[*PG291]AN EXCLUSIVE RIGHT TO EVOKE
Abstract: Ten years ago, in White v. Samsung Electronics America, Inc., the U.S. Court of Appeals for the Ninth Circuit held that a robot violated Vanna White’s publicity rights. Since White, the tendency to equate evocation with infringement in trademark and right of publicity cases has only grown. In contrast to this expansionist trend in trademark and right of publicity law, however, courts in recent copyright cases have arguably backed away from a strong right to evoke. This Article identifies these trends and suggests some reasons for concern over an exclusive right to evoke. The author argues that if we wish to preserve a rich commons and avoid significantly chilling free expression, courts should at least cabin the right to evoke and ensure that, when utilized, it serves the law’s normative goals.1
evoke \i-’vOk\ . . . 1: to call forth or up: as . . . c: to bring to mind or recollection . . . .2
Ten years have passed since the U.S. Court of Appeals for the Ninth Circuit, in White v. Samsung Electronics America, Inc.,3 held that a robot violated Vanna White’s publicity rights.4 In White, the court held [*PG292]that the right of publicity extends beyond a celebrity’s name or likeness to cover virtually any symbol that might “evoke” a celebrity’s identity for commercial gain.5 Judge Kozinski argued in his dissent from a denial of rehearing en banc, “Instead of having an exclusive right in her name, likeness, signature or voice, every famous person now has an exclusive right to anything that reminds the viewer of her. After all, that’s all Samsung did: It used an inanimate object to remind people of White, to ‘evoke [her identity].’”6
Kozinski portrayed this so-called “right to evoke” as a radical departure from prior precedent,7 but in fact, intellectual property laws had been used for some time to prevent the use of non-proprietary symbols to evoke personal identities, trademarks, or creative works. The right of publicity law was applied to look-alikes,8 sound-alikes,9 and a racecar that made people think of its driver;10 state trademark [*PG293]dilution laws had been invoked against marks that resembled other marks without causing confusion;11 and copyright law had been interpreted to protect the “total concept and feel” of creative works.12 Certainly, trademark and copyright law sometimes offered safe harbor if the evocation was designed to parody or criticize the intellectual property;13 but if the evocation was not so designed, courts frequently granted relief based on a party’s use of material that did not consist of, but somehow evoked, another’s original work, trademark, or celebrity image.14
After White, the tendency to equate evocation with infringement in trademark and right of publicity cases only grew. After the passage of the Federal Trademark Dilution Act of 1995 (“FTDA”), some courts concluded that an owner of a famous mark could prevent others from using marks that merely “conjure[d]” the famous mark,15 and actors extended the right of publicity to devices that evoked characters that [*PG294]the actors played on television.16 Courts reached these decisions with little consideration of their impact on the cultural commons—i.e., on the pool of communicative resources available for use by the public.17
In contrast to this expansionist trend in trademark and right of publicity law, however, courts in recent copyright cases have arguably backed away from a strong right to evoke. For example, the “look and feel” doctrine, which initially threatened to fence off highly abstract features of copyrighted works, has matured into a fairly narrow theory of limited applicability.18 Just as significantly, courts in copyright cases have repeatedly invoked a series of limiting doctrines designed to preserve a rich public domain. The idea/expression dichotomy,19 the scenes-a-faire doctrine,20 and the merger rule21 all enable courts to carve out dimensions of creative works that should not be owned. Fair use allows the copying of protected material if necessary to achieve a laudable goal.22 By applying a robust version of these doctrines in [*PG295]look-and-feel cases, most courts have resisted a strong right to evoke in copyright law.
The absence of such muscular limits in trademark and right of publicity law has fostered the growth of a right to evoke, to the clear detriment of the public. As intellectual property laws extend beyond the protection of particular, identifiable elements and into the more metaphysical realm, they threaten to overprotect in ways that stifle speech.23 In addition, as parties succeed in enjoining behavior based on the use of non-proprietary signals that make people think of them, they risk chilling not only those who wish to appropriate their work or to trade on their reputation, but also those whose evocation is purely referential or even unknowing.24 An exclusive right to evoke therefore should concern us.
