, the Lanham Act provisions in Section 2(a) banning registration of disparaging, immoral, and scandalous matter also did not restrict the speech of Simon Tam and Erik Brunetti. Perhaps the Court will acknowledge this fact in Elster, and explain that its holding in Tam and Brunetti was primarily (or solely) based on the viewpoint-discriminatory nature of the Section 2(a) laws banning registration of hate speech, swear words, and other offensive messages. The opinions in Tam and Brunetti focused on the fact that the First Amendment does not allow the government to discriminate based on viewpoint, and Justice Alito’s concurring opinion in Brunetti suggests this was his main critique of the immoral and scandalous mark provisions.

In the Supreme Court’s 8-0 Tam decision in favor of Tam, the Court (I believe wrongly) equated the denial of a registration with a restriction of speech throughout the opinion. In Brunetti, however, Justices Roberts, Breyer, and Sotomayor noted in opinions that concurred in part and dissented in part that Erik Brunetti could still display his proposed “FUCT” mark on clothing without a registration. (Note Brunetti later applied to register “FUCK” for various goods and retail store services, but the USPTO denied that application on failure to function grounds and rejected Brunetti’s argument that the Court’s First Amendment analysis in Tam and Brunetti required the USPTO to register this mark.) In addition, Justices Roberts, Breyer, and Sotomayor concluded in Brunetti that the government could ban registration of matter that is scandalous without violating the First Amendment as long as the trademark law targeted marks that are obscene, vulgar, or profane. After the oral argument in Elster, I believe most—if not all—of the Justices will note that Section 2(c) does not restrict Elster’s speech when they uphold this law as a valid constitutional regulation of expression.

Elster seeks a trademark right to restrict the speech of others. I was delighted when some Justices noted during oral argument that Elster seeks a trademark right to chill or suppress the speech of others who want to use an identical or similar phrase in the same way as him. I have been arguing for years that a trademark registration can lead to self-censorship, private censorship, and government censorship of truthful and nonmisleading expression claimed by another as a trademark. (See “Increasing First Amendment Scrutiny of Trademark Law” (2008), “A Free Speech Right to Trademark Protection?” (2016) written after the Federal Circuit’s decision in In re Tam, “Free Speech Challenges to Trademark Law After Matal v. Tam” (2018), and my guest post on Eric Goldman’s Technology and Marketing Law Blog written after the Federal Circuit’s decision in In re Elster.) Prominent free speech advocates and trademark attorneys agree that a trademark registration for this political message would stifle the speech of others who want to sell shirts displaying similar words. See, for example, Paul Levy of Public Citizen, and John Welch, who noted in commentary after his TTABlog post about the Federal Circuit’s decision in Elster that “the granting of exclusive rights to TRUMP TOO SMALL seems to hamper the First Amendment rights of everyone else to comment on the former [President], lest they be sued for trademark infringement.”

I hope the Supreme Court will clarify in the Elster opinion that the First Amendment prevents application of trademark infringement and dilution law when someone displays another’s name or mark (such as “Trump”) within a political, social, or religious message, artwork, parody, satire, mashup, meme, pun, or other non-product-source-identifying expression on the front or back of a shirt or other expressive merchandise. Regardless of whether these goods are sold for profit, such messages are protected by the First Amendment when they do not convey a false statement about the accused infringer’s products and are unlikely to mislead a reasonable person about the source of the goods, services, or message.

In addition, the speech-protective exclusions from dilution liability in the Lanham Act (15 USC § 1125(c)(3)(A)&(C)) for parody, criticism, commentary, and noncommercial uses of marks should apply to such messages in this context since the ornamental, informational, or expressive display of words, names, and symbols on the outside surface of clothing typically fails to function as a trademark that designates the source of the product for sale. (Unfortunately, after the Jack Daniel’s decision, Elster may not be able to take advantage of the noncommercial use exclusion for his critical use of the famous “Trump” marks if he is sued for dilution since he claims trademark rights in “Trump Too Small” for shirts and the Court added a non-source-identifying use requirement to the noncommercial use exclusion when the mark is used in parody, criticism, or commentary.) This type of expression should remain in the public domain available for everyone to use on expressive merchandise to convey ideas, information, and other messages.

