The First Amendment Limits Trademark Rights, But How?–Jack Daniel’s v. Bad Spaniels (Guest Blog Post)
by guest blogger Lisa P. Ramsey
[Lisa P. Ramsey is a Professor of Law at the University of San Diego School of Law. She writes and teaches in the trademark law area, and was one of the signatories of the First Amendment Professors amicus brief filed in Jack Daniel’s Properties, Inc. v. VIP Products LLC]
During oral arguments at the Supreme Court on March 22, 2023 in Jack Daniel’s Properties, Inc. v. VIP Products LLC, the Justices’ questions suggested that they believe the First Amendment limits the scope of trademark rights. They may also be looking for a speech-protective trademark doctrine that can be invoked on a motion to dismiss to stop broadly-worded infringement statutes from chilling expression that is not false or misleading. Christine Haight Farley and I recently wrote an article that discusses the options for the Court if it wants to hold that the First Amendment limits trademark rights, but not adopt the Rogers test applied by the Ninth Circuit Court of Appeals in the Jack Daniel’s dispute (I will discuss the Rogers test later).
Professor Farley and I propose a trademark fair use defense that would apply to any informational or expressive use of language or designs claimed by another as a mark in connection with any goods or services. Examples include words and symbols used to identify or describe goods or services; use of another’s mark to refer to the trademark owner or its products in news reporting, posts on blogs or social media, consumer product reviews, teaching, scholarship, and comparative advertising; parody, satire, mashups, memes, and other jokes; and political, social, and religious messages that do something more than identify the trademark owner as the source of the goods, services, or expression. If this threshold requirement of our test is satisfied, the informational or expressive use of the mark should be presumptively a fair use unless it is (1) a false statement about defendant’s products (including false claims of origin, sponsorship, endorsement, approval, affiliation, connection, or association) or (2) is likely to mislead reasonable persons about the source of the goods, services, or message. Our proposed fair use defense is a speech-protective interpretation of the “likely to cause confusion” language in the trademark infringement statutes, 15 U.S.C. §§ 1114(1), 1125(a)(1)(A), which furthers trademark law’s goals of preventing fraudulent, deceptive, or misleading uses of trademarks.
As this test can be applied on a motion to dismiss when a defendant is using another’s trademark to communicate information, ideas, or viewpoints, our proposed fair use defense is similar to the Rogers test first developed by the Second Circuit in Rogers v. Grimaldi, and later adopted and expanded by the Ninth Circuit. The Rogers test requires a trademark owner to satisfy a higher standard for infringement if the defendant proves that its allegedly infringing use is part of an expressive work protected by the First Amendment. Once this threshold legal showing is satisfied, the federal trademark law cannot apply in the dispute unless the plaintiff establishes that the defendant’s use of the mark is either (1) not artistically relevant to the underlying work or (2) explicitly misleads as to the source or content of the work. Thus, in trademark disputes in the Ninth Circuit that implicate First Amendment interests, the Rogers test requires courts to interpret the infringement statutes narrowly in a manner which makes them constitutional by focusing on whether the expression is explicitly misleading rather than an evaluation of numerous factors of a judge-made, fact-intensive test typically used to assess whether the defendant’s use of the mark is “likely to cause confusion” in the marketplace.
During oral argument in Jack Daniel’s, some Justices expressed concern about the “artistic relevance” prong of the Rogers test—which might require courts to judge whether something is art—while others noted that “explicitly” does not appear in the infringement statute. Still, a few Justices seemed to like the fact that Rogers is an objective analysis with a reasonable person standard that allows courts to dismiss weak trademark claims early in a lawsuit. In the Ninth Circuit, there is no need for the parties to engage in costly discovery or a trial about multiple likelihood of confusion factors if this artistically relevant use of the mark is not explicitly misleading. A consumer survey is not required. Thus, it is easier for the defendant to win on a motion to dismiss under this test. This rule benefits artists, authors, students, non-profit organizations, small businesses, and others with limited financial resources who cannot afford to litigate through discovery and trial to a successful result. The Rogers test protects the rights of trademark owners by allowing courts to ban certain false or misleading uses of their marks.
