Symposium: Free speech comes to trademark law

Christine Haight Farley is a professor at American University Washington College of Law who teaches intellectual property law. 

In Matal v. Tam, formerly Lee v. Tam, the court settled certain aspects of First Amendment law while it opened up new issues in trademark law. It is a challenge for the uninitiated to follow a coherent path through the court’s First Amendment jurisprudence. I speak from personal experience as a trademark expert who has tried to make sense of First Amendment law. Free-speech absolutists will of course always find it easy to get to a particular end point. But I found certain questions, such as the approach to commercial speech and the treatment of viewpoint discrimination, to have some frustrating gaps. A First Amendment lawyer therefore will no doubt appreciate the court’s opinion in Tam, as it bears on both of these issues.

A trademark lawyer, in contrast, will come away with more questions raised than answered. The court definitively answered the question of whether the U.S. Patent and Trademark Office can refuse registration to disparaging marks. It cannot. (And yes, this is a decisive victory in the REDSKINS case too.) But what should the PTO do with applications for marks that are “scandalous” or “immoral,” as both categories are also barred registration by the Lanham Act? Although the court was silent as to these bars, there may not be any way to distinguish them from the provision at issue in Tam. This lingering question was to be expected because the court does not pronounce on issues that are not before it. The court’s opinion, however, raises a far more fundamental question: What is the relationship between the First Amendment and trademark law?

In striking down the challenged law as unconstitutional, a unanimous court (Justice Neil Gorsuch did not participate) was definitive on two points, which clarify First Amendment law beyond its application to trademark law: The disparaging marks provision of the Lanham Act constitutes viewpoint discrimination, and the program of federal trademark registration is not government speech. The second point was perhaps the most central issue in the case for the court. Having recently decided another government speech case, certain justices were presumably eager to revisit that doctrine. In 2015, in Walker v. Texas Division, Sons of Confederate Veterans, Inc., the court split 5-4, ultimately finding constitutional a law that denied the Sons of Confederate Veterans specialty license plates. Justice Samuel Alito wrote the dissent in Walker (joined by Chief Justice John Roberts, Justice Antonin Scalia and Justice Anthony Kennedy) and the majority opinion in Tam. Tam was therefore an opportunity for the dissenters to limit this doctrine, noting it is “susceptible to dangerous misuse,” and declare that Walker “marks the outer bounds of the government-speech doctrine.” The court in Tam fervently rejected the proposition that when the government puts its seal of approval on certain private speech, that speech is transformed into government speech.

Tam therefore may be of interest to a First Amendment lawyer as a curtailment of the government-speech doctrine. The court rejected the application of that doctrine because the government neither creates nor independently cancels these marks, the messages that registered marks convey are varied and even inconsistent, and the public does not understand the registration process. It may be difficult to fit trademark registration into the category of government speech, but it is not clear why any of the characteristics listed above should be essential. One quibble with the court’s analysis of trademark registration as government speech: The court stated that the PTO “does not edit marks submitted for registration.” I would argue that when the PTO requires an applicant to officially disclaim words in a mark that the applicant is already using in commerce, it is editing the mark that becomes registered. (A broader critique is that the court seems often to be speaking of trademarks when the issue before it is limited to the registration of trademarks.)

The court gave comparatively short shrift to the other arguments advanced by the government as to why the disparaging marks provision is constitutional — that the provision was a feature of a government subsidy or a government program in which participants’ constitutional rights outside the program were not leveraged by the program restrictions. Too bad as I thought these were better arguments in favor of the provision than the government-speech doctrine.

As to viewpoint discrimination, Tam instructs that it is a subcategory of content discrimination and that when it is present, the limited public forum doctrine is unavailable. More importantly, we learn in Tam about the breadth of the category of viewpoint-based speech. A regulation may be deemed viewpoint based even when neither the government nor the speaker has a viewpoint. The disparagement provision was unsuccessfully argued to be viewpoint neutral because the government did not pick sides and was not swayed by the applicant’s intentions. The court found, however, that it is enough if the law considers the viewpoint of the audience. Moreover, the court bluntly stated that “[g]iving offense is a viewpoint.” These broad pronouncements about viewpoint discrimination will surely keep First Amendment lawyers and scholars busy.

