In Alan Jay Lerner’s lyric, Professor Henry Higgins laments: “Why can’t a woman be more like a man?”1×1. Alan Jay Lerner & Frederick Loewe, A Hymn to Him, in My Fair Lady (1956). Professor Rebecca Tushnet’s provocative article2×2. Rebecca Tushnet, Registering Disagreement: Registration in Modern American Trademark Law, 130 Harv. L. Rev. 867 (2017). in effect urges that a U.S. trademark should be more like a European Union trademark, at least with respect to the relationship of registration to substantive protection.3×3. Other aspects of the European Union trademark regime may be less compelling — for example the regime’s broader protection against taking unfair advantage of the repute of the trademark. See Council Regulation 207/2009, on the Community Trade Mark, art. 9(2)(c), 2009 O.J. (L 78) 1, 5 (EC) (as amended by Regulation 2015/2424, 2015 O.J. (L 341) 21 (EU)) [hereinafter EUTMR]; see also infra note 56. The article convincingly exposes the current incoherence in U.S. trademark law — a hybrid between “procedural” and “substantive” registration regimes, in which the traditional emphasis on use-based trademark rights undermines the business-planning benefits that flow from registration.
Before elaborating on the similarities between Tushnet’s suggested reforms of U.S. trademark law and current EU trademark law, this Response will articulate the premises underlying use-based and registration-based systems, premises implicit in Tushnet’s arguments, but which may not be apparent to readers unfamiliar with trademark law. This Response will then address ways in which the EU trademark registration system offers a model to implement some of Tushnet’s prescriptions.
I. The Underlying Premises of Use-Based and Registration-Based Systems
Use-based and registration-based systems rely on distinct, and opposing, underlying conceptions of trademark rights. At one extreme, exemplified by use-based systems, trademark law derives from the traditional common law action of passing off; enforcement of rights in the mark protects the public against a competitor’s attempts to mislead the public regarding the source or qualities of goods. The trademark owner’s claims are proxies for the public interest in honest marketing. But a trademark fills its role of advancing the public’s interest in avoiding confusion or deceit only if the public perceives that the trademark symbolizes a particular producer’s goods. Without the consumer-perception link, the symbol does not “identify and distinguish”4×4. See Lanham Act § 45, 15 U.S.C. § 1127 (2012) (defining “trademark”). one producer’s goods, and it therefore is not a trademark. The link arises from the use of the mark in trade. As the U.S. Supreme Court emphasized in 1916 in Hanover Star Milling Co. v. Metcalf5×5. 240 U.S. 403 (1916). :
[A] man’s right to the continued enjoyment of his trade reputation and the good-will that flows from it, free from unwarranted interference by others, is a property right, for the protection of which a trade-mark is an instrumentality. . . . [T]he right grows out of use, not mere adoption. . . . [T]he trade-mark is treated as merely a protection for the good-will, and not the subject of property except in connection with an existing business.6×6. Id. at 413–14; see also Trade-Mark Cases, 100 U.S. 82, 94 (1879) (“At common law the exclusive right to it grows out of its use, and not its mere adoption.”).
