Moke America LLC v. Moke International Limited, No. 23-1634 (4th Cir. 2025)

Annotate this Case

Moke America LLC and Moke International Limited, along with Moke USA, LLC, are competing for the U.S. trademark rights to the "MOKE" mark, used for their low-speed, open-air vehicles. The U.S. District Court for the Eastern District of Virginia found that "MOKE" is a generic term for these vehicles, meaning it cannot be a trademark owned by either party. This finding was based on the history of the Moke vehicles, which were originally produced by the British Motor Corporation (BMC) and later by other manufacturers, and the term "Moke" becoming synonymous with a style of vehicle.

The district court's decision followed a bench trial where Moke America failed to prove its priority of use. The court then considered whether the MOKE mark was distinctive or generic. Both parties argued that the mark was inherently distinctive, but the court found it to be generic based on the evidence presented, including the parties' marketing efforts and the testimony of a Moke America witness.

The United States Court of Appeals for the Fourth Circuit reviewed the case and concluded that the district court correctly placed the burden on the parties to prove that "MOKE" is not a generic term. However, the Fourth Circuit found that the evidence was insufficient to either affirm or outright reverse the district court's finding of genericness. The court noted that more evidence is needed to determine whether "MOKE" is a generic term or an inherently distinctive mark that was abandoned by its original owner, BMC.

The Fourth Circuit vacated the district court's judgment and remanded the case for further proceedings to gather additional evidence on the distinctiveness or genericness of the "MOKE" mark. The parties will continue to bear the burden of proving that the mark is not generic. The court suggested that appointing a disinterested expert witness might be helpful in resolving the issue.

Download PDF

PUBLISHED UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT No. 23-1634 MOKE AMERICA LLC, a New York limited liability company, Plaintiff – Appellee, v. MOKE INTERNATIONAL LIMITED, Counter Claimant – Appellant, and AMERICAN CUSTOM GOLF CARS, INC., a California corporation; MOKE USA, LLC, a Delaware limited liability company, Defendants, and BLACK AND WHITE CORPORATIONS I-X; ABC PARTNERSHIPS I-X; JOHN AND JANE DOES I-X; DANIELS INDUSTRIES, LLC, an Ohio limited liability company; ANTOINE VERGLAS, INC., a New York corporation, Third Party Defendants. No. 23-1675 MOKE AMERICA LLC, a New York limited liability company, Plaintiff – Appellant, v. MOKE INTERNATIONAL LIMITED, Counter Claimant – Appellee, and MOKE USA, LLC, a Delaware limited liability company, Defendant – Appellee, and AMERICAN CUSTOM GOLF CARS, INC., a California corporation, Defendant, and BLACK AND WHITE CORPORATIONS I-X; ABC PARTNERSHIPS I-X; JOHN & JANE DOES I-X; DANIELS INDUSTRIES, LLC, an Ohio limited liability company; ANTOINE VERGLAS, INC., a New York corporation, Third Party Defendants. No. 23-1720 MOKE AMERICA LLC, a New York limited liability company, Plaintiff – Appellee, v. MOKE USA, LLC, a Delaware limited liability company, Defendant – Appellant, and 2 AMERICAN CUSTOM GOLF CARS, INC., a California corporation, Defendant, MOKE INTERNATIONAL LIMITED, Counter Claimant, and BLACK AND WHITE CORPORATIONS I-X; ABC PARTNERSHIPS I-X; JOHN & JANE DOES I-X; DANIELS INDUSTRIES, LLC, an Ohio limited liability company; ANTOINE VERGLAS, INC., a New York corporation, Third Party Defendants. Appeals from the United States District Court for the Eastern District of Virginia, at Richmond. David J. Novak, District Judge. (3:20-cv-00400-DJN-EWH) Argued: May 8, 2024 Decided: January 15, 2025 Before KING and RICHARDSON, Circuit Judges, and Gina M. GROH, United States District Judge for the Northern District of West Virginia, sitting by designation. Vacated and remanded by published opinion. Judge King wrote the majority opinion, in which Judge Groh joined. Judge Richardson wrote a dissenting opinion. ARGUED: Craig S. Mende, FROSS ZELNICK LEHRMAN & ZISSU, P.C., New York, New York, for Appellant/Cross-Appellee. Ryan C. Williams, AKERMAN LLP, Chicago, Illinois, for Appellee/Cross-Appellant. ON BRIEF: Andrew Nietes, FROSS ZELNICK LEHRMAN & ZISSU, P.C., New York, New York; Maya M. Eckstein, Trevor S. Cox, HUNTON ANDREWS KURTH LLP, Richmond, Virginia, for Appellant/Cross-Appellee. David S. Brafman, J. Andre Cortes, Mark D. Passler, AKERMAN LLP, West Palm Beach, Florida, for Appellee/Cross-Appellant. 3 KING, Circuit Judge: The parties to these cross-appeals from the Eastern District of Virginia — on one side, Moke America LLC, and on the other, related entities Moke International Limited and Moke USA, LLC — are competing for the United States trademark rights to the “MOKE” mark adorning the similar low-speed, open-air vehicles that they each sell. 1 Below, for example, is an image of one of their vehicles provided by Moke International and Moke USA to the district court. Although the parties are adversaries, their cross-appeals equally contest the district court’s primary finding: that MOKE is a generic term for the parties’ vehicles, meaning that it cannot be a trademark and cannot be owned by either party or anyone else. The court rendered the genericness finding in its Memorandum Opinion of May 3, 2023, setting forth findings of fact and conclusions of law following a truncated bench trial and multiple Like the district court, we sometimes allude to there being two parties. When we do so, we are referring to one party being Moke America and the other being Moke International and Moke USA. There were, however, additional parties to the district court proceedings, including another party aligned with Moke International and Moke USA by the name of American Custom Golf Cars, Inc. 1 4 rounds of post-trial briefing. See Moke Am. LLC v. Am. Custom Golf Cars, Inc., 671 F. Supp. 3d 670 (E.D. Va. 2023) (the “Opinion”). Upon consideration of the genericness finding, we are unable to either affirm or outright reverse it. Simply put, because the issue of genericness emerged only after the trial’s conclusion — and because neither party took the position that MOKE is a generic term or seriously endeavored to prove otherwise — there is a dearth of relevant evidence in the record. Accordingly, we vacate and remand for further proceedings. I. A. We begin, as the district court did in its Opinion of May 2023, by outlining the pertinent principles of trademark law. See Moke, 671 F. Supp. 3d at 677-79. It is helpful to do so because the “procedural background of this case is best digested with an understanding of the legal context in which the case arises.” Id. at 677. 1. Of utmost significance here, to qualify as a valid and protectable trademark, a designation “must be used to identify the source of the goods to potential customers.” See George & Co. v. Imagination Ent. Ltd., 575 F.3d 383, 400 (4th Cir. 2009) (citing MicroStrategy Inc. v. Motorola, Inc., 245 F.3d 335, 341 (4th Cir. 2001)). That is, the “designation must perform the job of identification: to identify one source and distinguish it from other sources.” See 1 J. Thomas McCarthy, McCarthy on Trademarks and Unfair 5 Competition § 3:4 (5th ed. May 2024 Update); see also MicroStrategy, 245 F.3d at 341 (same). Unless a designation is “distinctive” within the meaning of trademark law, it does not perform the job of identification and thus cannot be “a legally protectable ‘mark.’” See 1 McCarthy, supra, § 4:13. In simple terms, “No distinctiveness — no validity — no mark.” Id. As we have recognized, there are three types of marks that “are considered inherently distinctive and therefore valid without the holder having to make any other showing.” See OBX-Stock, Inc. v. Bicast, Inc., 558 F.3d 334, 340 (4th Cir. 2009). Those three types are “arbitrary,” “fanciful,” and “suggestive.” Id. In that order, “‘arbitrary’ marks are based on existing words used in ways unconnected with their common meaning, such as APPLE computer or SHELL gasoline.” Id. “‘Fanciful’ marks are made-up words that are invented to describe the product or source, such as KODAK or EXXON.” Id. And “suggestive” marks are those that “connote, without describing, some quality, ingredient, or characteristic of the product, such as L’EGGS pantyhose and GLASS DOCTOR window repair.” Id. (internal quotation marks omitted). Another type of mark is “descriptive.” See OBX-Stock, 558 F.3d at 340. “‘Descriptive’ marks merely describe a function, use, characteristic, size, or intended purpose of the product, such as YELLOW PAGES telephone directories and 5 MINUTE glue.” Id. (internal quotation marks omitted). Although a descriptive mark is not inherently distinctive, it is eligible for protection if it has “acquired a ‘secondary meaning.’” See Sara Lee Corp. v. Kayser-Roth Corp., 81 F.3d 455, 464 (4th Cir. 1996). 6 A descriptive mark has secondary meaning “if in the minds of the public, the primary significance of [the mark] is to identify the source of the product rather than the product itself.” Id. (internal quotation marks omitted). Finally, there are “generic” terms — often the common name of the product itself — which always lack distinctiveness and therefore “can never be trademarks.” See Sara Lee, 81 F.3d at 464; see also Ashley Furniture Indus., Inc. v. SanGiacomo N.A. Ltd., 187 F.3d 363, 369 (4th Cir. 1999) (“A generic mark refers to the genus or class of which a particular product is a member and can never be protected.”). “Examples of brand names held to be ‘generic’ terms are CONVENIENT STORE retail stores, DRY ICE solid carbon dioxide, and LIGHT BEER ale-type beverages.” See Sara Lee, 81 F.3d at 464 (cleaned up). Importantly, by withholding validity and protectability from generic terms, the law of trademarks “protects for public use those commonly used words and phrases that the public has adopted, denying to any one competitor a right to corner those words and phrases by expropriating them from the public ‘linguistic commons.’” See Am. Online, Inc. v. AT&T Corp., 243 F.3d 812, 821 (4th Cir. 2001). It is also important that, through a process sometimes called “genericide,” marks that begin as inherently distinctive may become generic as a result of the public’s pervasive use of them. See Am. Online, 243 F.3d at 820-21. Specifically, “[w]hen a trademark ceases to identify in the public’s mind the particular source of a product or service but rather identifies a class of product or service, regardless of source, that mark has become generic and is lost as an enforceable trademark.” See Glover v. Ampak, Inc., 74 F.3d 57, 59 (4th Cir. 1996). Prominent examples of marks that have undergone genericide include 7 THERMOS, ASPIRIN, CELLOPHANE, and ESCALATOR. See Am. Online, 243 F.3d at 821. 