This Article seeks to call attention to the right to evoke and to identify the costs associated with its expansion. “Right to evoke,” as used in this Article, means a right to prevent others from calling to mind a particular piece of intellectual property even if they have not replicated the intellectual property or deceived the public in any way. The critique therefore targets evocation as a right in and of itself, and not as a piece of evidence to be considered in an infringement suit. Thus, one work’s evocation of another is certainly relevant to proof of substantial similarity in a copyright case,25 but it does not alone establish that a defendant has wrongfully appropriated copyrighted expression.26 Likewise, to dilute a famous mark, a trademark should come closer than merely calling the mark to mind,27 and evocation of a per[*PG296]son’s identity should be a necessary, but not sufficient, prerequisite to a right of publicity claim.28
To be sure, courts have recognized the public’s affirmative right to evoke trademarks or personal identities in certain circumstances. So, for example, individuals may not only evoke, but also use trademarks for parody29 or criticism;30 parties may employ others’ trademarks for comparative advertising or other “fair” uses;31 and reporters may evoke individuals’ identities for news reporting without violating their right of publicity.32 Each of these cases, however, involves defendants who, for one reason or another, warrant special license to engage in behavior that would otherwise infringe. My point here is that by allowing evocation to serve as a proxy for infringement of trade[*PG297]mark and publicity rights, courts risk unduly broadening the kinds of behavior that one needs a license to commit.
Judging from a recent decision of the United States Supreme Court, the tide of trademark law may well be turning away from a strong-form right to evoke. As this Article was going to press, the Court, in Moseley v. V Secret Catalogue, Inc., announced that “the mere fact that consumers mentally associate [a] junior user’s mark with a famous mark is not sufficient to establish actionable dilution.”33 If lower courts follow the Supreme Court’s cue in Moseley, trademark law may follow copyright in limiting the legal consequences of evocation. Because Moseley rested on an interpretation of the FTDA, however, it reserved no affirmative evocation right to the public, and has no impact on state dilution law or the right of publicity.34
Part I describes the evolution of the right to evoke in copyright, right of publicity, and trademark law.35 It begins with copyright, in which courts initially flirted with a right to evoke, but later backed off by applying a series of doctrines designed to make copyright true to its normative goals.36 After describing this progression, Part I turns to right of publicity and trademark, in which a right to evoke has been alive and growing.37 Part II outlines the dangers of a right to evoke—including its overbreadth and the risk that it will chill legitimate behavior—and contends that the public has an affirmative interest in evoking cultural reference points in our society.38 Part III suggests a number of vehicles through which courts might accommodate these interests in trademark and right of publicity cases.39
Two caveats are in order here. First, this Article is preliminary and exploratory. I do not pretend to catalog every case recognizing a right to evoke, nor do I discuss all of the threats presented by such recognition. My goals are more modest—to raise consciousness of the existence of a trend and to warn of its dangers. Second, despite my concerns about a right to evoke, my opposition is not absolute; I can [*PG298]imagine circumstances in which an evocation might offend the interests that our intellectual property laws are designed to promote. Such cases, however, will be rare, and courts must do a better job of guarding against the other, far more common, cases in which the public interest in evocation outweighs the interests of intellectual property holders in the protected aspects of their “property.”
