United States Patent and Trademark Office v. Booking.com B.V.: How Do We Know When Something Is a Name?

July 2, 2020

United States Patent & Trademark Office v. Booking.com B.V., 591 U.S. ___ (2020) (Ginsburg, J.).
Response by Laura A. Heymann
Geo. Wash. L. Rev. On the Docket (Oct. Term 2019)
Slip Opinion | SCOTUSblog

United States Patent and Trademark Office v. Booking.com B.V.: How Do We Know When Something Is a Name?

“The Container Store” is the name of a U.S. retail chain of stores that sell storage boxes, organizing supplies, and the like. The company currently holds a federal trademark registration for the mark “The Container Store,”¹ presumably based on the conclusion by the U.S. Patent and Trademark Office that when consumers encounter that phrase—perhaps typically with the benefit of the capitalization above—they understand it to be the name of the store and not simply a description. If this is true, should the fact that the mark consists of a generic term (“container store”) preceded by a definite article (“the”) preclude it from being recognized as a name? In other words, which should take precedence: consumer recognition of “The Container Store” as a trademark or concerns about competitors who want to describe themselves as “container stores” without fear of a lawsuit from The Container Store? Can these two concerns be reconciled?

United States Patent and Trademark Office v. Booking.com B.V.² is framed as a simple case on its facts. Booking.com is a company that offers travel reservation services over the Internet. It applied to register “Booking.com” as a trademark. The U.S. Patent and Trademark Office (“USPTO”) rejected that application on the grounds that “booking” is generic for the services that the company provides and that the addition of “.com” does nothing to change that status. The U.S. District Court for the Eastern District of Virginia, on review, concluded that consumers understood “Booking.com” to be descriptive of the company’s services and that the term had acquired secondary meaning and was therefore distinctive.³ The USPTO appealed only the determination that “Booking.com” is not generic; the U.S. Court of Appeals for the Fourth Circuit then affirmed the district court’s ruling.⁴ Thus, as the case came to the U.S. Supreme Court, the question was this: What should be the basis for determining whether a claimed trademark is, in fact, a generic term or a proper name—the difference between “apple” for fruit and “Apple” for computers?

The Court concluded that consumer perception is the touchstone. (Justice Ginsburg wrote the opinion for an eight-member majority, with Justice Sotomayor writing a separate concurring opinion and Justice Breyer dissenting.) Because a generic term is a term for “a class of goods or services,” and because consumers, according to the record below, “do not perceive the term ‘Booking.com’ to signify online hotel-reservation services as a class,” the term “Booking.com” is a name for a particular service rather than a generic term for such services as a group.⁵ In other words, noted the Court, because consumers don’t ask each other who their favorite Booking.com is, and don’t refer to competitors such as Travelocity as a Booking.com, the term “Booking.com” is not, as a whole, a generic term.⁶ Accordingly, the Court concluded that the U.S. Patent and Trademark Office erred when it refused registration for the term on the grounds that, as a rule, “[generic].com” is necessarily also generic.⁷

But the Court’s opinion isn’t as straightforward as it purports to be. Booking.com is the name of the corporate party to the case. It is also the name of a travel reservation service on the Internet. When news articles discuss the company, they refer to it as Booking.com. Booking.com is also, in shorthand form, the URL for the website where one can access the company’s services (the full URL of which is https://www.booking.com/). And the failure to clarify these distinctions, I think, is what complicates the Court’s opinion. Consumers might, for example, see “Booking.com” as a trademark, given the way that the company has marketed its services under this name, but would not see www.booking.com as a trademark. In much the same way, consumers might see Nike as a trademark, but not www.nike.com or www.sneakers.com (if that URL were to resolve to the Nike website). Indeed, in a truly hypothetical world, the company could have called itself “Booking.com” and sought a trademark registration for that name without ever having secured the domain name www.booking.com. (This decision, of course, would have been senseless from a consumer acquisition perspective, but I raise it here to highlight that although the name of the company is derived from its URL, the two are, in fact, distinct.)

