What Trademark Law Is Learning from the Right of Publicity

Barton Beebe* In 2006, Stacey Dogan and Mark Lemley published the now classic article "What the Right of Publicity Can Learn from Trademark Law." 1 There they urged us to "[r]econceiv[e] the right of publicity as a trademark-like right," 2 particularly because "looking at the right of publicity through the lens of trademark law offers logical ways to limit the right." 3 Right of publicity law's failure to incorporate trademark law's limitations had resulted, they argue, in the right of publicity functioning as "a mutant version of trademark policy . . . ." 4 I argue here in response that regrettably, notwithstanding Dogan and Lemley's good advice, the reality is that it is trademark law that has become more like right of publicity law. Indeed, trademark law is in danger of becoming a "mutant version" of right of publicity policy. To defend this claim, I will first briefly survey the conventionally-recognized similarities and differences between trademark law and right of publicity law. I will then propose a more complicated-and, I think, more accurate-comparison between the two areas of law and argue that they are converging in many important ways, giving us the worst of both worlds. In what follows, I will focus more on trademark law, not only because many other contributions to this Symposium thoroughly discuss right of publicity law, but also because I think we should be especially concerned with trademark law's mutation in the age of Burwell v. Hobby Lobby Stores 5 and Citizens United v. Federal Election Commission 6 into a kind of right of publicity law for corporate personhoods.

Before I proceed, I should emphasize one point from the start. I very much doubt that Dogan or Lemley would disagree with the general thrust of my argument, not least because much of it draws upon other work by them. COLUMBIA JOURNAL OF LAW & THE ARTS [42:3

I first compare the policy goals understood to be guiding the two areas of law and then turn to a comparison of their subject matter, scope of rights, defenses, and limitations.

In terms of the customary policy justifications for the two areas of law, there are apparent differences. We conventionally say that trademark law protects trademarks from infringement and thereby maintains the integrity of the trademark system in order (1) to minimize consumer search costs and (2) to encourage firms to maintain consistent levels of product quality. 7 This is the orthodox law and economics justification for trademark protection. 8 It makes obvious good sense with respect to trademarks on products like pharmaceuticals or food. If we acknowledge only this justification for the protection of trademarks, then we will be in a position to conclude that trademark law, with its emphasis on search costs, is very different from right of publicity law.

But the orthodox law and economics justification makes less sense when applied to goods whose primary purpose is to convey social distinction, such as certain kinds of apparel. 9 This social distinction may be conventionally hierarchical in nature, and this is how lawyers almost invariably think about social distinction-such that society consists essentially of one giant U.S. News & World Report ranking. 10 But social distinction may more generally take the form simply of non-hierarchical, horizontal difference from mass society. 11 In light of the role that trademarks play in facilitating the consumer's sense of identity, we sometimes admit that we protect trademarks from infringement to preserve firms' incentives to produce signs of social difference and to preserve the distinguishing function of those signs of social difference. 12 It is conventionally thought that different justifications motivate right of publicity law, and Jennifer Rothman reviews them quite effectively in her book The Right of Publicity: Privacy Reimagined for a Public World. 13 As she points out, the notion that right of publicity law strives to protect consumers from deception is unpersuasive, not least because right of publicity law typically does not require a showing of likelihood of consumer confusion. 14 Instead, a variety of other justifications have been adduced over the years for right of publicity protection: It incentivizes the creation of celebrity personas; 15 it prevents the overgrazing or tarnishment of those personas and protects the public's interest in stable celebrity identities; 16 it provides an appropriate reward for the labor that goes in to the development of a valuable personal identity; 17 and it prevents misappropriation or unjust enrichment of that value. 18 And finally, a rationale that emerges in part from the earliest origins of right of publicity law is that the law protects the individual liberty and dignitary interests of identity holders. It enables identity holders to engage in what Mark McKenna calls "autonomous self-definition" 19 and protects them from certain forms of emotional harm, a rationale which Rothman also endorses. 20 We care about rationales because they guide-or at least tend to have some influence on-the specific rules that make up trademark law and right of publicity law. An example of their influence is found in the differing subject matters of trademark law and right of publicity law. Individual identities, and especially individual celebrity identities, routinely qualify for trademark protection. 21 Corporations, by contrast, cannot qualify for right of publicity protection (at least not yet), apparently because corporations do not have privacy, dignitary, or autonomous self-definition interests, and do not suffer emotional harms. 22 A CULTURAL ANALYSIS OF THE RIGHT OF PUBLICITY AND PASSING OFF 45-62 (Cambridge Univ. Press 2017) (reviewing theoretical justifications for right of publicity).