“Right to evoke” is not a term used by courts in intellectual property cases. Instead, courts describe copyright infringement in terms of the rights granted by section 106 of the Copyright Act,40 depict trademark entitlements as non-exclusive rights to prevent confusion and/or dilution of protected marks,41 and say that the state-law right of publicity applies only if a defendant has appropriated an individual’s identity for commercial gain.42
Yet as intellectual property doctrine has evolved, courts have increasingly loosened the strictures of what it means to copy, to confuse, and to appropriate.43 I am not the first to observe this phenomenon; countless scholars have traced the expansionist trends of copyright,44 trademark,45 and right of publicity46 doctrine. This Part critiques a [*PG299]particular piece of that expansion: the use of evocation as a proxy for proof of violation of a right under copyright, right of publicity, or trademark law.47
Historically, copyright law did not support claims based on mere evocation.48 The standard for infringement—which required both copying and substantial similarity—seemed to contemplate claims only against parties that replicated identifiable, tangible, and protectable features of copyrighted works.49 Although the law protected against the copying of non-literal expression (such as characters and plot structures),50 plaintiffs were required to identify with precision the high-level expression that the defendant appropriated.51 A plaintiff claiming substantial similarity in plot structure, for example, had to convince a fact-finder that the defendant’s plot replicated the original components of the plaintiff’s own narrative flow.52
Beginning in the late 1970s, however, a series of cases began eroding the traditional requirement of a particularized showing of similarities between works. In the first such case, the U.S. Court of Appeals for the Ninth Circuit, in Sid & Marty Krofft Television Productions, Inc. v. McDonald’s Corp.,53 held that a plaintiff could prove infringement based solely on similarities in the “total concept and feel” of two works, even if the expressive details were quite different. Krofft itself involved a claim of infringement of the look and feel of a chil[*PG300]dren’s television show.54 The two-step test that the court adopted55 virtually invited fact-finders to find infringement based on evocation.56
A decade later, the Ninth Circuit, in Roth Greeting Cards v. United Card Co.,57 re-endorsed “total concept and feel” protection in a case involving the “mood” of a series of greeting cards.58 The defendant in the case had used its own artwork on its cards, and had copied text too inconsequential to merit copyright protection.59 All of the particularized expression in the two lines of cards was thus either unprotected or not copied.60 The court nonetheless found infringement because “the characters depicted in the art work, the mood they portrayed, the combination of art work conveying a particular mood with a particular message, and the arrangement of the words on the greeting card are substantially the same . . . .”61
Like Krofft, Roth essentially recognized an exclusive right to evoke in copyright law.62 Under the “total concept and feel” doctrine as originally conceived, a party could prevail in a copyright case without identifying any protected similarities between a copyrighted and an infringing work; a plaintiff only needed to convince a fact-finder that the works exuded a similar mood.63
[*PG301] To some extent, this exclusive right to evoke in copyright law arose because of the difficulty in articulating exactly what copyright law protected, and why. The United States Supreme Court has stated repeatedly that copyright in the United States exists for utilitarian purposes—”to stimulate artistic creativity for the general public good.”64 Thus, copyright extends only to “intellectual conceptions of the author,”65 and protects only against the copying of the author’s contributions.66 Yet deciding what kinds of original contributions merit protection had long flummoxed the courts.67 Learned Hand’s classic formulation of copyright’s scope—that it extends to expression, but not ideas68—only begged the question of how to draw a distinction between the two.69 In the absence of any clear guidance, the [*PG302]early look-and-feel courts concluded that the fairly abstract “feel” or “mood” of a work could constitute protected expression.70
A number of recent developments, however, have suggested a greater willingness of courts to interpret “expression” more strictly and thereby to swing the pendulum away from a right to evoke in copyright law. Perhaps stimulated by the Supreme Court’s decision in Feist Publications, Inc. v. Rural Telephone Service Co.,71 holding that actionable copying requires “copying of constituent elements of the work that are original,” lower courts have used greater rigor in evaluating look-and-feel infringement claims.72 The Ninth Circuit, in particular, has modified the intrinsic/extrinsic approach laid out in Krofft to require an initial showing of similarities between expressive elements of two works before a case can reach the jury.73 In addition, if a plaintiff rests a claim only on similarities in the combination of otherwise unprotectable features, the Ninth Circuit has required “virtual identity” in selection and arrangement.74 Other jurisdictions have similarly in[*PG303]sisted that plaintiffs identify with particularity similarities in the original and allegedly infringing works, even in cases involving look-and-feel claims.75
Complementing this stricter conception of “expression,” courts have relied upon copyright’s limiting doctrines to narrow the scope of look-and-feel claims. They have used merger,76 scenes-a-faire,77 and the originality requirement78 to filter out unprotected aspects of works in look-and-feel cases. The cumulative effect of these doctrines is that a plaintiff in a copyright case must persuade a court not merely that the defendant has done something to remind the public of the plaintiff’s work, but that the defendant has actually copied an identifiable component of the plaintiff’s original expression.