2. As for trademark ownership, such ownership is acquired at common law “by actual use of the mark in a given market.” See Emergency One, Inc. v. Am. Fire Eagle Engine Co., 332 F.3d 264, 267 (4th Cir. 2003). Under the basic rules of trademark priority, ownership is premised on priority of use, such that “the party claiming ownership must have been the first to actually use the mark in the sale of goods or services.” Id. (internal quotation marks omitted). “The party claiming ownership must also use the mark as a trademark,” i.e., “the mark must be used to identify the source of the goods to potential customers.” See George & Co., 575 F.3d at 400. “Thus, so long as a person is the first to use a particular mark to identify his goods in a given market, and so long as that owner continues to make use of the mark, he is entitled to prevent others from using the mark to describe their own goods in that market.” Id. (internal quotation marks omitted). Notably, “a party can become a senior user of a mark by acquiring from a previous owner an assignment of rights.” See RXD Media, LLC v. IP Application Dev. LLC, 986 F.3d 361, 370 (4th Cir. 2021). Moreover, “[t]he priority to use a mark . . . can be lost through abandonment.” See Emergency One, 332 F.3d at 268. “Once abandoned, a mark may be seized immediately and the person doing so may establish priority of use and ownership under the basic rules of trademark priority.” quotation marks omitted). 8 Id. (alteration and internal 3. Pursuant to the Lanham Act, the United States Patent and Trademark Office (the “PTO”) administers a federal registration system for trademarks. See Iancu v. Brunetti, 588 U.S. 388, 390-91 (2019) (citing 15 U.S.C. §§ 1051-1052). Although such registration “is not mandatory,” it “gives trademark owners valuable benefits.” Id. at 391. For example, registration constitutes “prima facie evidence of the validity of the registered mark and of the registration of the mark, [and] of the registrant’s ownership of the mark.” See 15 U.S.C. § 1115(a). But see Emergency One, 332 F.3d at 267 (cautioning that “[t]o acquire ownership of a trademark it is not enough to have invented the mark first or even to have registered it first” (internal quotation marks omitted)). Registration decisions are made by PTO examining attorneys, subject to review by the agency’s Trademark Trial and Appeal Board (the “TTAB”). See Iancu, 588 U.S. at 392. In a registration proceeding, if there is a question as to whether the proposed mark is a generic term, “the PTO always bears the burden of proving genericness.” See In re Cordua Rests., Inc., 823 F.3d 594, 600 (Fed. Cir. 2016). “When the PTO believes that an applicant’s mark is entitled to registration, it publishes the mark” and thereby “gives other interested parties the opportunity to oppose registration.” See Xactware Sols., Inc. v. Buildxact Software Ltd., 95 F.4th 810, 813 (4th Cir. 2024). If another party opposes registration, the “[o]pposition is referred to the [TTAB] ‘to determine and decide the respective rights of registration.’” See Xactware Sols., 95 F.4th at 813 (quoting 15 U.S.C. § 1067(a)); see also Rosenruist-Gestao E Servicos LDA v. Virgin Enters. Ltd., 511 F.3d 437, 443 n.4 (4th Cir. 2007) (explaining that an opposition 9 “proceeding before the TTAB is an adversarial action between parties regarding the registrability of a proposed trademark”). In an opposition proceeding, “[t]he party opposing registration bears the burden of proof.” See B&B Hardware, Inc. v. Hargis Indus., Inc., 575 U.S. 138, 144 (2015). Instead of being an opposition proceeding, an adversarial action before the TTAB regarding a mark’s registrability may be a cancellation proceeding, wherein a party seeks to cancel its adversary’s existing registration of a mark. See 15 U.S.C. § 1067(a); Rosenruist-Gestao, 511 F.3d at 443 n.4. In a cancellation proceeding, the party seeking to cancel a mark’s registration bears the burden of proof and must overcome the presumptions of the mark’s validity, the registration’s validity, and the registrant’s ownership of the mark. See Cold War Museum, Inc. v. Cold War Air Museum, Inc., 586 F.3d 1352, 1356 (Fed. Cir. 2009). 4. Under 15 U.S.C. § 1071(b), “[a] party who is dissatisfied with [a TTAB] decision may ‘have a remedy by a civil action’ by filing a suit in district court.” See Xactware Sols., 95 F.4th at 818 (quoting 15 U.S.C. § 1071(b)(1)); see also Swatch AG v. Beehive Wholesale, LLC, 739 F.3d 150, 155 (4th Cir. 2014) (explaining that a § 1071(b) action is an alternative to a § 1071(a) appeal of the TTAB decision to the U.S. Court of Appeals for the Federal Circuit). Rather than being an appeal, a § 1071(b) action “is a wholly separate suit, where new evidence is allowed and (in cases where new evidence is introduced) a de novo standard governs.” See Xactware Sols., 95 F.4th at 818-19. In a § 1071(b) action, “[t]he district court has authority independent of the PTO to grant or cancel registrations 10 and to decide any related matters such as infringement . . . claims.” See Swatch, 739 F.3d at 155. An infringement claim may concern either a registered trademark, see 15 U.S.C. § 1114(1), or an unregistered trademark, see id. § 1125(a). And infringement claims — along with claims involving registrability and ownership issues — may be pursued in district court actions under Lanham Act provisions other than § 1071(b). See, e.g., Glover, 74 F.3d at 58-59 (in non-§ 1071(b) district court action following registration of trademarks, plaintiff registrant sued defendant for trademark infringement and defendant counterclaimed for cancellation of plaintiff’s registrations). Insofar as a party to a § 1071(b) or other district court action is opposing a new registration or seeking to cancel an existing registration, that party generally bears the burden of proof. See Interprofession du Gruyere v. U.S. Dairy Exp. Council, 61 F.4th 407, 416 (4th Cir. 2023); Glover, 74 F.3d at 59. In the district court, a party opposing a new registration or seeking to cancel an existing registration on genericness grounds bears the burden of proving genericness by a preponderance of the evidence. Id. Where a district court action concerns the registrability or ownership of a registered trademark, the registrant enjoys the rebuttable presumptions of the mark’s validity, the registration’s validity, and the registrant’s ownership of the mark. See Emergency One, 332 F.3d at 268-69; Glover, 74 F.3d at 59. But where a district court action concerns ownership of an unregistered trademark, any party claiming ownership bears the burden of proving that it owns a valid and protectable mark. See Emergency One, 332 F.3d at 269. A claim that an unregistered trademark has been infringed requires the plaintiff to show 11 that it owns a valid and protectable mark, plus that the alleged infringer’s use of the mark creates a likelihood of consumer confusion. See Variety Stores, Inc. v. Wal-Mart Stores, Inc., 888 F.3d 651, 660 (4th Cir. 2018). Those showings must be made by a preponderance of the evidence. See Ga. Pac. Consumer Prods., LP v. von Drehle Corp., 618 F.3d 441, 451 (4th Cir. 2010). Of course, a showing of ownership of a valid and protectable mark requires proof of both priority of use and the mark’s distinctiveness. See TBL Licensing, LLC v. Vidal, 98 F.4th 500, 505 (4th Cir. 2024) (summarizing that “[t]he first to use the distinctive mark acquires rights to that mark, including the right to prevent others from using it”). With respect to non-genericness, this Court has recognized that a party claiming trademark infringement bears the burden of proving non-genericness when the alleged infringer claims genericness as a defense. See Ale House Mgmt., Inc. v. Raleigh Ale House, Inc., 205 F.3d 137, 140 (4th Cir. 2000). But we have not specified who, if anyone, bears the burden of proving non-genericness when two parties vie for ownership of an unregistered trademark and thus neither takes the position that the proposed mark is actually a generic term. B. Having outlined the foregoing legal principles, we turn to the procedural background of this case, up to the district court’s Opinion of May 2023. We limit this account to the details germane to our decision to vacate and remand. A fuller history is provided in the Opinion. See especially Moke, 671 F. Supp. 3d at 679-86, 699-700. Even 12 that account, however, is somewhat circumscribed, as the procedural background of this case is sizable and complex. 1. To start, in August 2015, another entity related to Moke International and Moke USA asserted ownership of the MOKE mark and applied for its registration with the PTO. 2 The PTO published the MOKE mark in late 2016, prompting Moke America to file a notice of opposition to the proposed registration. Therein, Moke America claimed that it is the owner of the MOKE mark based on priority of use. Several years of proceedings before the TTAB ensued, culminating in the TTAB’s dismissal of Moke America’s opposition by a decision of April 2020. Notably, no question of genericness was raised in the PTO proceedings by the PTO examining attorney. Moreover, the TTAB decision left pending the registration application benefiting Moke International and Moke USA, such that the MOKE mark has remained unregistered. Moke America turned to the district court for relief from the TTAB decision, initiating this civil action in June 2020 pursuant to 15 U.S.C. § 1071(b). Relevant here, in addition to requesting reversal of the TTAB decision, Moke America has sought a declaration of its trademark ownership and asserted a trademark infringement claim against Moke International and Moke USA. By their counterclaims, Moke International and Moke The entity that applied for registration of the MOKE mark in August 2015 is American Custom Golf Cars, see supra note 1, which assigned its interests in the MOKE mark to Moke USA in June 2019. In turn, Moke USA assigned the same interests to Moke International in April 2021. 