14. The influence of underlying rationales is also apparent in the two areas' different tests for infringement and defenses to liability. Commentators conventionally emphasize, as I did above, that under right of publicity law the plaintiff need not show that the defendant's conduct is likely to confuse consumers as to source or sponsorship. Instead, a defendant may be liable by merely invoking the identity holder's identity, even if consumers do not believe that the identity holder is endorsing a product. 23 The result is that right of publicity law is generally thought to offer a broader scope of protection to identity holders than trademark law has offered to trademark owners. 24 As for defenses, the various defenses available in right of publicity law, which are also thoroughly reviewed in Rothman's book, 25 tend to be relatively plaintifffriendly. By comparison, trademark law's Rogers v. Grimaldi test 26 (which strictly speaking functions as a limitation on the plaintiff's rights rather than a defense) is more than just defendant-friendly. In operation, it has seemed until recently to be defendant-dispositive. As the Ninth Circuit observed in Gordon v. Drape, "on every prior occasion in which we have applied the [Rogers] test, we have found that it barred an infringement claim as a matter of law." 27 Though, as I will discuss shortly, in Gordon, the Ninth Circuit has finally hinted at some possible limits on the Rogers test.

There are other dimensions along which we can compare right of publicity law and trademark law-for example, their terms of protection or their approaches to assignability-that I will skip over here. Instead, I will conclude this very brief survey of the conventional similarities and differences between the two laws by suggesting overall that trademark law has greater built-in limits, especially in its likelihood of confusion requirement and the very powerful Rogers test. As Dogan and Lemley explained in 2006, "the structure and content of trademark law provide a theoretical justification for a bounded right of publicity." 28 23. Id. § 5:11. 24. See, e.g., Hart v. Elec. Arts, Inc., 717 F.3d 141, 158 (3d Cir. 2013) (stating "the right of publicity is broader and, by extension, protects a greater swath of property interests").

25. See ROTHMAN, supra note 13, at 145-53. 26. See Rogers v. Grimaldi, 875 F.2d 994, 999 (2d Cir. 1989) (stating that "[i]n the context of allegedly misleading titles using a celebrity's name," infringement will not be found "unless the title has no artistic relevance to the underlying work whatsoever, or, if it has some artistic relevance, unless the title explicitly misleads as to the source or the content of the work."). The Rogers test has since expanded to include a defendant's use of a mark within the body of the defendant's work.

Despite all of the conventionally-recognized differences between trademark law and right of publicity, and Professors Dogan and Lemley's hopes in 2006, I want to propose that it is trademark law that has increasingly taken on the characteristics of right of publicity law rather than the other way around.