The right of publicity79 is, at core, a business right to control use of one’s identity in commerce.80 Although courts and commentators [*PG304]have offered various rationales for the right, most modern cases defend it on some combination of incentive, just deserts, and unjust enrichment grounds: celebrities should have an incentive to develop valuable public personas, their efforts should be rewarded, and others deriving financial benefits from those valuable identities should pay.81 Publicity claims arise most often in celebrity endorsement cases, but the right extends to any use of a celebrity image calling attention to commercial interests or products.82 The publicity right originally applied only to the use of a celebrity’s name or photograph; over time, however, courts extended it to the use of a celebrity’s “identity,”83 reasoning that to limit the right to the use of particular features of the celebrity would defeat the objective of giving celebrities “a protected pecuniary interest in the commercial exploitation of [their] identit[ies].”84
Even before White v. Samsung Electronics America, Inc., courts found that use of symbols, phrases, look-alikes, or other devices to evoke the image of a celebrity in advertisements violated the right of publicity. Motschenbacher v. R.J. Reynolds Tobacco Co.,85 for example, involved a claim by a racecar driver who objected to an advertisement that featured a car with markings similar to his, along with an indistinct im[*PG305]age of a driver.86 The Ninth Circuit in Motschenbacher held that, even though “the ‘likeness’ of plaintiff is itself unrecognizable,” a publicity claim could proceed because the car’s “markings were not only peculiar to the plaintiff’s cars but they caused some persons to think the car in question was plaintiff’s and to infer that the person driving the car was the plaintiff.”87 The U.S. Court of Appeals for the Sixth Circuit, in Carson v. Here’s Johnny Portable Toilets, Inc.,88 found Johnny Carson’s right of publicity violated by a portable-toilet company operating under the name “Here’s Johnny.”89 And the New York Appellate Court, in Lombardo v. Doyle, Dane & Bernbach, Inc.,90 held that a “combination of New Year’s Eve, balloons, party hats, and ‘Auld Lang Syne’ . . . might amount to an appropriation of [Guy Lombardo’s] carefully and painstakingly built personality.”91 Over time, the courts appeared to be converging on a simple rule: if a commercial actor ran an advertisement that obtained value by referencing a celebrity, that commercial actor should pay.92
White, therefore, involved not so much a departure as a fortification and detailed articulation of a trend that had begun twenty years before. The facts of the case are well known. Samsung, [*PG306]the electronics manufacturer, had run a series of futuristic ads intended to show how its products would fare over time in comparison to various pop culture references.93 One of the ads, for Samsung VCRs, showed a robot, “dressed in a wig, gown, and jewelry,” standing “next to a game board which is instantly recognizable as the Wheel of Fortune game show set, in a stance for which Vanna White is famous.”94 A caption read, “Longest-running game show. 2012 A.D.”95 Vanna White sued, claiming that the ad violated her publicity rights.96 The district court granted summary judgment for Samsung, but the Ninth Circuit reversed.