What traditional First Amendment test (if any) could be used to evaluate the constitutionality of viewpoint-neutral trademark laws? While the Supreme Court will likely uphold the constitutionality of Section 2(c) for one or more of the reasons discussed above, it is unclear whether the Justices will also apply one of the Court’s traditional constitutional scrutiny tests to Section 2(c) in Elster. Perhaps the Court will hold without further analysis that Section 2(c) is constitutional because it is a viewpoint-neutral law that does not restrict speech. Based on questions asked by the Justices during oral argument, it did not appear that a majority of them are interested in having courts always apply rational basis, intermediate scrutiny, or strict scrutiny analysis when trademark laws are challenged under the First Amendment.

Justices Sotomayor and Kagan are the Justices most likely to agree with the government that the trademark registration system—which does not restrict speech—is akin to a limited public forum and/or that the benefits of registration of a trademark by the USPTO are comparable to the benefits received through a government subsidy or other government program. They may support using a rational basis analysis or reasonableness standard of review for viewpoint-neutral trademark registration laws. In Tam, Justices Alito, Roberts, Thomas, and Breyer rejected application in the trademark registration context of these more deferential First Amendment tests. (Moreover, during oral argument in Elster, Justice Alito said he has not changed his mind on this point.) Those four Justices held in Tam that the disparagement clause did not even satisfy Central Hudson’s intermediate scrutiny test used to evaluate the constitutionality of regulations of nonmisleading commercial speech. All of the Justices declined to take a position in Tam on whether this test or a more rigorous constitutional analysis should be used when other trademark laws are challenged under the First Amendment. As the Court is likely to uphold the constitutionality of the viewpoint-neutral Section 2(c) provision, the approach used by Justice Alito in his plurality opinion in Tam won’t work in the Elster opinion.

Trademark registration and enforcement laws apply to commercial and noncommercial uses of trademarks, and the Supreme Court’s traditional First Amendment jurisprudence suggests that strict scrutiny analysis is usually required when a law regulates noncommercial speech based on its content. Section 2(c) is a content-based regulation of commercial and noncommercial expression, but most (if not all) substantive trademark registration and enforcement laws regulate the content of speech. Except for trademark dilution law (15 USC § 1052(f) and 15 USC § 1125(c)), most of these content-based laws would likely satisfy intermediate or strict scrutiny analysis. However, several Justices seem to be moving away from the “tiers of scrutiny” approach in other areas of constitutional law, and some Justices appeared reluctant to use one of these traditional tests during the Elster oral argument. In Tam, Justice Thomas thought strict scrutiny analysis should be used to evaluate the constitutionality of the disparagement provision, but that law discriminated based on viewpoint. I believe the Court is unlikely to apply strict scrutiny analysis in Elster when it upholds the constitutionality of the viewpoint-neutral Section 2(c). It may be looking for a new First Amendment test to use in trademark disputes.

A proposed constitutional test for trademark law. I propose that the Court consider adopting a new three-prong First Amendment test in Elster that can be used to evaluate the constitutionality of trademark laws that do not discriminate based on viewpoint. 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 proposed a similar type of First Amendment analysis for trademark law in “Increasing First Amendment Scrutiny of Trademark Law” (pages 421-446) and “Free Speech Challenges to Trademark Law After Matal v. Tam” (pages 432-469). At a minimum, courts should discuss the purpose(s) of this trademark law, the fit between the law’s means and ends, and that rule’s potential harm to free speech when they are evaluating whether a certain trademark statute, regulation, or common law doctrine is consistent with the right to freedom of expression. The Court arguably used this approach in the Jack Daniel’s opinion, as explained at the end of my blog post about that decision.