Consider the requirement-if it may still be called a requirement-in trademark law that a plaintiff must show that the defendant's mark creates a likelihood of consumer confusion. As many commentators have suggested, the likelihood of confusion test has arguably mutated into a mere likelihood of association test, which finds liability if consumers merely associate the defendant's mark with the plaintiff, even if consumers are not ultimately confused as to the true source of the defendant's goods and even if consumers' association has no impact on their purchasing behavior. 29 In its recent opinion in Int'l Info. Sys. Sec. Certification Consortium, Inc. v. Sec. Univ., the Second Circuit emphatically embraced this view: "The modern test of infringement is whether the defendant's use [is] likely to cause confusion not just as to source, but also as to sponsorship, affiliation or connection." 30 As the Int'l Info. court observed, the shift toward a mere association standard began at least with the 1962 amendment to Lanham Act § 32 that deleted from the section the limiting phrase "purchasers as to the source or origin of such goods or services." 31 Formerly, conduct was actionable only when it confused "purchasers" and only when it confused them as to the source of the goods they were purchasing. With this language removed, explained the Int'l Info. court, "the Act's protection against infringement is not limited to any particular type of consumer confusion, much less exclusively to confusion as to source." 32 Subsequent developments have only quickened the shift toward a mere association standard for confusion. As Professors Dogan and Lemley recognize in their article on the merchandising right in trademark law, the merchandising cases of the 1970s and '80s consolidated the principle that a defendant may be liable for merely invoking the plaintiff's identity and suggesting to consumers that the plaintiff has endorsed or merely acceded to the defendant's conduct. 33 the Trademark Law Revision Act of 1988 amended section 43(a) of the Lanham Act to include extraordinarily capacious language covering the kind of conduct that violates the act. 34 The case law under section 43(a) now seems to suggest that anyone who does anything referential shall be liable to any entity that believes that it is "likely to be damaged" 35 by that reference. The Fourth Circuit Court of Appeal's recent opinion in Belmora LLC v. Bayer Consumer Care AG further suggests that the language of section 43(a) refers to any entity in the world, regardless of whether it is actually using a trademark within the territorial borders of the United States. 36 In sum, what is conventionally recognized as one of the fundamental distinctions between trademark law and right of publicity law-that the former requires a showing of consumer confusion while the latter does not-has arguably become a distinction without a difference. One objection to this characterization of the evolution of the likelihood of confusion test in trademark law may stem from the claim that only the strongest marks benefit from the shift toward a mere association test for confusion. But even if this claim is true, and it likely is not, it simply shows how trademark law has further followed-or at least paralleled-the example of right of publicity law. Both areas of law increasingly express the legal logic of the economics of superstars, in which superstars-be they the strongest marks or the most famous celebrities-dominate their markets and further enable themselves to do so by promoting legal rules designed to protect and enhance their superstar status. 37 In both areas of law, the stronger benefit from broader protection. 38 But probably the best expression of the mutation of trademark law into a right of publicity law for persons both natural and corporate is antidilution protection. 39 Here, the majority of courts interpret the anti-blurring provisions of the Lanham Act to require that the plaintiff show only that consumers associate the defendant's mark with the plaintiff's in order to trigger a finding of blurring. 40 The leading survey method for dilution by blurring simply asks respondents what comes to their minds when they are exposed to the defendant's mark; if respondents identify the plaintiff, then this is taken as evidence of blurring. 41 The parallels with right of publicity protection are even clearer in anti-tarnishment protection, where corporations appear to possess something akin to dignitary interests. This has been apparent in cases where the defendant has had the temerity to associate the New York Stock Exchange with gambling 42 or to associate the Victoria's Secret corporation with "sex-related products." 43 In these and other anti-tarnishment cases, courts rule according to rationales very similar to those in right of publicity law. 44 Still, I do not want to overstate my case. Doing so would risk what I think is a common problem in academic intellectual property law writing, including in my own. This is what might be called the problem of the self-fulfilling description, in which a scholar exaggerates how bad the law is on a particular issue (in order to sell the importance of the scholar's topic). In doing so, the scholar risks persuasively restating the law in such a way that judges and lawyers come to believe that the law is in fact as bad as the scholar says it is, and they decide cases and counsel clients accordingly. I mean to refer to the practice of citing horrible edge cases as representative of the center of the law-which thereby helps to bring these peripheral cases into the center.

Trademark law is certainly still its own thing, especially in the area of limitations and defenses. The best example of this is the Rogers test. But even here, we are beginning to see a breakdown. The Ninth Circuit's Gordon v. Drape probably still counts as an edge case, but it should give us cause for concern. The author of the Gordon opinion was Judge Jay Bybee, who just happened also to author the Ninth Circuit's opinion in the trademark case Brown v. Electronic Arts, Inc., which applied the Rogers test to find no infringement. 45 Judge Bybee likewise authored the majority opinion in the right of publicity case In re NCAA Student-Athlete Name & Likeness Licensing Litigation, which held that Rogers cannot be applied to a right of publicity claim. 46 In Gordon, Judge Bybee needed to find some limit on the Rogers test. (And who can blame him? The test is probably too defendant-friendly.) He found that limit in something like a transformativeness test, which analyzes "the degree to which the junior user uses the mark in the same way as the senior user" and the "extent to which the junior user has added his or her own expressive content to the work beyond the mark itself." 47 There would be much more to say along these lines if I were adequately to persuade you of my point, but I want to take what time I have left instead to speculate on why we are seeing trademark law mutate into something like right of publicity law.