2 13 USA have advocated affirmance of the TTAB decision, sought a declaration of their trademark ownership, and alleged that Moke America has engaged in trademark infringement. In other words, beyond their dispute over whether the TTAB decision should be affirmed or reversed, the parties have advanced in the district court competing trademark ownership and trademark infringement claims, obliging each of them to prove its ownership of a valid and protectable mark. As such, each party must prove not only priority of use, but also the MOKE mark’s distinctiveness. Additionally, the trademark infringement claims necessitate a showing by each party that the other party’s use of the MOKE mark creates a likelihood of consumer confusion. As authorized in a § 1071(b) action, the district court accepted new evidence from the parties and conducted a de novo review of their dispute. Prior to the court’s bench trial, the parties agreed that the sole issue as to their competing trademark ownership and trademark infringement claims is priority of use, i.e., which of them first used MOKE as a trademark in the sale of goods. The parties ignored the question of the MOKE mark’s distinctiveness, and they agreed that the use of MOKE by whichever of them is not the mark’s owner creates a likelihood of consumer confusion. The bench trial proceeded accordingly and began in late January 2023. On the trial’s first day, at the close of Moke America’s case-in-chief, Moke International and Moke USA moved for judgment on partial findings under Federal Rule of Civil Procedure 52(c) based on Moke America’s failure to demonstrate its priority of use by a preponderance of the evidence. See Fed. R. Civ. P. 52(c) (providing, inter alia, that “[i]f a party has been fully heard on an issue during a nonjury trial and the court finds against the party on that issue, 14 the court may enter judgment against the party on a claim or defense that, under the controlling law, can be maintained or defeated only with a favorable finding on that issue”). Having determined that Moke America indeed failed to prove its priority of use, and relying on Moke America’s “pretrial representations that a finding against it regarding priority of use would resolve all claims and counterclaims,” the court resolved to rule against Moke America on all three of its claims and to rule in favor of Moke International and Moke USA on two of their three counterclaims — all but their counterclaim for trademark infringement. See Moke, 671 F. Supp. 3d at 683-84. As further explained below, the district court “reserved judgment” as to the trademark infringement counterclaim because of a “last-minute argument” made by Moke America in opposition to its adversaries’ Rule 52(c) motion. See Moke, 671 F. Supp. 3d at 684. That argument called into question the MOKE mark’s distinctiveness. Later that same day, the district court entered a written order granting Moke International and Moke USA’s Rule 52(c) motion based on Moke America’s failure to prove its priority of use; directing post-trial briefing on the proper resolution of the trademark infringement counterclaim; and announcing the court’s rulings on the five other claims and counterclaims. In the circumstances, however, the court delayed issuing its findings of fact and conclusions of law as to those five claims and counterclaims. See Fed. R. Civ. P. 52(c) (specifying that “[a] judgment on partial findings must be supported by findings of fact and conclusions of law as required by Rule 52(a) [generally requiring findings of fact and conclusions of law in bench trials]”). 15 2. Again, prior to the January 2023 bench trial, the parties had ignored the question of the MOKE mark’s distinctiveness. That is, they had never “mentioned the category of mark — fanciful, arbitrary, suggestive, descriptive or generic — into which the MOKE mark falls.” See Moke, 671 F. Supp. 3d at 680. But in its Rule 52(c) argument at trial, Moke America asserted that the district court’s adverse priority-of-use determination was premised on an implicit finding that — rather than being fanciful, arbitrary, or suggestive and thus inherently distinctive and presumptively valid and protectable — the MOKE mark is descriptive and therefore eligible for trademark protection only with proof of secondary meaning. Under Moke America’s theory, the court could not rule in favor of Moke International and Moke USA on their trademark infringement counterclaim unless and until they demonstrated secondary meaning, i.e., that in the minds of the public, the primary significance of the MOKE mark is to identify Moke International and Moke USA as the source of the MOKE-adorned vehicles. 3 In brief, to prove priority of use at trial, Moke America relied on evidence that it had acquired ownership in the MOKE mark from a nonparty entity named Mini Mania, Inc., which had allegedly used MOKE as a trademark in the sale of goods — vehicle replacement parts — since at least the late 1980s. As part of its adverse priority-of-use determination, however, the district court found that Mini Mania’s use of the MOKE mark did not qualify as trademark use. Specifically, the court “found as a matter of fact that Mini Mania had used the MOKE mark to describe the nature of its products, not to identify their provenance.” See Moke, 671 F. Supp. 3d at 683. From there, Moke America asserted in its Rule 52(c) argument that the court “had found not only that Mini Mania used the MOKE mark descriptively, but further, that the MOKE mark was itself a descriptive trademark.” Id. at 684. And thus, according to Moke America, its adversaries were obliged to demonstrate secondary meaning and “could not prevail on their trademark infringement (Continued) 3 16 Thereafter, the post-trial brief of Moke America reiterated the position it had advanced at trial, while Moke International and Moke USA’s post-trial briefs “largely sidestepped the issue of the MOKE mark’s distinctiveness.” See Moke, 671 F. Supp. 3d at 684. Upon consideration of those briefs, the district court raised the question of genericness and whether the MOKE mark wholly lacks distinctiveness and therefore is invalid and unprotectable, advising the parties during a status conference in late February 2023 “that there existed a ‘pretty compelling argument’ that the MOKE mark is generic.” Id. at 685 (citation omitted). The district court nonetheless went on to agree with Moke America that implicit in the court’s priority-of-use determination under Rule 52(c) “was its finding that the MOKE mark is descriptive.” See Moke, 671 F. Supp. 3d at 685. The court thus ordered that the bench trial would resume so that Moke International and Moke USA could present evidence as to secondary meaning. The court further ordered, however, that there would first be a second round of post-trial briefing, in which Moke International and Moke USA “were to make their position as to the MOKE mark’s typology — whether descriptive or otherwise — explicit.” Id. In their second-round post-trial brief, Moke International and Moke USA disputed the proposition that the district court’s priority-of-use determination was premised on an implicit finding that the MOKE mark is descriptive. Moke International and Moke USA counterclaim without first demonstrating that consumers associated the MOKE mark with [the adversaries’] vehicles.” Id. 17 therefore argued that they were not required to show secondary meaning in order to prevail on their trademark infringement counterclaim, and they informed the court “that they intended to present no further evidence as to secondary meaning or the MOKE mark’s distinctiveness more broadly.” See Moke, 671 F. Supp. 3d at 685. Moke International and Moke USA also insisted that the court “needed no further evidence concerning distinctiveness, because the MOKE mark is arbitrary, i.e., inherently distinctive, and thus presumptively valid and protectable.” Id. 4 Because of Moke International and Moke USA’s disinclination to present further evidence, the district court decided in mid-March 2023 not to resume the bench trial after all. Around the same time, the court ordered third and fourth rounds of post-trial briefing. For the third round, the court directed the parties to address the question of genericness and the import of this Court’s decision of early March 2023 in Interprofession du Gruyere v. U.S. Dairy Export Council. See Gruyere, 61 F.4th at 411-12 (concluding that the term GRUYERE is generic and thus cannot be registered with the PTO as a geographic certification mark under the Lanham Act, see 15 U.S.C. § 1054, for cheese produced in the Gruyère region of Switzerland and France). The fourth and final round of briefing related to Moke America’s filing of a cross-motion for judgment on partial findings pursuant to Rule 52(c). The crux of Moke International and Moke USA’s theory was “that an inherently distinctive mark like MOKE may nonetheless be used descriptively, which is exactly what Mini Mania did when employing the MOKE mark.” See Moke, 671 F. Supp. 2d at 685; see also supra note 3. 4 18 By their third- and fourth-round briefs, the parties expressly agreed that the MOKE mark is not generic and that, as illustrated by the Gruyere decision, there was insufficient evidence in the record to support a genericness finding. See Gruyere, 61 F.4th at 425-26 (explaining that evidence of genericness “may come from purchaser testimony, consumer surveys, listings and dictionaries, trade journals, newspapers, and other publications” (quoting Glover, 74 F.3d at 59)). Neither party advocated further development of the evidence as to genericness specifically or distinctiveness more generally. Rather, Moke America contended that it is apparent that the MOKE mark is fanciful and thus inherently distinctive and presumptively valid and protectable. Moke America also argued that — if the district court disagreed and adhered to its previous implicit finding that the MOKE mark is descriptive — the court should rule against Moke International and Moke USA under Rule 52(c) on their unresolved trademark infringement counterclaim as a result of their failure to demonstrate secondary meaning. Additionally, Moke America maintained that the lack of proof of secondary meaning should cause the court to reverse its prior Rule 52(c) ruling in favor of Moke International and Moke USA on their counterclaim for a declaration of trademark ownership. For their part, Moke International and Moke USA agreed with Moke America that the MOKE mark is inherently distinctive and presumptively valid and protectable. But rather than characterizing the MOKE mark as fanciful (as Moke America did), Moke International and Moke USA described the mark as arbitrary (as they had previously done 19 in their second-round brief). 