The Ninth Circuit agreed that Samsung did not make use of White’s name or likeness, but it nonetheless held that White had alleged facts sufficient to support a publicity claim.97 True, the robot did not replicate any of White’s features or use her name or voice; nonetheless, the court announced, “[t]he right of publicity does not require that appropriations of identity be accomplished through particular means to be actionable.”98 Indeed, by limiting the right to particular types of appropriation, “we would not only weaken the right but effectively eviscerate it” because “[t]he identities of the most popular celebrities are not only the most attractive for advertisers, but also the easiest to evoke without resorting to obvious means such as [*PG307]name, likeness, or voice.”99 By evoking White’s image, the court held, Samsung had exploited the value of White’s celebrity, and therefore it should pay.100
Despite two hefty dissents101 and rancorous scholarly criticism,102 White remains the law, at least in the Ninth Circuit. Although other courts have discredited or distinguished certain aspects of the Ninth Circuit’s reasoning in White,103 none has entirely abandoned the notion that a celebrity has a claim whenever an advertiser gets commercial value from evoking the celebrity’s persona. For example, the U.S. Court of Appeals for the Sixth Circuit, in Landham v. Lewis Galoob Toys, Inc.,104 considered a movie actor’s claim that a toy based on his movie character violated his right of publicity.105 The court announced that the right of publicity “is now generally understood to cover anything that suggests the plaintiff’s personal identity,”106 but held that in that case the actor’s identity was not suggested because the public did not associate him with his character.107 Had the public made such an association, the plaintiff would presumably have had a publicity claim.108
[*PG308] As it stands, then, the right of publicity extends to any commercial, unauthorized use of a device or symbol—including but not limited to the individual’s name, likeness, or voice—that brings to mind a celebrity.109 Although theoretically the use must also commercially benefit the defendant, courts generally assume that if the celebrity link is made, such a benefit exists.110 The defendant need not use any particular attribute of the celebrity or deceive the public into thinking that it has.111 Evocation, alone, is enough.112
For most of its history, federal trademark law protected only against misleading or deceptive uses of confusingly similar marks.113 It was well accepted that a trademark holder had no rights against par[*PG309]ties whose marks merely evoked the protected mark.114 Indeed, at least with respect to trade dress, some courts viewed it as desirable to allow competitors to use enough of the trade dress to evoke a competing product, as long as the overall appearance of the product did not create confusion.115
The FTDA shifted this landscape. Intended to protect “famous marks” against loss of their distinctiveness,116 the FTDA created a new cause of action that no longer depends upon the risk of customer deception. The FTDA protects famous, distinctive marks against commercial uses of other marks that “lessen[]” the famous mark’s capacity “to identify and distinguish goods or services, regardless of the presence or absence” of competition among the parties or any confusion as to source.117 The House report offered several examples of uses that could, over time, reduce a mark’s uniqueness—”the use of [*PG310]DUPONT shoes, BUICK aspirin, and KODAK pianos would be actionable under this legislation.”118
In the eight years since the FTDA’s passage, the courts of appeals have had differing views on everything from the degree of distinctiveness required for dilution protection119 to the requisite level of fame120 to the question of whether the statute applies to claims be[*PG311]tween competitors.121 The courts of appeals have not yet resolved whether the statute protects trade dress,122 and despite a recent United States Supreme Court decision, the standard for proving dilution remains elusive.123 Any resolution of the scope of protection against dilution thus appears quite a ways off.