Saul Kripke has written that a name functions as a “rigid designator”—in other words, it has a denotative function, conferred in a “baptismal” moment, that remains constant despite changes to the underlying characteristics of the thing that is named; it continues so long as a chain of reference does.⁸ As Michel Foucault once wrote, if “Pierre Dupont does not have blue eyes, or was not born in Paris, or is not a doctor, the name Pierre Dupont will still always refer to the same person; such things do not modify the link of designation.”⁹ Arguably, although the Court didn’t frame its opinions in these terms because of the First Amendment framework of the cases, its decisions in Tam¹⁰ and Brunetti¹¹ can be understood as consistent with such a theory—that concerns about the deeply offensive connotations associated with a trademark should not be relevant to whether the mark functions as a name. As Justice Kennedy wrote in his concurrence in Tam, “[t]he central purpose of trademark registration is to facilitate source identification. . . . Whether a mark is disparaging bears no plausible relation to that goal.”¹² The Court’s decision in Booking.com can be read to be consistent with this approach: the question of trademark validity is whether the term at issue facilitates source identification—that is, whether it is a proper name for the good or service at issue. Relying on the evidence below (unchallenged by the USPTO) that consumers saw “Booking.com” as the name of a company offering travel-related reservation services, the Court concluded that “Booking.com” is seen as a proper name, even though it comprises generic elements and, when part of a URL, functions as an address.

This is where I think both the majority and Justice Breyer, in his thoughtful dissent, get off track. The majority, for its part, writes that a [generic].com term can be source identifying because it might “convey to consumers a source-identifying characteristic: an association with a particular website,” given that “only one entity can occupy a particular Internet domain name at a time.”¹³ Certainly consumers understand, because of how domain names work, that a URL is associated with only one website. But that’s not the determination the USPTO was charged with making. The company didn’t apply to register “Booking.com” as a URL; it applied to register the term “Booking.com” as a mark for travel-reservation services. The fact that, in its function as a URL, “www.booking.com” can be associated only with one website doesn’t make a URL in and of itself a mark any more than the fact that there is only one building at 123 Main Street makes “123 Main Street,” without more, a mark for the services rendered in that building. Justice Breyer also takes up this framework, writing, “The meaning conveyed by ‘Booking.com’ is no more and no less than a website associated with its generic second-level domain, ‘booking.’”¹⁴ But here, the trademark isn’t “Booking”—it’s “Booking.com.” Consumers asked to name both a specific company that provides travel reservation services online and a specific company that allows users to sell goods online might well respond by saying “Booking.com” and “eBay”—but not “Booking” and “eBay.com.” The fact that both opinions (and the USPTO) seemed to agree that “tennis.net” would be a special case supports this point.¹⁵ “Tennis.net” is understood to be a special case because it is assumed that consumers would understand that the wordplay indicates a source-indicating name rather than simply a URL.

Thus, the implications of the case, in my view, are not what the Court had to say about names and source identification. The opinion reaffirms the concept that trademark validity is about source identification and that consumer perception is central to that determination. But consumer perception has to be assessed carefully. The Court accurately notes, citing an amicus brief filed by several trademark scholars, that surveys, for example, “require care in their design and interpretation.”¹⁶ Questions may be framed imprecisely, and the impulse to guess rather than admit lack of understanding may skew the results. Justice Breyer, for example, points to the survey results introduced by Booking.com in the case, which showed that 33 percent of respondents believed “Washingmachine.com”—“which does not correspond to any company”—to be a brand name.¹⁷ But consumers who have heard of companies like Booking.com or the erstwhile Pets.com might have guessed that Washingmachine.com, as presented, is a trademark. Might there have been a different response if the question had been, “Do you think that www.washingmachine.com is a trademark?,” even from consumers who understand that typing [word].com often redirects to www.[word].com?¹⁸ Conversely, would a similar percentage of respondents have identified any made-up word that sounded “trademarky” as a trademark, simply because they were guessing? This survey might tell us less about the possibility of [generic].com terms acquiring trademark status and more about the pitfalls of survey design.¹⁹