5 Otherwise, Moke International and Moke USA continued to dispute that there had been an implicit finding that the MOKE mark is descriptive and that they therefore were required to demonstrate secondary meaning. During a hearing in late April 2023, the district court conducted oral argument on Moke America’s Rule 52(c) cross-motion. At that point, it was clear to the court that “the MOKE mark’s distinctiveness, or lack thereof, now stands as the central and dispositive issue in the case.” See Moke, 671 F. Supp. 3d at 686. As the court saw it, the case’s posture was as follows: if the MOKE mark is fanciful, arbitrary, or suggestive and thus inherently distinctive, then the “analysis of all six claims and counterclaims must restart”; if the MOKE mark is descriptive, then the court “must assess whether [Moke International and Moke USA] have established secondary meaning” and thereby proved their trademark ownership and trademark infringement counterclaims; but if MOKE is a generic term, then “neither party can claim common law rights to the mark.” Id. The court recognized that each possibility would necessitate at least some revisions to its rulings on the five previously resolved claims and counterclaims. The court also observed that the parties’ “mutual failure to address the mark’s validity before trial explains a great deal of the posttrial procedural morass.” Id. at 686 n.11. In describing the MOKE mark as arbitrary, Moke International and Moke USA contended that “moke” is an existing word — “moke” being British slang for a donkey — used in a way unconnected with its common meaning. Meanwhile, Moke America characterized the MOKE mark as fanciful on the premise that U.S. consumers are not familiar with “moke” as a word referring to a donkey and thus understand “moke” to be a term created out of whole cloth. 5 20 C. That brings us to the Opinion of May 2023, in which the district court found — contrary to the unanimous position of the parties — that MOKE is a generic term for the parties’ vehicles and therefore cannot be a trademark. In rendering its genericness finding, the court deemed it unnecessary to further develop the evidentiary record and instead relied on the following: the existing trial evidence, which focused on Moke America’s alleged priority of use; the records proffered by the parties of the PTO proceedings; stipulations that had been made by the parties; and the parties’ publicly available websites, of which the court took judicial notice. 1. As indicated by the district court, the premise of its genericness finding is that MOKE was once an inherently distinctive mark but, through the process of “genericide,” it became a generic term before the parties first sold a vehicle bearing the MOKE mark. See Moke, 671 F. Supp. 3d at 679 (explaining that “[g]enericide occurs when the public appropriates a trademark and uses it as a generic name for particular types of goods or services irrespective of its source” (quoting Elliott v. Google, Inc., 860 F.3d 1151, 1156 (9th Cir. 2017))); see also Am. Online, 243 F.3d at 820-21 (similarly describing genericide). In indicating that the MOKE mark underwent genericide, the district court observed that “[a]t first blush,” the MOKE mark appears to be inherently distinctive. Id. 21 at 691. 6 The court found, however, that MOKE is now a generic term for the general style of vehicles sold by the parties in the United States. The genericness finding stemmed from these other findings of fact: “In the early 1960s, the now-defunct British Motor Corporation (‘BMC’) introduced a vehicle known as the ‘Mini Moke’ into the British car market”; “The BMC Mini Moke — colloquially, the ‘Moke’ — is an open-air, lightweight vehicle that sits low to the ground,” that was “initially designed for military purposes,” and that was marketed to the public by BMC “as a low-cost utility vehicle”; “From the Moke’s introduction in the 1960s until 1993, BMC and several other manufacturers produced Mokes in at least three countries — the United Kingdom, Australia, and Portugal”; “Mokes changed in name and style over time, with variations on the original known as the ‘Austin Mini Moke,’ the ‘Morris Mini Moke,’ the ‘Leyland Moke,’ and the ‘Moke Californian’”; “Irrespective of their style, name, manufacturer or country of origin, however, all variations were and are referred to colloquially as ‘Mokes’”; “BMC and its apparent progeny produced approximately 50,000 Mokes before production ceased in 1993”; “Notwithstanding their limited production, Mokes garnered a small but devoted following for use as a beach buggy in the United States, the Caribbean and Australia”; and The district court acknowledged that there is “a plausible argument” that MOKE falls into each of the three types of inherently distinctive marks: fanciful because “moke” lacks a defined meaning in American English; arbitrary because “moke” is British slang for a donkey; and suggestive because of the connotation that, like a donkey, “a MOKE vehicle carries people and things at a leisurely pace, perhaps with a little attitude.” See Moke, 671 F. Supp. 3d at 691-92; see also supra note 5. 6 22 “A meaningful secondary market for Mokes developed in the United States at least as early as 1988.” Id. at 687-88. The district court also found that Moke International and Moke USA’s first sale in the United States of a vehicle bearing the MOKE mark was in August 2015, and that Moke America thereafter began U.S. sales of vehicles bearing the MOKE mark in 2016. According to the court, the parties’ “vehicles appear largely indistinguishable from the twentieth-century Mokes that BMC and its apparent corporate successors sold.” See Moke, 671 F. Supp. 3d at 690 (specifically describing Moke America’s vehicles); id. at 691 (similarly describing Moke International and Moke USA’s vehicles). Moreover, the parties’ respective websites provide “recapitulation[s] of the Moke’s twentieth-century history” and describe their vehicles as being reengineered and redesigned versions of the predecessor Mokes. Id. at 690-91. As the district court explained, it was led to its ultimate genericness finding by the foregoing findings of “the existence of nearly identical products bearing the same mark and referred to by the same name — products that the parties themselves explicitly reference when marketing their own vehicles.” See Moke, 671 F. Supp. 3d at 692-93. In the court’s view of the parties’ marketing efforts, the parties implicitly disclaim that the MOKE mark identifies their products and their products alone. Rather, to hear the parties tell it, the twentiethcentury Mokes, many of which are still in circulation in the United States, are so iconic that MOKE has become synonymous with a style of vehicle — a style mimicking the “Mokes” produced by BMC and its [apparent] corporate successors. For [the parties], then, their products are merely a species within the MOKE genus, a song within the MOKE genre. The 23 MOKE mark thus serves not as an indication of source, but as a talisman of a particular style. Id. at 694 (citation omitted). “This,” the court concluded, “is quintessential genericism.” Id. 7 2. Additionally, the district court concluded that neither of the dual purposes of trademark law — which the court described as “(1) protecting consumers against fakes and counterfeiters and (2) safeguarding markholders’ investments of time and money in building the goodwill associated with their mark” — would be served by “reserving the MOKE mark for one of the parties and excluding all others, including third parties, from employing the mark in connection with motor vehicles.” See Moke, 671 F. Supp. 3d at 695 (citing George & Co., 575 F.3d at 392-93). Starting with the second purpose, the court observed that the evidence before it of “consumer goodwill associated with the MOKE mark largely suggest[s] that consumers direct such goodwill towards the Moke vehicles Similarly, the district court found that Mini Mania — the entity that purportedly had conveyed its rights in the MOKE mark to Moke America — only ever used MOKE “as a generic term, synonymous with a style of vehicle produced by different companies.” See Moke, 671 F. Supp. 3d at 693. Consequently, the court adhered to its prior finding that Mini Mania did not use MOKE as a trademark — the finding previously tied to what the court had characterized as Mini Mania’s “descriptive” use of the MOKE mark. See supra note 3. The court cited evidence that, beginning by the late 1980s, Mini Mania sold replacement parts in the United States for the Moke vehicles produced by BMC and other foreign manufacturers, and that Mini Mania had used the MOKE mark in advertising “not to indicate the source or origin of Mini Mania’s products, but to reference the Mokes that its target customers already owned.” See Moke, 671 F. Supp. 3d at 693. 