Nonetheless, at least in some early interpretations, the FTDA appeared to give trademark holders a broad right to obtain an injunction merely by showing that someone else’s mark brought their mark to mind.124 A number of appellate decisions suggested that any symbol that reminded the public of a trademark posed a threat to that mark’s distinctiveness.125 The U.S. Court of Appeals for the Second Circuit, for example, stated that to dilute, a mark “must be of sufficient similarity so that, in the mind of the consumer, the junior mark will conjure an association with the senior.”126 Although the court considered other factors in evaluating the evidence of dilution, the focus of the inquiry remained whether the junior use would “remind” the public of the senior mark.127 At the other extreme, the U.S. Court of Appeals of the [*PG312]Fourth Circuit refused to find dilution unless a plaintiff showed an actual, consummated adverse impact on its mark’s selling power as a result of the junior use.128
These two divergent interpretations of the dilution standard reflected radically different conceptions of Congress’s goals in enacting the FTDA. On the one hand, a standard requiring proof of actual, consummated injury views the dilution statute as targeting only cases in which loss of selling power has already occurred, and imposes a virtually insurmountable hurdle of proof: even holders of distinctive, famous marks would have to wait for evidence of loss of brand recognition before bringing an injunction against use of the same mark by another party.129 On the other hand, a standard of dilution based solely on a mark conjuring another reflects a view that a famous mark has a singular association in people’s minds and is entitled to protection against any background noise that might interfere with the purity of that association.130 The upshot is that the owner of a famous mark [*PG313]would have in gross property rights, not only in the mark itself but in a broad penumbra of associated words.131 The loose dilution standard, in other words, represents a strong right to evoke.
In contrast to these extreme interpretations of the FTDA, some courts of appeals adopted a more nuanced view of the FTDA’s objectives. The U.S. Court of Appeals for the Third Circuit, for example, described the overall objective of dilution analysis as protecting not against attenuated weakening of brand value but against the loss of the singularity of a famous mark: “Blurring occurs when the defendant’s use of its mark causes the public to no longer associate the plaintiff’s famous mark with its goods or services; the public instead begins associating both the plaintiff and the defendant with the famous mark.”132 In this view, a trademark holder—of even a famous mark—is not entitled to absolutely pure, interference-free connections; although trademark holders may prevent use by others that will eventually make their marks lose their singularity, the circle should not be drawn beyond users of the mark itself and others that the public views as the same.133
[*PG314] The United States Supreme Court, in Moseley v. V Secret Catalogue, Inc., adopted a version of this intermediate view of the FTDA. Although the Court agreed doctrinally with the Fourth Circuit that the FTDA requires “actual” rather than “likelihood of” dilution, for practical purposes the decision forged a middle ground between the two standards.134 In Moseley, the owners of the “Victoria’s Secret” mark sued a party that adopted “Victor’s Little Secret” as the name of an adult gift store in Kentucky. The plaintiffs claimed that defendants’ use of “Victor’s Little Secret” both blurred and tarnished the “Victoria’s Secret” mark.135 The Sixth Circuit had ruled in favor of the plaintiffs, finding that consumers hearing the name of the gift store were “likely automatically to think of the more famous store and to link” the two.136 Finding such linking “a classic instance of dilution by tarnishing (associating the Victoria’s Secret name with sex toys and lewd coffee mugs) and by blurring (linking the chain with a single, unauthorized establishment),” the Sixth Circuit upheld summary judgment for the plaintiffs.137 In reaching this conclusion, the court rejected the Fourth Circuit’s requirement that plaintiffs prove consummated loss of selling power to prevail in an FTDA claim.138
The Supreme Court granted certiorari to resolve the circuit split over the “actual harm” standard.139 Noting that the FTDA applies to uses that “‘cause[] dilution of the distinctive quality’ of the famous mark,” the Court held that the statute “unambiguously requires a showing of actual dilution, rather than a likelihood of dilution.”140 A plaintiff, in other words, must show that a defendant’s use in fact dilutes its mark, rather than merely that some future dilution is likely.141 Although the Court offered little guidance on how one might prove “actual dilution,” it made clear that “the mere fact that consumers [*PG315]mentally associate the junior user’s mark with a famous mark is not sufficient.”142 According to the Court, “[S]uch mental association will not necessarily reduce the capacity of the famous mark to identify the goods of its owner, the statutory requirement for dilution under the FTDA.”143
While nominally adopting an actual dilution standard, however, the Court liberally constructed the meaning of “actual dilution.”144 The Court not only rejected the Fourth Circuit’s requirement that parties prove loss of sales or profits, it also suggested that dilution did not depend on survey or other evidence of the actual effect of a challenged use in the marketplace.145 Instead, the Court indicated that “actual dilution can reliably be proven through circumstantial evidence—the obvious case is one where the junior and senior marks are identical.”146