Under the first prong of my proposed constitutional test, courts should identify the goals of the specific trademark law—not the purposes of trademark law generally—that are important government interests. The primary goals of most trademark registration and enforcement laws are preventing fraudulent, misleading, or deceptive uses of trademarks; preventing unfair competition; promoting fair competition; and facilitating the communication of source-identifying information about goods or services. When applied to source-confusing uses of another’s name, portrait, or signature, Section 2(c) furthers these goals just like the Lanham Act’s bar against registration of marks that falsely suggest a connection with a person (15 USC § 1052(a)) and of marks that are likely to cause confusion with a prior mark (15 USC § 1052(d)). However, another purpose of Section 2(c) is to protect the rights of publicity and privacy of living individuals and deceased Presidents by preventing others from registering trademarks that contain their name, portrait, or signature without permission. As noted by Justice Gorsuch during the Elster oral argument, and documented by Professor Rothman in her paper “Navigating the Identity Thicket” (pages 1307-1309), there is a historical tradition of government limits on the unauthorized use as trademarks of the names or images of Presidents and other public figures. It is likely the Court will conclude that protecting the rights of publicity and privacy within a viewpoint-neutral trademark registration law are important government interests.

Section 2(c) would also satisfy the second prong of my proposed constitutional test for trademark law, which requires courts to confirm this trademark law directly and materially furthers those trademark goals. By barring the trademark registration of a living individual’s or deceased President’s name, portrait, or signature without consent, this trademark registration provision is narrowly tailored to prevent the registration of certain trademarks that are likely to cause source confusion and to protect the rights of publicity and privacy of government officials, public figures, and other recognized individuals, as explained by INTA in its amicus brief filed in Elster.

The third prong of my proposed test is also satisfied by Section 2(c), as this law does not endanger free speech at all; it actually promotes freedom of expression. As discussed above, Elster and other trademark applicants can display their expression on shirts regardless of whether Section 2(c) bars registration of the proposed mark without the named person’s written consent. Registering a political phrase or other non-source-identifying message as a mark for clothing actually stifles more protected speech than the denial of that registration. The trademark owner could then invoke the registration in a cease-and-desist letter or trademark complaint submitted to an online marketplace or court, and thereby chill or suppress the speech of the named person and other individuals, organizations, and companies that want to express themselves in the same way on competing goods. The current defensive doctrines in trademark infringement law are not sufficient to protect free expression in this context, as explained by Professor Christine Farley and me in our paper “Raising the Threshold for Trademark Infringement to Protect Free Expression” (2023). Thus, the Court should reject Elster’s argument that a registration for his “Trump Too Small” political message will not stifle nonmisleading and/or noncommercial expression protected by the First Amendment.

My proposed three-prong constitutional test for trademark law will better protect free speech values in trademark law than the Court’s approach in San Francisco Arts and Athletics v. United States Olympic Committee (1987). During the oral argument in Elster, Justice Alito asked whether “the constitutionality of [Section 2(c)] could be sustained on a theory similar to the one in San Francisco Arts and Athletics” since he does not approve of the “government benefits theory” proposed by the government. The government said no, and I agree. The problem with looking to that case for guidance in Elster is that it involved a First Amendment challenge to a special statute—Section 110 of the Amateur Sports Act—which granted the United States Olympic Committee the right to prohibit certain commercial and promotional uses of words and symbols associated with the Olympic games. Like copyright law (which was also discussed at oral argument in Elster), this provision of the Amateur Sports Act has different goals and rules than Section 2(c) and other trademark registration and enforcement laws in the Lanham Act. This is just one of the reasons the Elster Court should decline to adopt the First Amendment approach used in these different contexts. (See “A Free Speech Right to Trademark Protection?” (pages 879-880) and “Free Speech Challenges to Trademark Law after Matal v. Tam” (pages 433 n. 150 and 449 n. 228) for a brief discussion of how the Lanham Act provisions are different than copyright law and the Amateur Sports Act, and cites to other commentary on this point.)

In conclusion, the Supreme Court will likely hold that Section 2(c) is consistent with the First Amendment since it is a viewpoint-neutral trademark law that does not restrict speech. It may also clarify that a trademark applicant does not have a free speech right to a trademark registration, especially when the proposed mark is a political message that fails to function as a source-identifying trademark. To increase certainty and predictability in trademark law, the Court should also provide more guidance in Elster on how to balance an accused infringer’s right to freedom of expression against another’s claims of trademark rights or the rights of publicity and privacy.