7 24 manufactured in the twentieth century by BMC and others, not the vehicles produced by either [Moke America or Moke International and Moke USA].” Id. The district court’s discussion of “goodwill” emphasized that neither party “put forth any evidence that consumers associate MOKE with their particular products,” leaving the record with “no consumer surveys, no customer testimonials, and no news stories or trade journals evidencing an association between the MOKE mark and [the parties’ respective] vehicles.” See Moke, 671 F. Supp. 3d at 695. “On the other side of the ledger,” however, was the evidence “that Moke ‘enthusiasts’ existed long before any party to this suit” and that the parties had a “shared understanding that by employing the MOKE mark in connection with their vehicles,” they could “capitaliz[e] on consumer goodwill that already exists, not goodwill that they themselves have built.” Id. As such, the court reasoned that neither party “can plausibly maintain that it is their investments of time and money that imbue MOKE with a ‘cult’ following and thus warrant the exclusionary benefits accompanying trademark status.” Id. Turning to the first purpose of trademark law, the district court concluded that that purpose — “protecting consumers against fakes and counterfeiters” — “would be equally disserved by reserving the MOKE mark for one party over the other.” See Moke, 671 F. Supp. 3d at 695. Relevant to that conclusion, the court highlighted that the record “contains no allegation or intimation that either party is a corporate relative of [BMC]” and thus neither party “can maintain that they manufacture the ‘One True MOKE.’” Id. The court elaborated: 25 And while the vehicles that the parties produce are, for all intents and purposes, indistinguishable, this is by design, as both explicitly note that their vehicles mimic the design of the “iconic” MOKE of yesteryear. To state that one party is producing a fake or counterfeit MOKE would be equally damning to the other, as both [parties] clearly market their Mokes as imitations of a pre-existing product sharing the same name. If anything, granting trademark protection to either [party] would itself give rise to customer confusion, as it would incorrectly signal to consumers that the trademark holder serves as the exclusive source of Mokes, despite that party’s acknowledgement that bona fide Mokes manufactured by other companies — British, Australian, Portuguese, or otherwise — remain in the marketplace. Id. at 695-96. “In sum,” the court determined “that bestowing trademark status on the MOKE mark would disserve the policy rationales undergirding trademark protections, irrespective of which party ultimately won the mark.” Id. at 696. With regard to policy considerations, the district court also thought it important to “consider the effect that granting trademark protection to either side would have on” thirdparty Moke manufacturers and the public, especially the other Moke manufacturers’ loss of “the ability to describe their products with the most accurate and precise term available” and the public’s loss of a word “from the ‘linguistic commons.’” See Moke, 671 F. Supp. 3d at 696 n.17 (quoting Am. Online, 243 F.3d at 821). That is, the court acknowledged and addressed the adverse consequences that would result from extending trademark protection to a generic term. “Here,” the court underscored, “the parties fail to show that their right to use the MOKE mark should prevail over that of third parties and at the expense of the public vocabulary.” Id. 26 3. Next, the district court confronted the parties’ contention that, as illustrated by this Court’s Gruyere decision, the evidentiary record is too thin to support any finding of genericness. As the district court recognized, that contention implicates not only our Gruyere decision, but also our earlier Glover decision, wherein we explained that evidence of genericness “may come from purchaser testimony, consumer surveys, listings and dictionaries, trade journals, newspapers, and other publications.” See Gruyere, 61 F.4th at 425-26 (quoting Glover, 74 F.3d at 59). Along with the adequacy of the evidence, the parties’ contention caused the district court to focus on the burden of proof. a. With respect to the burden of proof, the district court distinguished Gruyere and Glover on the ground that those cases involved parties who claimed genericness in situations that placed the burden on them to prove it. In Gruyere, they were parties who claimed genericness in opposing the application for registration of GRUYERE as a geographic certification mark. See Gruyere, 61 F.4th at 411. And in Glover, it was a party who claimed genericness in order to cancel the previously approved registrations of certain pocket- and hunting-knife-related trademarks. See Glover, 74 F.3d at 58. As the district court emphasized, this case — unlike Gruyere and Glover — has no party asserting that MOKE is a generic term and thus no party bearing the burden to prove it. Instead, each party is seeking to prove its ownership and the other party’s infringement of the unregistered MOKE mark. Hence, the court pronounced, “the burden rests with the parties . . . to prove that the mark is protectable” — meaning that the burden rests with the 27 parties to prove that MOKE is not a generic term. See Moke, 671 F. Supp. 3d at 697-98. For support of its burden-of-proof ruling, the court cited a Second Circuit decision explaining that the burden to prove the validity and protectability of an unregistered mark “necessarily implies that [the party claiming ownership] must bear the burden of proving that [the proposed] mark is not generic.” See Reese Publ’g Co. v. Hampton Int’l Commc’ns, Inc., 620 F.2d 7, 11 (2d Cir. 1980). b. Addressing the adequacy of the evidence, the district court observed that the parties invoked Gruyere and Glover “for the proposition that purchaser testimony and consumer surveys are de rigueur in cases denying trademark protection on genericness grounds.” See Moke, 671 F. Supp. 3d at 696; see also Gruyere, 61 F.4th at 426 n.8 (suggesting that “consumer surveys may be a preferred method of proving genericness” (internal quotation marks omitted)). “Without such evidence,” the parties insisted, there can be no finding of genericness in this case because there can be no finding “that the ‘primary significance’ of the MOKE mark in the minds of relevant consumers is its indication of type or style, rather than source.” See Moke, 671 F. Supp. 3d at 696. The district court recognized, however, that our Gruyere and Glover decisions do not require purchaser testimony and consumer surveys to prove genericness in every case. See, e.g., Gruyere, 61 F.4th at 426 n.8 (confirming that consumer surveys are not always needed). Indeed, in Gruyere, the parties claiming genericness did not present a single customer survey but prevailed on their genericness claim anyway based on other evidence that “strongly indicates that to the American purchaser, GRUYERE primarily signifies a 28 type of cheese (much like brie, swiss, parmesan or mozzarella) regardless of regional origin.” Id. at 423 (alteration and internal quotation marks omitted). And in Glover, the party claiming genericness was unsuccessful because of a “fail[ure] to present any evidence that [any of the] marks has become synonymous with ‘pocket knife’ or ‘hunting knife’ to purchasers in the relevant public.” See Glover, 74 F.3d at 60 (emphasis added). To the extent that the district court ascertained the evidentiary record to be too thin, the court’s determination was that there was a dearth of evidence demonstrating that the MOKE mark is distinctive and thus valid and protectable. See Moke, 671 F. Supp. 3d at 698 (concluding “that the parties’ characterizations of the record evidence concerning the MOKE mark’s distinctiveness prove wanting”). The court rejected the proposition that there was inadequate evidence of genericness, stating: Both [parties] repeatedly assert that the record stands devoid of any facts suggesting that the MOKE mark is generic, and specifically, devoid of any evidence that prospective purchasers understand MOKE as synonymous with a type or style of vehicle. While the Court agrees to a limited extent, in that the record contains no consumer survey evidence of genericness, the Court disagrees to the extent that the parties suggest or imply that [the trial] testimony [of a Moke America fact witness] sheds no light whatsoever on how the MOKE mark is perceived in the marketplace. Id. 8 The court specified that the Moke America witness’s “testimony could easily be interpreted as reflecting his understanding of MOKE as a generic term,” pointing to, inter alia, the witness’s repeated references “to vehicles manufactured by disparate, unrelated The Moke America fact witness invoked by the district court — the only witness called during the single day of the bench trial — is “Don Racine, former owner and CEO of [Moke America’s] purported predecessor-in-interest, Mini Mania.” See Moke, 671 F. Supp. 3d at 698; see also supra note 3. 8 29 companies as ‘Mokes.’” Id. (detailing that the witness used “Moke” to refer to vehicles produced by BMC and “an unnamed Chinese company,” as well as vehicles produced in Australia and Portugal by entities that may or may not have been “bona fide corporate successors to BMC”). The district court acknowledged that the Moke America witness “was not qualified as an expert witness on either the etymology or consumer perception of the word ‘Moke,’” but concluded that it was enough that the witness “no doubt fancies himself a prolific consumer of all things MOKE” and that “his testimony constitutes the evidence on which the parties chose to proceed.” See Moke, 671 F. Supp. 3d at 698. The court also reiterated that “[f]or the parties to suggest that the record contains no evidence of consumers’ perception of the mark simply ignores how [the witness’s] testimony conveys his understanding of the MOKE mark in the American marketplace.” Id. at 698-99. 9 At the close of its discussion of the burden of proof and the adequacy of the evidence, the district court deemed it irrelevant that no question of genericness was raised in the PTO proceedings by the PTO examining attorney. In so doing, the court cited its possession of “a factual record that was not before the [PTO examining attorney] when he reviewed the [MOKE] mark.” See Moke, 671 F. Supp. 3d at 699. The court explained that the PTO examining attorney’s “failure to raise a genericness objection thus tells the Court little in the current context, where . . . the Court ‘must evaluate the evidence in the present record to determine whether’ the mark proves valid and protectable.” Id. (quoting Cordua Rests., 823 F.3d at 600). Moreover, the court noted that “comparisons to registration proceedings before [PTO examining attorneys] prove inapt to the extent that the burden in such proceedings lies not with the applicant, but with the examiner.” Id. (citing Cordua Rests., 823 F.3d at 600, for the proposition that “in registration proceedings the PTO always bears the burden of proving genericness”). 