By suggesting that circumstantial evidence can be used to prove actual dilution—and by specifically stating that surveys proving loss of distinctiveness are unnecessary—the Court fell short of requiring proof of consummated dilution.147 In other words, if a plaintiff need not introduce evidence that a famous mark has suffered an actual loss of distinctiveness in consumers’ minds, or a loss of value associated with such loss of distinctiveness, then the plaintiff can prevail in a FTDA claim without direct proof of actual dilution.148 On the other hand, the plaintiff must show that the challenged use, either by its nature or through its actual effect, causes a loss of the famous mark’s singularity.149
Thus, under the Court’s suggested approach, proof of dilution requires either that a defendant’s use by its very nature reduces the sin[*PG316]gularity of the famous mark (as when the defendant uses an identical mark) or that a defendant’s use actually reduces the singularity of the famous mark (by, for example, reducing its selling power, as proven through surveys or direct financial evidence).150 The Court in Moseley has therefore left the FTDA intact, but deliberately rejected a strong right to evoke.
The above analysis explores some of the ways in which the law empowers people to prevent others from evoking their protected intellectual property. Assuming the case has been made, however, why should we care?
A rich body of scholarship points out many of the positive reasons that speakers need an ability to evoke cultural associations to make effective speech. Rochelle Dreyfuss, for example, has made a strong case for using trademarks themselves to evoke particular images,151 and even the U.S. Court of Appeals for the Ninth Circuit recently held that it was fair game to make fun of Barbie.152 Numerous scholars—and many courts—have emphasized the importance of a rich public domain filled with the tools for future authorship.153 Our culture would suffer if we were prevented not only from using these images, but from reminding people of them.
There are other good reasons to avoid an abstract right to evoke, even if we accept existing limitations on the public’s right to use identifiable, tangible features of existing intellectual property. First, as rights extend from the tangible to the ethereal, speakers will have a more difficult time evaluating whether their contemplated speech violates someone’s rights. The resulting uncertainty will undoubtedly [*PG317]chill speech.154 Second, as the scope of rights increases beyond features identified solely with the intellectual property at issue, we run the risk of generating conflicting claims among multiple stakeholders.155 We are already experiencing a conflict of interests between copyright holders and actors who play roles in their copyrighted works.156 Those conflicts could potentially multiply if certain words, marks, or symbols remind the public of multiple parties. Finally, particularly in the trademark realm, the public frequently benefits when a new market entrant uses product trade dress to evoke the strong trade dress of an entrenched market participant.157 So long as there is no confusion, however, the evocation serves an important public policy goal of market competition. With dilution protection for trade dress, some of this desirable activity might subside.158
[*PG318] In contrast with its negative effect on public discourse, an unencumbered right to evoke may not promote the core philosophical objectives of trademark or right of publicity laws. Take the right of publicity: under a utilitarian theory, celebrities should have an incentive to develop personae pleasing to the public.159 As scholars have pointed out, however, celebrities arguably have such an incentive anyway, without any right of publicity.160 In any event, it is at best questionable whether a right to profit from evocations would add measurably to whatever incentive the right of publicity already provides. The other theoretical justifications for the right—natural rights theory and unjust enrichment—should involve some consciousness of whether the alleged violator has taken something of value that the celebrity was responsible for creating. In many (though not all) cases, others are primarily responsible for the value of symbols or other reference points that happen to evoke celebrities.161 In light of these competing concerns, as well as the speech interests of the public outlined above, courts considering publicity claims based on mere evocation should at least engage in some balancing to determine whether the right of publicity should prevail in a particular case.162
[*PG319] Similar considerations apply to dilution. To the extent that dilution law seeks to prevent junior users from commandeering famous marks such that consumers no longer can identify their source,163 that problem can be adequately addressed by applying dilution law only against marks that are identical or legally equivalent to a famous mark.164 The FTDA, after all, was intended to protect not only owners of famous trademarks, but also the public.165 Extending dilution rights to all marks bearing any associational relationship to famous ones exceeds the narrow objectives of the statute, at a significant societal cost. To be sure, such an interpretation arguably protects the purity of the famous mark’s recall with consumers; but the existence and exclusivity of the recall is not threatened. In comparison with the public interests outlined above, the added value of a right to evoke seems insignificant.