At the other extreme, exemplified by registration-based systems, trademarks are property rights arising out of the government’s acceptance of the mark for registration.7×7. See EUTMR, supra note 3, art. 6. Use is not a prerequisite to registering a distinctive sign, though subsequent failure to exploit the mark in trade provides a basis for cancelling the registration.8×8. See id. arts. 15(1), 51(1)(a) (requiring genuine use within five years of registration but in fact allowing a mark to be maintained without use unless someone applies to have it revoked). As in use-based systems, the trademark is a distinctive sign that identifies the goods and their producer;9×9. See id. art. 4. trademarks therefore protect both the producer’s goodwill and the public interest, but it is the registration that creates the right. Registration attests to the sign’s potential to link public perception and the registrant,10×10. See id. art. 7 (declaring a sign’s inability to serve as a trademark if it is “devoid of any distinctive character,” id. art. 7(1)(b), or is a generic term). but acquisition of rights in a distinctive sign is not contingent on demonstrating that the mark has in fact acquired secondary meaning. Thus, an inherently distinctive mark need not have entered the marketplace (much less proved itself there) before legal rights attach. Equally importantly, registration defines the existence and largely delineates the extent of the rights: a sign denied registration — for example for lack of distinctiveness — is not an object of protection. Moreover, the “rule of speciality”11×11. See Joan Anfossi-Divol, L’Usage et L’enregistrement, Éléments Essentiels de L’harmonisation du Droit des Marques: Une Approche Comparative des Droits Franco-Communautaire et des Etats-Unis d’Amérique 138 (2003) (“[B]y application of [European] community law, registration is the sole source of rights, the rule of speciality circumscribes [rights] to the products or services for which the mark was registered.” (translated from French)). See generally id. at 138–63 (comparing the scope of the property right in registration-based and use-based systems). limits the scope of the rights to the goods or services named in the registration.12×12. Unless the mark is well known and qualifies for protection against dilution under EUTMR art. 9(2)(c). Moreover, the greater the range of goods or services named in the registration, the less restraining is the rule of speciality; on the other hand, the registration is subject to cancellation as to goods or services named in the registration but for which the registrant is not in fact using the mark after five years. Thus, a use of a mark on goods other than those named in the registration provides no basis to claim trademark infringement by a party who has adopted the same or similar mark for the unnamed goods;13×13. Except for famous marks receiving dilution protection under EUTMR art. 9(2)(c). in a use-based system, courts would inquire if the actual use of the mark (registered or not) has led to its public perception as a trademark for those goods, so that use by other parties would be likely to confuse the public concerning the source of marked goods.14×14. Similarly, Lanham Act section 2(d) directs that registration may be refused if the mark “[c]onsists of or comprises a mark which so resembles a mark registered in the Patent and Trademark Office, or a mark or trade name previously used in the United States by another and not abandoned, as to be likely, when used on or in connection with the goods of the applicant, to cause confusion, or to cause mistake, or to deceive.” 15 U.S.C. § 1052(d) (2012) (emphasis added). Dilution of a famous mark is a ground for opposition to registration. See Lanham Act § 13(a), 15 U.S.C. § 1063(a). In theory, use-based systems’ flexibility focuses primarily on the consumer while registration-based systems’ greater formalism favors enterprises because it ensures predictability in the acquisition and scope of rights.
As Tushnet explains, the United States has tried to have it both ways, grounding rights in use — hence preserving the defining role of the marketplace — but then undercutting the significance of use by enhancing a trademark’s scope through registration.15×15. Tushnet, supra note 2, at 870–71 (“American scholars . . . have often treated registration like a borrowed civil law coat thrown awkwardly over the shoulders of a common law regime. . . . It’s time to recognize that it’s our coat and to start making alterations so it fits better.”). The Lanham (Trademark) Act16×16. 15 U.S.C. §§ 1051–1072, 1091–1096, 1111–1127, 1141, 1141a–1141n (2012). provides incentives to register: perhaps most importantly by giving priority dating from filing rather than from first use in commerce, and also by making trademark rights enforceable nationwide17×17. Lanham Act § 7(c), 15 U.S.C. § 1057(c). (while common law rights would exist in only localities in which the mark is in fact used). Registration thus not only assists title holders, but also, by creating a publicly searchable record, aids other businesses’ planning by putting them on notice that the registrant claims the same or similar mark for the same or similar goods.18×18. Indeed, Congress shifted the Lanham Act’s emphasis toward security for enterprises when it instituted the intent-to-use system, Lanham Act section 1(b), enabling applicants to file on the basis of an intent to use, rather than on demonstrating actual use. See S. Rep. No. 100-515, at 4–6 (1988), as reprinted in 1988 U.S.C.C.A.N. 5577, 5581 (“The Lanham Act’s preapplication use requirement . . . creates unnecessary legal uncertainty for a U.S. business planning to introduce products or services into the marketplace. It simply has no assurance that after selecting and adopting a mark, and possibly making a sizable investment in packaging, advertising and marketing, it will not learn that its use of the mark infringes the rights another acquired through earlier use. In an age of national, if not global, marketing, this has a chilling effect on business investment.” Id. at 5.). As Tushnet also demonstrates, however, refusal to register does not necessarily prevent the unsuccessful applicant from using the mark and building up goodwill protectable by unfair competition claims, both at state law and under section 43(a) of the Lanham Act.19×19. Tushnet, supra note 2, at 878–80. The disparity between grounds for refusal to register and on-the-ground acquisition of rights undermines the utility of registration even as it honors the role of consumer perception in giving rise to trademark rights. Even the most significant recent development toward convergence of registration and enforceable rights in fact underscores the disconnect between the two regimes. In B & B Hardware, Inc. v. Hargis Industries, Inc.,20×20. 135 S. Ct. 1293 (2015). the Supreme Court ruled that a refusal to register on grounds of likelihood of confusion between the applicant’s mark and a mark previously registered or used in commerce could have preclusive effect in a subsequent infringement action.21×21. Id. at 1299. However, the Patent and Trademark Office’s (PTO) inquiry into likely confusion does not take into account how the marks are actually used in the marketplace, meaning that the decision conflates, rather than unifies, the two regimes.22×22. See Tushnet, supra note 2, at 893–99.