9 30 4. Lastly, the district court addressed its “evolution” from the initial implicit finding that MOKE is a descriptive mark to the ultimate finding that MOKE is a generic term, again criticizing the parties for their “shared failure to address the mark’s distinctiveness in a straightforward fashion” and to take advantage of the opportunities the court afforded them to present relevant evidence. See Moke, 671 F. Supp. 3d at 699-700 (repeating key points of this case’s procedural history). The court then summarized its decision as follows: The salient takeaway from all of the above — the cornerstone on which the Court’s decision rests — proves simple: this case turns on the burden of proof. [Moke America] and [Moke International and Moke USA] alike shoulder a burden to prove, by a preponderance of the evidence, that they possess common law trademark rights in the MOKE mark. Such a showing entails proving (1) priority of use of (2) a valid, protectable trademark. Without proving both elements, neither party can prevail on their respective trademark infringement claims or their respective actions for a declaration of trademark ownership rights. The parties chose to present very little evidence on the second half of the ownership equation, all but assuming that the Court would find the MOKE mark inherently distinctive. While the Court . . . acknowledges that sorting trademarks into the appropriate distinctiveness bucket is no easy task, the Court cannot find, on the record before it, that either party has demonstrated that the relevant consuming public associates the MOKE mark with their products. The limited record before the Court, to the extent probative, in fact suggests the opposite — that consumers see the mark as synonymous with a style of vehicle produced by one or more non-party entities a century ago. Based on the record evidence, the Court thus concludes that the MOKE mark, as used in connection with the parties’ vehicles, is generic. The mark therefore falls beyond the ambit of trademark protection. Id. at 700-01 (citations omitted). 31 D. Upon issuance of its Opinion in early May 2023, the district court made conforming revisions to its rulings on the five previously resolved claims and counterclaims. As a result, the court entered judgment against Moke America on its trademark ownership and trademark infringement claims, and against Moke International and Moke USA on their trademark ownership and trademark infringement counterclaims. The court also entered judgment against Moke International and Moke USA on their counterclaim for affirmance of the TTAB decision. The court entered judgment for Moke America on its claim for reversal of the TTAB decision, but rather than ordering the TTAB to rule in favor of Moke America, the court directed the TTAB to deny Moke International and Moke USA’s pending application for registration of the MOKE mark on the ground that MOKE is a generic term and therefore cannot be a trademark. In late May 2023, Moke America filed a motion pursuant to Federal Rule of Civil Procedure 59(e) to alter or amend the judgment. See Mayfield v. Nat’l Ass’n for Stock Car Auto Racing, Inc., 674 F.3d 369, 378 (4th Cir. 2012) (recognizing that Rule 59(e) relief may be granted only in limited circumstances, including “to correct a clear error of law or prevent manifest injustice,” and that “[i]t is an extraordinary remedy that should be applied sparingly” (internal quotation marks omitted)). Moke America argued, inter alia, that the district court erred by placing the burden on the parties to prove non-genericness and by finding genericness in the absence of adequate evidence to do so. Additionally, Moke America highlighted the concept of trademark abandonment and contended that — rather than supporting the district court’s genericness finding — the 32 evidence before the court establishes that MOKE is an abandoned, inherently distinctive mark. See Emergency One, 332 F.3d at 268 (explaining that “[t]he priority to use a mark . . . can be lost through abandonment” and that “[o]nce abandoned, a mark may be seized immediately and the person doing so may establish priority of use and ownership under the basic rules of trademark priority” (alteration and internal quotation marks omitted)). Pursuant to Moke America’s abandonment theory, the MOKE mark both began as and remains an inherently distinctive mark; the MOKE mark was originally owned but later abandoned by BMC, the British company that introduced Moke vehicles in the early 1960s and then ceased production in 1993; and BMC’s abandonment of the MOKE mark left the mark available to a new owner based on priority of use. Moke America also asserted victory in the race to become the MOKE mark’s new owner. The district court denied Moke America’s Rule 59(e) motion in early June 2023 for failure to satisfy the stringent requirements for relief. Meanwhile, the parties timely noted these cross-appeals, invoking our jurisdiction under 28 U.S.C. § 1291. II. In contesting the district court’s finding in its Opinion of May 2023 that MOKE is a generic term for the parties’ vehicles and therefore cannot be a trademark, both parties now maintain on appeal that the court erred by placing the burden on the parties to prove non-genericness and by finding genericness in the absence of adequate evidence to do so. Both parties also promote the abandonment theory, i.e., that the existing evidence establishes that MOKE began as and remains an inherently distinctive mark, which was 33 abandoned by its original owner and thereby left up for grabs to a new owner based on priority of use. Where the parties diverge is on their competing assertions of priority of use of the previously abandoned MOKE mark. Each party would have us proclaim it the MOKE mark’s new owner, with directions to the district court to revise its final rulings on the parties’ various claims and counterclaims accordingly. We conclude that the district court properly placed the burden on the parties to prove that MOKE is not a generic term for their vehicles. As stated at the outset of this decision, however, we are vacating and remanding for further proceedings because there is not enough evidence to either affirm or outright reverse the court’s genericness finding. That is, we conclude that the evidence is inadequate to resolve not only whether MOKE is a generic term (as the district court found), but also whether MOKE is instead an abandoned, inherently distinctive mark (as the parties now argue). We further address those issues below. A. We begin with the burden of proof, which is a legal issue that we review de novo. See Swatch AG v. Beehive Wholesale, LLC, 739 F.3d 150, 154-55 (4th Cir. 2014). Again, the district court reasoned that — because the parties’ competing trademark ownership and trademark infringement claims require each of them to prove that it owns a valid and protectable mark — each party bears the burden of proving that MOKE is a distinctive mark. And thus, each party bears the burden of proving that MOKE is not a generic term for its vehicles. 34 The parties contend, however, that because neither of them has asserted that MOKE is a generic term, neither of them bears any burden to prove genericness — or, for that matter, non-genericness. Rather, as the parties would have it, the district court must accept what is essentially the parties’ stipulation that MOKE began as and remains an inherently distinctive mark. 1. We agree with the district court’s sound conclusion that the competing trademark ownership and trademark infringement claims require each party to prove non-genericness. It is well-established that a party claiming ownership of an unregistered trademark bears the burden of proving that it owns a valid and protectable mark. See Emergency One, Inc. v. Am. Fire Eagle Engine Co., 332 F.3d 264, 269 (4th Cir. 2003). And to show that the mark is valid and protectable, the party claiming ownership must show that the mark is distinctive. See 1 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 4:13 (5th ed. May 2024 Update) (explaining that unless a designation is “distinctive” within the meaning of trademark law, it does not identify the source of the goods and cannot be “a legally protectable ‘mark’”). Therefore, as the district court concluded, the party claiming ownership is obliged to prove that the proposed mark is not actually a generic term lacking any distinctiveness at all. In the words of the Second Circuit’s Reese Publishing decision invoked by the district court, the burden to prove the validity and protectability of an unregistered mark “necessarily implies that [the party claiming ownership] must bear the burden of proving that [the proposed] mark is not generic.” 35 See Reese Publ’g Co. v. Hampton Int’l Commc’ns, Inc., 620 F.2d 7, 11 (2d Cir. 1980); accord Nat’l Conf. of Bar Exam’rs v. Multistate Legal Studies, Inc., 692 F.2d 478, 488 (7th Cir. 1982) (“Because the alleged trademark in this case is unregistered, the plaintiff had the burden of proving that its mark was not an unprotectable generic mark.” (cleaned up) (citing Reese Publ’g, 620 F.2d at 11)). Oftentimes, when a party claiming ownership has demonstrated that a mark is either “arbitrary,” “fanciful,” or “suggestive” — and thus inherently distinctive — the party will have satisfied its burden of showing non-genericness. See OBX-Stock, Inc. v. Bicast, Inc., 558 F.3d 334, 340 (4th Cir. 2009) (relating that inherently distinctive marks are generally “valid without the holder having to make any other showing”). But it is not so simple where, as here, there is reason to question whether the mark began as inherently distinctive but is now generic as a result of “genericide.” See Am. Online, Inc. v. AT&T Corp., 243 F.3d 812, 820-21 (4th Cir. 2001) (explaining that an inherently distinctive mark may undergo genericide by way of the public’s pervasive use of the mark as a generic term). In such circumstances, it is necessary for the party to prove non-genericness by showing that genericide has not in fact occurred. 