What remains to be seen, as both opinions suggest, are the tradeoffs involved and the extent to which the USPTO will engage in the necessary inquiry at the outset. The concerns raised by Justice Breyer (and some scholars and practitioners) about whether overreaching by trademark owners can be sufficiently thwarted by more robust affirmative defenses, particularly given that many chilling effects occur as the result of cease-and-desist letters alone, should not be dismissed. Likewise, there are questions about whether the USPTO should engage in more searching review at the outset to confirm that a [generic].com mark would be recognized as the name of a good or service rather than merely an address. This would include requiring proof that the mark is being used in a trademark sense and requiring disclaimers when necessary. (An application that seeks to register www.donuts.com as a trademark for a business called Main Street Bakery should typically be rejected, and a business called Donuts.com seeking a trademark registration for that term should be required to disclaim each part of that mark separately.) Reasonable observers may conclude that these efforts are not worth the candle.

But because the procedural posture of the case meant that the Court had no need to engage with the facts about consumer perception, it’s difficult to know how the case might have come out with a more nuanced inquiry. Certainly, the Court chose a route that rejected an easily administrable bright line rule of exclusion in favor of another general rule of trademark validity: that trademarks are, ultimately, proper names. That choice may end up being more difficult to manage, but it may have been the one more faithful to the concept of trademark validity.

1. See THE CONTAINER STORE, Registration No. 1,164,143 (for “Retail Store Services in the Area of Household Accessories, Storage Items, Storage Systems, and Space Organizers.”). The Principal Register also contains many live registrations for marks that take the form THE [generic term] COMPANY.
2. No. 19-46 (U.S. June 30, 2020).
3. See Booking.com B.V. v. Matal, 278 F.Supp.3d 891, 923–24 (E.D. Va. 2017).
4. See Booking.com B.V. v. United States Patent & Trademark Office, 915 F.3d 171, 175 (4th Cir. 2019).
5. Booking.com, slip op. at 1-2.
6. Id. at 7.
7. Id. at 7–8.
8. See Saul Kripke, Naming and Necessity 96 (1980).
9. Michel Foucault, What Is an Author?, in Textual Strategies: Perspectives in Post-Structural Criticism 141, 146 (Josué V. Harari ed., 1979).
10. Matal v. Tam, 137 S. Ct. 1744 (2017).
11. Iancu v. Brunetti, 139 S. Ct. 2294 (2019).
12. Tam, 137 S. Ct. at 1768 (Kennedy, J., concurring in part and concurring in the judgment). Justice Kennedy made this observation in connection with his view that trademark registration does not involve the use of private speakers “to assist the government in advancing a particular message.” Id.
13. Booking.com, slip op. at 9.
14. Id. at 6 (Breyer, J., dissenting).
15. See id. at 8 n.4 (majority opinion); id. at 6 (Breyer, J., dissenting).
16. Id. at 11 n.6 (majority opinion) (citing Brief for Trademark Scholars as Amici Curiae in Support of Neither Party, United States Patent & Trademark Office v. Booking.com B.V., No 19-46 (U.S. filed Jan. 13, 2020).
17. Id. at 9 (Breyer, J., dissenting).
18. Perhaps of note, www.washingmachine.com currently redirects to the site of the home furnishing company Wayfair.
19. Relatedly, I agree with Professor McCarthy that the doctrinal term “de facto secondary meaning”—used to describe the circumstances when a generic term is associated in some consumers’ minds with a particular producer—is unhelpful because it is used to describe not secondary meaning but rather market share or some other circumstance causing an association with the term. J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 12:47 (5th ed. 2020).