II. How the EU Registration-Based Trademark Regime Can Implement Tushnet’s Prescriptions
Tushnet’s solutions in several respects endorse the European approach to trademark registration, including the “double identity” rule, which bars the use of the same mark for the same goods as the registrant’s, without requiring a showing of likelihood of confusion.23×23. See id. at 932–34; EUTMR, supra note 3, art. 9(2)(a). Tushnet acknowledges the risk that the rule will prove overprotective in the absence of fair use and comparative advertising exceptions. Tushnet, supra note 2, at 68. The Court of Justice of the European Union (C.J.E.U.) has in fact tempered the double identity rule by establishing a requirement that the use have “an adverse effect on the functions of the trade mark” and then elaborating a range of elusively stated “functions.” Joined Cases C-236/08 to C-238/08, Google France SARL v. Louis Vuitton Malletier SA, 2010 E.C.R. I-2467, I-2502. In 2009, in L’Oréal SA v. Bellure NV, the C.J.E.U. explained:
These functions include not only the essential function of the trade mark, which is to guarantee to consumers the origin of the goods or services, but also its other functions, in particular that of guaranteeing the quality of the goods or services in question and those of communication, investment or advertising.
Case C-487/07, L’Oréal SA v. Bellure NV, 2009 E.C.R. I-5185, I-5187; see also Case C-323/09, Interflora Inc., v. Marks & Spencer plc, 2011 E.C.R. I-8625, I-8627 (addressing the function of “guaranteeing the quality of that product or service or that of communication, investment or advertising”); Google France, 2010 E.C.R. at I-2503 (addressing “the function of indicating origin and the function of advertising”); Case C-206/01, Arsenal Football Club plc v. Reed, 2002 E.C.R. I-10273, I-10316 (“[T]he essential function of a trade mark is to guarantee the identity of origin of the marked goods or services to the consumer or end user by enabling him, without any possibility of confusion, to distinguish the goods or services from others which have another origin.”). There are other features of the European Union Trade Mark Regulation (EUTMR) not discussed in Tushnet’s article, but whose adoption in the United States would be consistent with Tushnet’s prescriptions. The remainder of this Response will identify and compare those features with the current U.S. registration regime in light of Tushnet’s overall objective of making registration more “substantive.”24×24. Tushnet, supra note 2, at 932.
Tushnet advocates a more “substantive” role for registration not in order to strengthen the hand of trademark owners but on the contrary to reduce the potential for overreach by trademark owners that results from the current disparity between an administrative grant of rights through registration and judicial enforcement of registered and unregistered trademarks.25×25. Id. Among the reforms she urges, stricter assessment of the distinctiveness of the mark figures prominently.26×26. Id. at 921–26. In particular, she argues the PTO should more vigorously police the border between merely descriptive and inherently distinctive signs, for example, by “requiring the applicant to show real secondary meaning in the market.”27×27. Id. at 922. Under current U.S. trademark law, however, an applicant whose mark would be unregistrable on grounds of mere descriptiveness28×28. Lanham Act § 2(e), 15 U.S.C. § 1052(e) (2012). may seek registration on the basis that the mark has acquired distinctiveness, with “proof of substantially exclusive and continuous use thereof as a mark by the applicant in commerce for the five years before the date on which the claim of distinctiveness is made.”29×29. Id. § 2(f). According to the Trademark Manual of Examining Procedure (TMEP), “[t]he claim of five years of use is generally required to be supported by an affidavit or declaration under 37 C.F.R. § 2.20, signed by the applicant.”30×30. U.S. Patent & Trademark Office, U.S. Dep’t of Commerce, Trademark Manual of Examining Procedure § 1212.05(d) (Oct. 2016 ed.) [hereinafter TMEP] (citing 37 C.F.R. § 2.20 (2016)). The PTO may, but currently is not obliged to, seek proof not only of continuous use but also of acquired distinctiveness in fact.31×31. Id. §§ 1212, 1212.06. Thus, enforceable nationwide rights may attach to descriptive terms just on the applicant’s sworn statement of five years’ continuous use.32×32. In an infringement action, courts may seek proof of the actual strength of the mark; but as Tushnet underscores, some courts may treat the fact of registration as evidence that consumers perceive the sign as a trademark. See Tushnet, supra note 2, at 878–80. Moreover, as a result of a 1988 amendment, the five-year period need not have preceded the application date: “any five-year claim submitted on or after November 16, 1989, is subject to the new time period. This applies even if the application was filed prior to that date.”33×33. TMEP, supra note 30, § 1212.05 (citing Trademark Law Revision Act of 1988, Pub. L. No. 100-667, 102 Stat. 3935 (1988) (amending Lanham Act § 2(f))). Because rights in registered trademarks run from the date of the application,34×34. 15 U.S.C. § 1057(c). the amendment effectively confers trademark priority on signs that may not have been capable of being a trademark at the time of application.