2. According to the parties, however, controlling and persuasive authorities connote that no party can be required to prove genericness or non-genericness unless some party has affirmatively raised a genericness claim. For example, Moke International and Moke USA point to decisions requiring a party claiming genericness to prove it, including our Gruyere and Glover decisions. See Interprofession du Gruyere v. U.S. Dairy Exp. Council, 36 61 F.4th 407, 416 (4th Cir. 2023) (party opposing new registration of mark on genericness grounds); Glover v. Ampak, Inc., 74 F.3d 57, 59 (4th Cir. 1996) (party claiming genericness in order to cancel existing registration). They and Moke America also highlight decisions, such as our Ale House decision, requiring a trademark infringement plaintiff to prove the non-genericness of an unregistered mark where the alleged infringer has claimed genericness as a defense. See Ale House Mgmt., Inc. v. Raleigh Ale House, Inc., 205 F.3d 137, 140 (4th Cir. 2000). Alternatively, the parties posit that — even if a party claiming ownership of an unregistered mark may sometimes be obliged to prove non-genericness in the absence of another party’s genericness claim — the purported owner is required to do so only if there is reason to believe that the mark has always been a generic term, and not if there is suspected genericide. In other words, the parties assert that genericide never need be disproven by a party claiming ownership of an unregistered mark. The parties’ alternative theory relies on the Second Circuit’s decision in Murphy Door Bed Co. v. Interior Sleep Systems, Inc., 874 F.2d 95, 100-01 (2d Cir. 1989), shifting the burden to the alleged infringer of an unregistered mark to prove its defense that the mark has undergone genericide. The Murphy Bed court distinguished the Second Circuit precedent of Reese Publishing, 620 F.2d at 11, in which the court had recognized that the party claiming ownership of an unregistered mark bears the burden of proving that the proposed mark is not generic. Specifically, the Murphy Bed court distinguished Reese Publishing on the ground that the mark at issue therein, VIDEO BUYER’S GUIDE, apparently began as and remained generic, whereas in Murphy Bed, the disputed MURPHY 37 BED mark began as an inherently distinctive mark and then allegedly became generic by way of genericide. See Murphy Bed, 874 F.2d at 100. The Murphy Bed court deemed “this distinction important” and held that where an alleged infringer claims that a once distinctive mark has undergone genericide, the trademark infringement plaintiff enjoys a “presumption of non-genericness” that “is justified by the commercial protection a developer of innovations deserves.” Id. at 101. To be sure, none of the foregoing decisions squarely supports either of the parties’ theories. And we do not find ourselves persuaded by any of the cited decisions to embrace the theory that no party can be required to prove genericness or non-genericness absent some party’s genericness claim. Nor are we swayed to adopt the alternative theory that genericide never need be disproven by a party claiming ownership of an unregistered mark. 10 In effect, the parties contend that competitors for ownership of an unregistered mark can jointly and self-servingly stipulate to the mark’s distinctiveness — whether it be by an explicit stipulation or by an implicit stipulation engendered by the parties’ mutual failure to assert any question of genericness. Crucial to their contention, the parties further suggest We leave for another day the question — answered by the Second Circuit in its Murphy Bed decision — of which party bears the burden to prove genericness or nongenericness when an alleged infringer of an unregistered mark claims a genericide defense. See Murphy Bed, 874 F.2d at 101 (requiring the alleged infringer to prove genericide and thereby overcome a “presumption of non-genericness” enjoyed by the trademark infringement plaintiff). That issue is not implicated herein and arguably was not resolved by our Ale House decision, which addressed the defense that the allegedly infringed ALE HOUSE mark had always been generic. See Ale House, 205 F.3d at 140 (obliging the trademark infringement plaintiff to prove non-genericness). 10 38 that the district court must accept the stipulation of distinctiveness no matter how obvious or at least plausible it is that the proposed mark has always been a generic term or has become so by genericide. We reject the parties’ contention as contrary to the general rule that, in “its role as the evidentiary gatekeeper, a district court need not accept stipulations between parties.” See Wharton v. Superintendent Graterford SCI, 95 F.4th 113, 125 (3d Cir. 2024); see also 83 C.J.S. Stipulations § 18 (May 2024 Update) (“The court has the discretion to determine a stipulation’s accuracy or its validity and reasonableness, and the court must reject stipulations that contain demonstrable falsehoods or inconsistencies.” (citations omitted)). Moreover, we reject the parties’ contention for being incongruous with the court’s role in protecting third-party interests and preventing generic terms from being expropriated “from the public ‘linguistic commons.’” See Am. Online, 243 F.3d at 821; see also Baz v. Patterson, 100 F.4th 854, 867 (7th Cir. 2024) (recognizing that stipulations between the parties should not be allowed “to bind third parties or to dictate the district court’s factual determinations”). Rather, as heretofore expressed, we agree with the district court that the parties’ competing trademark ownership and trademark infringement claims require each of them to prove non-genericness. That is, each party bears the burden of proving that the MOKE mark has not undergone genericide. B. We next address the adequacy of the evidence to resolve whether MOKE is a generic term or an abandoned, inherently distinctive mark. The genericness or distinctiveness of a 39 mark is a question of fact that we generally review for clear error. See Swatch, 739 F.3d at 155. 1. a. To recap, the district court deemed the existing evidentiary record to be insufficient to satisfy the parties’ burden of proving that MOKE is not a generic term for their vehicles. The court also determined there to be ample evidence that MOKE is such a term. Specifically, the court identified and credited evidence that MOKE was once an inherently distinctive mark but became a generic term by way of genericide well before the parties first sold a MOKE-adorned vehicle, in 2015 (Moke International and Moke USA) and 2016 (Moke America). In thus rendering its genericness finding, the district court made a series of underlying findings of fact. Those included that the original Moke vehicles were produced in the early 1960s by British manufacturer BMC; that Moke vehicles were thereafter produced by additional foreign manufacturers, including BMC’s “apparent” corporate successors in Australia and Portugal; that BMC and the manufacturers in Australia and Portugal ceased production of Moke vehicles in 1993; and that by then, vehicles of the same general style produced by BMC and others had come to be known colloquially as “Mokes” and had garnered a small but devoted following and a meaningful secondary market in the United States. Based on the publicly available websites of Moke America and of Moke International and Moke USA, the district court found that the parties’ vehicles are largely 40 indistinguishable from the Moke vehicles previously produced by BMC and others, and that the parties have highlighted those similarities and the history of the predecessor Moke vehicles in marketing their own. Furthermore, the court interpreted the parties’ marketing efforts as espousing the notion that MOKE has become synonymous with the style of their vehicles, and not with any particular source. The district court similarly construed the trial testimony of Moke America’s fact witness to reflect an understanding that MOKE is a generic term for the parties’ vehicles, in that the witness repeatedly used the term “Moke” to refer to vehicles produced not only by BMC and its apparent corporate successors in Australia and Portugal, but also by at least one other, unnamed manufacturer in China. Although the court acknowledged that the witness was called by Moke America for the sole purpose of establishing its priority of use and was not qualified as an expert witness on the etymology or consumer perception of the MOKE mark, the court was satisfied with the witness’s status as “a prolific consumer of all things MOKE.” Indeed, the district court deemed the existing record to be sufficient to render its genericness finding even without some types of evidence — particularly consumer surveys and conventional purchaser testimony — that is often used to establish genericness. The court observed both that such evidence is not required to demonstrate genericness in every case, and that the lack of such evidence in this case was the choice and the fault of the parties. The district court also emphasized policy considerations, specifying that reserving the MOKE mark to one party or the other would not serve the dual purposes of trademark 41 law. That is, the court concluded that because both parties are basically copying the Moke vehicles popularized by BMC and others — and because there is no evidence that consumers associate the MOKE mark with the parties’ particular vehicles or that either party is a corporate successor to BMC — reserving the MOKE mark to one of the parties would neither (1) protect consumers against fakes and counterfeiters nor (2) safeguard some investment of time or money by the prevailing party in building the goodwill associated with the mark. Additionally, the court addressed the adverse consequences that would result from extending trademark protection to a generic term, citing the inability of third-party Moke manufacturers to describe their products as “Mokes” and the loss of “Moke” from the public vocabulary. b. For their part, the parties maintain that the existing evidentiary record belies the district court’s genericness finding and instead establishes that MOKE is and always has been an inherently distinctive mark. Under the parties’ view of the evidence, beginning in the early 1960s, the MOKE mark identified BMC as the source of Moke vehicles and continued to do so even after the other vehicle manufacturers in Australia and Portugal began using the MOKE mark. The parties say that is because those other manufacturers were — or were at least believed to be — corporate successors of BMC. The parties assert that BMC and its apparent corporate successors thereafter abandoned the MOKE mark by ceasing production of Moke vehicles in 1993, leaving the MOKE mark available to a new owner based on priority of use. Further, the parties assert that the evidence fails to show that, either before or after the abandonment, the MOKE