I think there are several different explanations for the mutation of trademark law into a right of publicity law for corporate personas, many of which may be obvious to you. I will briefly focus on two primarily to set out what might be an agenda for further research.

The first relates to the legal concept of misappropriation and to the concept, now increasingly fashionable in the humanities, of "affect," 51 a term perhaps best defined broadly as intensities of feeling. I very much doubt that courts are consciously patterning trademark law after right of publicity, and I suspect you doubt this too. Instead, courts applying trademark law and courts applying right of publicity law have both long been converging toward the general prevention of conduct they perceive to constitute misappropriation. 52 What is it that is being misappropriated? Right of publicity law teaches us that it is something more essential than source or goodwill, or even what I mentioned above-commodified social difference. The thing of value that is being misappropriated is raw positive affect in commodified form, which is especially valuable in a world filled with affectless mass-produced commodities but whose media is otherwise "characterized by a surfeit of [affect]." 53 Just as a celebrity persona can endow a product with positive feeling, so too can a corporate persona as embodied in its trademark. One need think only of Apple or of Nike. More generally, so often when trademark law uses the term "aesthetic," as in the Ninth Circuit cases struggling with the trademark law concept of "aesthetic functionality," 54 what we mean to refer to is affect and affective value. 55 The antimisappropriation impulse in trademark law, as in right of publicity law, seeks to protect this affective value from free riders.

The second explanation for the shift in trademark law toward a right of publicity framework for corporate personas is perhaps more interesting but also more difficult to prove. Trademark law is both reflecting and contributing to the continued rise of corporate rights and attention to corporate personhood. 56 If the merchandising cases of the '70s and '80s were one chapter in the recent evolution of trademark law, 57 another chapter-the one we find ourselves in the midst of right now-is the corporate personhood chapter. Courts today are solicitous of corporations' entitlement to "autonomous self-definition," 58 [C]orporate personhood has played only a secondary role in the corporate rights movement. While the Supreme Court has on occasion said that corporations are people, the justices have more often relied upon a very different conception of the corporation, one that views it as an association capable of asserting the rights of its members. This alternative way of thinking about the corporation has paved the way for the steady expansion of corporate rights. Indeed . . . corporate personhood has traditionally-and surprisingly-been used to justify limits on the rights of corporations. ADAM WINKLER, WE THE CORPORATIONS 7 (2018). See also Elizabeth Pollman, Reconceiving Corporate Personhood, 2011 UTAH L. REV. 1629, 1630 (2011) ("[C]orporate personhood should be understood as merely recognizing the corporation's ability to hold rights in order to protect the people involved. This concept is the only common thread in the case law."). One question this raises in the trademark law context is the extent to which courts may begin similarly to justify trademark rights as based on shareholder interests. perhaps even corporations' right of privacy. 59 Instead of corporate persona, we say "brand," and this may explain why a good deal of scholarship would prefer to describe trademark law as "brand law" rather than trademark law. 60

I will briefly conclude by asking how we might try to counteract trademark law's shift toward a right of publicity sensibility. This issue brings us to one important remaining difference between right of publicity law and trademark law. Trademark law seeks, in essence, to promote competition. By contrast, right of publicity law does not try to promote competition in the production of celebrity personas; much of right of publicity doctrine has been developed without any attention to optimizing competition for the benefit of consumers. 61 This distinction may go far toward explaining why right of publicity rights are generally so much more expansive than trademark rights. But just as competition concerns represent an important check on the expansion of corporate rights, so in trademark law, an emphasis on competition may help to limit the expansion of trademark rights stemming from-or at least coincident with-the rise of corporate rights. As the federal courts shift ideologically more to the right and toward finding greater protections for corporate interests, it may be that economic arguments will find a more receptive audience than arguments driven by the First Amendment or a disaffection for consumer culture.