Tam does not, however, elucidate on the standard applied to commercial speech. The question of when, if ever, strict scrutiny is appropriate when the regulation at issue is directed to commercial speech was provoked by the court’s 2011 decision in Sorrell v. IMS Health Inc. Another question, which arose for the first time in this case, is whether trademarks are properly categorized as commercial speech. Neither of these questions was answered by the court in Tam. The majority opinion acknowledged the latter question, although not the former, and stated that the court did not need to resolve the issue because the disparaging marks provision fails even the more lenient test for commercial speech established in Central Hudson Gas & Electric Corp. v. Public Service Comm’n of New York. Both Kennedy’s concurrence (joined by Justices Ruth Bader Ginsburg, Sonia Sotomayor, and Elena Kagan) and Justice Clarence Thomas’ concurrence answered the former question directly and unequivocally. These five justices all agreed that when viewpoint discrimination is present, strict scrutiny should apply even in the context of commercial speech. Thus, on this point, the majority was actually in the minority as only Roberts, Breyer, and Alito seemed to resist this step.

Alito’s opinion found that the disparaging marks provision failed the Central Hudson test, I believe, chiefly because the government was not able to identify a substantial government interest that the provision advances. For example, during oral argument, the PTO’s deputy solicitor general surprised legal commentators with his response to Breyer’s question about the purpose of the disparagement provision. He responded that the provision was consistent with broader trademark policy because disparaging marks were a distraction to the public. There was much tittering about this argument within my echo chamber. The lack of a clear rationale for this provision in the legislative history is something I have addressed in my scholarship.

Although this is not the decision I was hoping for in this case — I filed an amicus brief co-authored with Rebecca Tushnet on behalf of 31 law professors in support of the government — it is the decision I was expecting. What I could not anticipate — and what I was most eager to discover — is what the court would say about the relationship between the First Amendment and trademark law. I was wondering whether Tam would be to trademark law what Eldred v. Ashcroft was to copyright law in terms of rationalizing how intellectual-property law does not run afoul of the First Amendment.  It was not.

The Tam opinion goes a certain distance in suggesting the applicability of First Amendment protections within trademark law, but it does not explain the relationship between the two or how the burdens that trademark law place on speech can survive. In this regard it is significant that the court chose not to categorize trademarks as commercial speech. Previously, courts have justified injunctions on the non-misleading use of trademarks as a permissible restriction on commercial speech. As the court expands the protection of commercial speech, or the speech that formerly belonged in that category, it will become more difficult for courts to enjoin non-misleading speech.

Although it may have been unnecessary to reach the question of whether trademarks are commercial speech, the court’s silence may be telling. Rather than answering that question directly, the court hinted at the proposition advanced by Tam and some amici that trademarks are expressive speech. For example, the court twice noted that the band adopted the mark in an effort to reclaim the derogatory term and also observed that “trademarks often consist of catchy phrases that convey a message.”

Justice Kennedy’s concurrence explicitly noted that the “case does not present the question of how other provisions of the Lanham Act should be analyzed under the First Amendment.” Immediately following this statement, however, he observed, almost as if to answer the question anyway, that what is settled is that the law can prohibit the use of confusing and misleading trademarks. This raises the question of whether trademark law runs afoul of the First Amendment when it prohibits the non-misleading use of trademarks or denies registration for other reasons.

Either if trademarks are not commercial speech, or if content-based regulations of commercial speech are subject to strict scrutiny, then certain provisions of trademark law become vulnerable. Section 2 of the Lanham Act contains other registration bars that are not directed at the misleading use of trademarks. For instance, Section 2 also prohibits the registration of inaccurate but not necessarily misleading geographical indications of origin for wine or spirits, names of specific living persons absent written consent, and government insignia. Would these prohibitions survive strict scrutiny? Likewise, the dilution provisions in Section 43(c), which enable non-confusing speech to be enjoined, could be challenged based on the broad understanding of speech protection in trademark law advanced by the Tam court. These questions will have to be answered in another case, as I am sure they will be, because the Tam opinion invites trademark litigants to raise constitutional claims and courts to experiment with free-speech doctrines in trademark law.