Compare the EUTMR’s approach to merely descriptive signs that have allegedly acquired distinctiveness. Article 7(1)(c) bars registration of “trade marks which consist exclusively of signs or indications which may serve, in trade, to designate the kind, quality, quantity, intended purpose, value, geographical origin or the time of production of the goods or of rendering of the service, or other characteristics of the goods or services.”35×35. EUTMR, supra note 3, art. 7(1)(c). Article 7(3), however, provides: “Paragraph 1 . . . (c) . . . shall not apply if the trade mark has become distinctive in relation to the goods or services for which registration is requested in consequence of the use which has been made of it.”36×36. Id. art. 7(3). The European Union Intellectual Property Office (EUIPO) Guidelines for Examination of European Union Trade Marks’s section on acquired distinctiveness37×37. European Union Intellectual Prop. Office, Guidelines for Examination of European Union Trademarks, Part B Section 4, § 2.6 (2016) [hereinafter EUIPO Guidelines]. explains:
Distinctive character acquired through use means that although the sign ab initio lacks inherent distinctiveness with regard to the goods and services claimed, owing to the use made of it on the market, the relevant public has come to see it as identifying the goods and services claimed in the EUTM application as originating from a particular undertaking.38×38. Id. § 2.6.1.
The Guidelines require that the applicant prove “a link between the sign and the goods and services for which the sign is applied for, establishing that the relevant class of persons, or at least a significant proportion thereof, identify goods as originating from a particular undertaking because of the trade mark.”39×39. Id. § 2.6.5. Thus, the EUIPO demands what Tushnet would have the PTO require: placement of the burden of proof on the applicant to demonstrate through evidence of actual use that the sign has in fact acquired secondary meaning.40×40. Tushnet, supra note 2, at 923. In addition, as Tushnet observes, id. at 923–24 & nn.246–47, European authorities police against bootstrapping a “limping” mark to the registered mark in connection with which the weaker sign (often an element of trade dress) appears. See, e.g., Case C-215/14, Société des Produits Nestlé SA v. Cadbury UK Ltd, 2015 E.C.R. 604, 614 (citing Directive 2008/95/EC, art. 3(3), 2008 O.J. (L 299) 25, 27) (“[A]lthough the trade mark for which registration is sought [the three-dimensional form of the Kit Kat chocolate bar] may have been used as part of a registered trade mark or in conjunction with such a mark, the fact remains that, for the purposes of the registration of the mark itself, the trade mark applicant must prove that that mark alone, as opposed to any other trade mark which may also be present, identifies the particular undertaking from which the goods originate.”). Relevant evidence includes:
[T]he market share held by the mark in question; how intensive, geographically widespread and long-standing use of the mark has been; the amount invested by the undertaking in promoting the mark; the proportion of the relevant class of persons who, because of the mark, identifies goods or services as originating from a particular undertaking; and statements from chambers of commerce and industry or other trade and professional associations.”41×41. Joined Cases C-217/13 & C-218/13, Oberbank AG v. Deutscher Sparkassen- und Giroverband eV, 2014 E.C.R. 2012, 2019 (discussing Directive 2008/95/EC, art. 3(3), which addresses acquired distinctiveness in the same terms as EUTMR, supra note 3, art. 7(3)). In that case, a survey of relevant consumers showed almost 70% recognition of term as a service mark. Id. at 2015; see also Case C-542/07 P, Imagination Technologies Ltd v. OHIM, 2009 E.C.R. I-4937 (affirming Case T-461/04, Imagination Technologies v. OHIM, 2007 E.C.R. II-122 ¶¶ 70, 74, 79 (finding £30 million worth of sales of radios under the Pure Digital mark insufficient to prove acquired distinctiveness before date of application for registration because “the applicant should have shown not only use of the sign applied for, but in addition the fact that the relevant consumer understood the sign not as a descriptive term, but also as a trade mark,” id. ¶ 74)).