Constraining the right to evoke requires neither legislative action nor radical judicial re-interpretation of existing law. To the contrary, as in the copyright context, courts could easily cabin the expansion of an evocation right merely by giving force to the limits inherent in trademark and right of publicity doctrines. The following are some possible vehicles to move in that direction.
Two generalities apply to both trademark and right of publicity cases. First, courts should resist the temptation to equate evocation with infringement, absent some rationale to do so in a particular case.166 Second, even if such a rationale exists in a particular case, courts should not accept it blindly, but should consider whether the public’s interest in evocation outweighs the intellectual property holder’s concern in that case.
[*PG320] My specific doctrinal suggestions are somewhat more guarded. Courts might consider, in right of publicity cases, limiting an evocation right to cases involving likelihood of confusion as to endorsement. In other words, if a commercial actor does not use a name or likeness but merely evokes a celebrity’s identity, the publicity right should apply only if the public is likely to believe that the celebrity endorsed the advertisement or product at issue.
In the dilution context, Moseley v. V Secret Catalogue, Inc.167 has arguably begun a move away from a right to evoke, but more remains to be done. For one thing, Moseley only applies to the FTDA, and many state dilution laws have been invoked in cases involving mere evocation.168 Even in federal cases, however, Moseley’s lack of clarity leaves some open questions. For one, the meaning of “dilution” remains unclear. In light of the legislative history of the FTDA, it appears that courts should seek to distinguish between diluting loss of singularity, on the one hand, and mere background noise, on the other.169 By articulating dilution’s objective as the preservation of singularity of famous marks, courts would not only avoid unjustified intrusions into the public domain, but would arguably limit dilution claims to those that Congress appears to have contemplated in passing the FTDA.170 One effect of this approach might be to require near identity between marks before allowing a dilution claim. In cases involving non-identical marks, Moseley appropriately suggests a need for proof of actual, consummated dilution because of the risk of unduly restricting the public domain under any approach that allows a presumption of actual dilution.171 Even with identical marks, this approach would counsel in favor of a broad version of the “nominative fair use” doctrine, to make clear that uses referring to the trademark holder itself do not reduce the singularity of the mark because they reinforce, rather than weaken, association between a mark and its source.172
[*PG321] Beyond clarifying the objective of the dilution inquiry, courts should be mindful of the positive reasons for allowing parties to evoke others’ products or services in pro-competitive ways. They might also weigh the public’s interest in such evocation in certain cases or with respect to certain types of marks. With trade dress, for example, it seems likely that the positive benefits of evocation outweigh the interest in protecting against dilution.173 For that reason, courts should not presume that Congress intended to extend dilution protection to trade dress.
We should be concerned about the trend toward allowing intellectual property holders to prevent others from using non-proprietary words, expression, or symbols to evoke their intellectual property. This Article identifies the trend and suggests some of the reasons for concern. It establishes that a right to evoke is not a notion limited to a rogue opinion of the U.S. Court of Appeals for the Ninth Circuit or to right of publicity law. If we wish to preserve a rich commons and avoid significantly chilling free expression, courts should at least cabin the right to evoke and ensure that, when utilized, it serves the law’s normative goals.SERTED BLANK PAGE