Our proposed fair use defense has these same advantages as the Ninth Circuit’s version of the Rogers test, but it addresses the concerns about the Rogers test raised by the Justices during oral argument. Our test avoids consideration of the “artistic relevance” of this use of the plaintiff’s mark, is more clear than the “explicitly misleads” test, and contains an infringement standard that incorporates the text of the Lanham Act. Moreover, our defense does not require courts to determine whether the defendant is selling an expressive work or commercial product, or whether this use of another’s mark is noncommercial or commercial speech. Both types of speech are protected by the First Amendment if they are not fraudulent, deceptive, or misleading. They cannot be regulated by the government using trademark law without a sufficient justification under the Supreme Court’s First Amendment jurisprudence.
During oral argument, a few Justices seemed concerned that a court applying the regular likelihood of confusion factors could ban use of another’s mark in a political message or parody displayed on the front of a T-shirt. Some questions suggested the Justices were trying to figure out whether the Bad Spaniels design was a parody, if it was used as a trademark to identify the source of the dog toy, or if it was commercial or noncommercial speech. The Justices may be considering whether other judge-made defenses to infringement should be used to protect First Amendment interests in trademark disputes, like parody, non-trademark use, or noncommercial use. As we discuss in our article, these defenses have limitations and other problems. For example, the parody defense currently requires case-by-case analysis of multiple likelihood of confusion factors, so it may be difficult to invoke this speech-protective rule on a motion to dismiss. The expense of litigation may chill speech. Judges could protect speech interests by dismissing infringement claims as a matter of law in disputes where there is no trademark use or commercial use of the mark, but a non-trademark or non-commercial use defense to infringement liability might allow certain false or misleading uses of another’s mark that are harmful.
One benefit of our proposed fair use defense is that it does not require courts to determine whether a defendant’s use of another’s mark is a parody, a trademark use, or a commercial use of the mark. It applies to parody, criticism, and commentary about the trademark owner, and to other forms of expression like satire, jokes, mashups, memes, or other communications about the world around us. Judges need not decide whether the message is funny; they must only consider whether this is an informational or expressive use of another’s mark, and then apply our higher standard for infringement liability. It is clear after Matal v. Tam that trademarks contain expression protected by the First Amendment. Thus, our fair use defense is not limited to trademark disputes where the expression is being used otherwise than as a mark. The fact this is a parody, a trademark use, or commercial use of the mark is relevant, but not determinative. When courts evaluate whether a defendant’s use of another’s mark is likely to mislead reasonable people about the source of the product or message, the following evidence is relevant under our proposed test: (a) the content of the defendant’s expression (e.g., use of a term for its inherent informational or expressive meaning, or in parody, satire, or another non-reputation-related message); (b) the context of its use (e.g., a trademark use or non-trademark use); and (c) the products sold or provided by the defendant (e.g., a political service, movie, T-shirt, dog toy, or can of peas).
Courts in the Ninth Circuit have applied the Rogers test outside the context of parody, non-trademark uses, and uses of marks in the title of an artistic or literary work, but the “no artistic relevance” and “explicitly misleads” language in the test might not appeal to certain Justices. It is not clear from the oral arguments whether most of the Justices approve of the Rogers test and, if so, whether they would only apply it in trademark disputes involving expressive works with titles, or more broadly to any expression protected by the First Amendment. If the Justices are looking for a speech-protective rule that can be applied on a motion to dismiss when trademark laws implicate the First Amendment, we think our test does a better job balancing the public’s interest in avoiding consumer confusion against the public’s interest in free expression than the Rogers test, the multi-factor likelihood of confusion test, or constitutional scrutiny of infringement laws as applied in a specific trademark dispute.