Moreover, in contrast to the 1988 amendment to the Lanham Act discussed above, the EUIPO stresses that “[t]he evidence must prove that distinctiveness through use was acquired prior to the EUTM application’s filing date.”42×42. EUIPO Guidelines, supra note 37, § 2.6.3; see also id. § 184.108.40.206. But see EUTMR, supra note 3, art. 52(2) (stating that if a mark is registered and is not inherently distinctive when cancellation proceedings are started, the proprietor can resist by showing it has subsequently acquired distinctiveness).
Finally, the EUIPO Guidelines elaborate a requirement that lacks a counterpart in the TMEP but that also responds to Tushnet’s concern that trademark registration in the United States may hamper competition because the PTO may be establishing rights in signs that do not in fact identify and distinguish the trademark owner’s goods or services.43×43. See Tushnet, supra note 2, at 918–21. Lanham Act registration confers nationwide rights from the date of filing; these benefits attach even to a section 2(f) registration that issues following the applicant’s affidavit of continuous use (without actual proof of acquired distinctiveness),44×44. See 15 U.S.C. § 1057(c) (2012). but neither the statute nor the TMEP requires that the applicant have used the mark throughout the United States. Thus, a U.S. trademark applicant could acquire nationwide rights in a descriptive term without showing that, throughout the United States, the mark in fact has come to “identify and distinguish” its goods or services in the marketplace.45×45. Id. § 1127. By contrast, the “unitary character” of the European Union Trade Mark46×46. EUTMR, supra note 3, art. 1(2); see also id. art. 7(2) (stating that if an absolute ground of refusal exists with respect to any part of the EU, registration must be refused as to the entire EU). implies that the sign must be distinctive throughout the EU.47×47. On the other hand, the EUTMR does not preempt national trademark registration in member states (nor use-based protection in member states that recognize rights on that basis); as a result, a mark that does not qualify for registration as an EU trademark may still be protected at the national level in some subset of the EU’s member states. The EUIPO draws the conclusion that “acquired distinctiveness must be established throughout the territory in which the trade mark did not, ab initio, have such character . . . . [T]he unitary character of the European Union trade mark requires a sign to possess a distinctive character, inherent or acquired through use, throughout the European Union.”48×48. EUIPO Guidelines, supra note 37, § 2.6.6 (citations omitted). An equivalent U.S. doctrine would similarly oblige the applicant not only to prove to the PTO that the mark had in fact come to symbolize the applicant’s goods or services, but also that the relevant consumers throughout the United States drew the necessary connection between the mark and the goods or services.49×49. Under the Dawn Donut doctrine, discussed in Tushnet, supra note 2, at 910–11, a federal trademark registrant’s nationwide rights are not enforceable against junior users operating in localities where the registrant is not yet doing business. See Dawn Donut Co. v. Hart’s Food Stores, Inc., 267 F.2d 358, 364 (2d Cir. 1959). More recently, however, courts may be more inclined to extend injunctive relief to “suppress junior users” (Tushnet’s phrase) in localities to which the registrant has not yet expanded. See Guthrie Healthcare Sys. v. ContextMedia, Inc., 826 F.3d 27 (2d Cir. 2016) (“[I]t does not follow . . . that a senior user who has proven entitlement to an injunction affecting one geographic area by reason of the junior user’s infringement must show the same high degree of probability of harm in every further area into which the injunction might extend, thus allowing the infringer free use of the infringing mark in all areas as to which the senior user has not shown a substantial probability of confusion.” Id. at 47).
Importing these EU rules would promote Tushnet’s goal of countering anticompetitive behavior because they would ensure registration of only marks of demonstrated distinctiveness. The EU approach also buttresses one of Tushnet’s related recommendations. She remarks that “requiring the applicant to show real secondary meaning in the market . . . would have positive effects on the ITU [intent to use] system as well, since fewer marks would qualify for ITU protection in the absence of use and market recognition.”50×50. Tushnet, supra note 2, at 922. This observation deserves further elaboration. A U.S. application based on intent to use a mark will qualify for registration once the applicant makes actual use of the mark, and if none of the Lanham Act’s bars to registration applies. Structurally, the mark must be inherently distinctive: if the mark is registrable immediately upon first use in commerce,51×51. 15 U.S.C. § 1051(d) (2012). then its registrability cannot depend on demonstration of distinctiveness acquired through some period of use.52×52. See TMEP, supra note 30, § 1212.09(a) (“Section 2(f), 15 U.S.C. § 1052(f), is limited by its terms to ‘a mark used by the applicant.’ A claim of distinctiveness under § 2(f) is normally not filed in a § 1(b) application before the applicant files an allegation of use, because a claim of acquired distinctiveness, by definition, requires prior use.”). But given the market advantages descriptive terms confer,53×53. See, e.g., Tushnet, supra note 2, at 918 & n.220 (citing Carol M. Rose, Crystals and Mud in Property Law, 40 Stan. L. Rev. 577, 596 (1988)). and because the line between inherent distinctiveness and mere descriptiveness often wavers,54×54. See 2 McCarthy on Trademarks and Unfair Competition § 11:18 (4th ed. 2016). the system does not necessarily discipline ITU applicants to select marks solidly within the spectrum of inherent distinctiveness. Were an applicant required to demonstrate that borderline marks had acquired distinctiveness, those marks could not be the subjects of ITU applications; the ITU applicant therefore would be obliged to select marks far from the line of demarcation. The line’s uncertainty, rather than encouraging applicants to approach it as closely as possible, would instead push applicants even further away from the zone of disqualified marks.
The EU approach would achieve a similar result. There are no ITU applications in the EU regime because use is not a prerequisite to registration of an inherently distinctive mark. Instead, the registrant has five years to make actual use in trade.55×55. See EUMTR, supra note 3, arts. 15(1), 51(1)(a). In effect, the EUTMR incorporates ITU for inherently distinctive marks. But if the mark lacks distinctiveness, then the mark cannot be registered unless the applicant proves that prior to filing the application, the mark had already acquired distinctiveness through actual use. As a result, no rights arise without actual use unless the sign is inherently distinctive.
In conclusion, while I doubt Tushnet would endorse importing every aspect of EU trademark law, particularly regarding the scope of protection of the mark once registered,56×56. For example, compared with U.S. trademark doctrine, courts in the EU have yet to develop a robust trademark parody exception. Compare Rebecca Tushnet, Make Me Walk, Make Me Talk, Do Whatever You Please: Barbie and Exceptions, in Intellectual Property at the Edge: The Contested Contours of IP 405, 406 (Rochelle Cooper Dreyfuss & Jane C. Ginsburg eds., 2014), with Robin Jacob, Parody and IP claims: A Defence? — A Right to Parody?, in Intellectual Property at the Edge, supra, at 427, 431. In addition, the scope of EU protection against dilution, applicable where
the [defendant’s] sign is identical with, or similar to, the EU trade mark irrespective of whether it is used in relation to goods or services which are identical with, similar to or not similar to those for which the EU trade mark is registered, where the latter has a reputation in the Union and where use of that sign without due cause takes unfair advantage of, or is detrimental to, the distinctive character or the repute of the EU trade mark,
EUTMR, supra note 3, art. 9(2)(c) (emphasis added), likely is too broad for Tushnet — no fan of the dilution doctrine, see generally Rebecca Tushnet, Gone in Sixty Milliseconds: Trademark Law and Cognitive Science, 86 Tex. L. Rev. 507 (2008) — to wish its importation into U.S. law. the EU Trademark Regulation and Directive offer a model of a “substantive” registration regime whose adoption could promote fairer competition by better policing the claimed sign’s role as a trademark.
* Morton L. Janklow Professor of Literary and Artistic Property Law, Columbia Law School. Thanks to Lionel Bently for comments on EU trademark law, and to Nathalie Russell, Columbia Law School class of 2018, for research assistance.