mark underwent genericide and 42 became a generic term for vehicles of the same general style as BMC’s Moke vehicles. That is, the parties insist there is scant evidence that all such vehicles have ever come to be known colloquially known as “Mokes,” as the district court found. In particular, Moke International and Moke USA criticize the district court’s interpretation of certain snippets of the trial testimony of Moke America’s fact witness “to suggest many unrelated users of the historical MOKE mark existed.” See Br. of Appellants 34-35. Moke International and Moke USA contend — based on different snippets of the Moke America witness’s testimony — that the testimony actually reflects that there was “only a single chain of ownership of MOKE vehicle makers before manufacturing stopped in the 1990s.” Id. at 35 (pointing to the Moke America witness’s testimony that he understood the manufacturers in Australia and Portugal to be related to BMC). Moreover, Moke International and Moke USA argue that even evidence that the MOKE mark was used by “more than one historical line of manufacturers” would not necessarily establish genericness. Id. Moke International and Moke USA also criticize the district court’s emphasis on policy considerations, asserting that it led the court to erroneously “conflate appropriation of an abandoned mark with genericness.” See Br. of Appellants 36. For example, Moke International and Moke USA assert that, by giving attention to their lack of investment of time or money in building the goodwill associated with the MOKE mark, the court improperly disregarded “the fact that BMC, the party that originally invested in building the goodwill in the MOKE mark, abandoned the mark long ago, has not used the MOKE mark in decades, has not re-introduced MOKE vehicles itself, and has taken no action to 43 prevent [Moke International and Moke USA] from using and registering the MOKE mark for their new vehicles.” Id. at 40 (internal quotation marks omitted). Moke International and Moke USA further complain that the court sought to protect the ability of third-party manufacturers and the public to use the term “Moke,” despite evidence that only Moke America has raised a competing claim to the MOKE mark and that there are sellers of similar vehicles who do not refer to their products as “Mokes.” Id. at 41-42 (invoking evidence indicating, inter alia, that one seller of similar vehicles uses E-MERGE and BEACHCOMBER, rather than MOKE). 2. We conclude that — although there is some evidence that MOKE began as an inherently distinctive mark but has since undergone genericide, as well as some evidence that MOKE is and always has been an inherently distinctive mark — there is not enough evidence on which the district court could make any proper finding of genericness or distinctiveness at all. Consequently, we are constrained to ascertain clear error in the court’s genericness finding and remand for further proceedings. Cf., e.g., Wall v. Rasnick, 42 F.4th 214, 220 (4th Cir. 2022) (recognizing that another circumstance in which clear error may be ascertained is when “there is evidence to support [the trial court’s finding],” but “the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed” (internal quotation marks omitted)). Even accepting that the existing evidentiary record sufficiently establishes that MOKE began as an inherently distinctive mark and that it was originally owned but later abandoned by BMC, the record leaves too many other questions unanswerable by a 44 preponderance of the evidence. On that score, we recognize that an inherently distinctive mark does not necessarily convert to a generic term upon abandonment. See Emergency One, 332 F.3d at 268 (explaining, to the contrary, that “[o]nce abandoned, a mark may be seized immediately and the person doing so may establish priority of use and ownership under the basic rules of trademark priority” (alteration and internal quotation marks omitted)). Nevertheless, we further recognize that abandonment of a once inherently distinctive mark does not foreclose the possibility that the mark has at some point become generic by way of genericide. So, there are questions that may or may not ultimately be significant, including the questions of whether BMC’s “apparent” corporate successors in Australia and Portugal were actually its corporate successors, and of who else used the MOKE mark, when, and in what circumstances. Of clear importance, however, are questions concerning how potential customers perceive the MOKE mark. Has MOKE been adopted by the public as a generic term for the general style of vehicles once produced by BMC? Or can MOKE yet be used to identify a particular source of such vehicles? See Glover, 74 F.3d at 59 (explaining that “[t]o become generic, the primary significance of the mark must be its indication of the nature or class of the product or service, rather than an indication of source”). Those are questions that cannot properly be answered without additional evidence — evidence that may, but is not required to, include “purchaser testimony, consumer surveys, listings and dictionaries, trade journals, newspapers, and other publications.” See Gruyere, 61 F.4th at 425-26 (quoting Glover, 74 F.3d at 59)). 45 In ruling as we do, we mean no criticism of the district court, which was left in a difficult position by the parties’ mutual insistence that MOKE is an inherently distinctive mark and their mutual failure to proffer ample relevant evidence. Indeed, we commend the court for heeding the indicators that MOKE may be a generic term and demanding proof of non-genericness, rather than simply accepting the parties’ implicit stipulation of distinctiveness. As the court rightly recognized and we once again reiterate, this trademark dispute implicates not just the rights of Moke America and of Moke International and Moke USA, but potentially also the rights of third parties and especially of the public to preserve a word in its “linguistic commons.” See Am. Online, 243 F.3d at 821. We also appreciate that a finding of genericness carries significant consequences, in that it precludes any and all claims of ownership of the proposed mark. See Sara Lee Corp. v. Kayser-Roth Corp., 81 F.3d 455, 464 (4th Cir. 1996) (recognizing that generic terms “can never be trademarks”). Put succinctly, for the sake of Moke America, Moke International and Moke USA, other sources of similar vehicles, and the public at large, it is crucial that the district court have enough evidence before it to get any finding of genericness or distinctiveness right. Because the existing evidentiary record is inadequate to that task, we vacate and remand for further proceedings in which the parties will continue to bear the burden of proving that the MOKE mark is not generic. In light of the absence of any party claiming genericness, we note that it might be helpful to the district court to, e.g., appoint a disinterested expert witness. See Fed. R. Evid. 706. We further note that, if on remand the parties yet again fail to prove non-genericness by a preponderance of the evidence, the 46 court should consider whether it is appropriate to adhere to its finding that MOKE is a generic term, or to stop short of such a finding and thus leave genericness an open question. 11 III. Pursuant to the foregoing, we vacate the judgment of the district court and remand for such other and further proceedings as may be appropriate. VACATED AND REMANDED With respect to the question of the proper disposition of this dispute, Moke International and Moke USA suggest on appeal that — even if the district court correctly ruled that the parties failed to satisfy a burden of proving non-genericness for purposes of their competing trademark ownership and trademark infringement claims — the court should not have directed the TTAB to deny Moke International and Moke USA’s application for registration of the MOKE mark on the ground that MOKE is a generic term. Rather, because they bore no burden to prove genericness or non-genericness in the PTO registration and opposition proceedings, Moke International and Moke USA contend that the court should have affirmed the TTAB decision dismissing Moke America’s opposition and leaving Moke International and Moke USA’s application pending. See Br. of Appellants 19 (citing B&B Hardware, Inc. v. Hargis Indus., Inc., 575 U.S. 138, 144 (2015), for the proposition that “[t]he party opposing registration bears the burden of proof, and if that burden cannot be met, the opposed mark must be registered” (citations omitted)). We do not unnecessarily reach and resolve that contention or several additional appellate arguments raised by the parties herein. 11 As a final matter, in August 2023, after these cross-appeals had been noted, Moke International and Moke USA filed a motion in the district court to supplement the record under Federal Rule of Appellate Procedure 10(e). Specifically, they sought to add all records of the PTO proceedings, along with various records of prior PTO trademark application examinations related to the MOKE mark. The district court, however, promptly denied that motion. Moke International and Moke USA then made a similar request in this Court, which we now deny as moot. 47 RICHARDSON, Circuit Judge, dissenting: I agree with much of the majority’s opinion. The majority correctly holds that each party asserting trademark infringement bears the burden to show that the mark is distinctive. When a mark’s distinctiveness is uncertain, this may require proffering empirical evidence that consumers do not perceive the mark as generic. And distinctiveness can be uncertain regardless of whether a party pleads genericness as a defense—for instance, when the putative mark owner’s own evidence is dubious or contradictory. So the majority was also right to reject the assertion that parties to a trademark infringement suit can always stipulate a mark’s distinctiveness. But while we agree on the overarching framework governing each party’s burden, I part ways with its application to the case before us. This is not a case where the mark’s distinctiveness is uncertain. And when distinctiveness is clear and uncontroverted, a party need not produce any more evidence to carry its burden. The district court’s erroneous conclusion that “Moke” was a generic term followed directly from two mistakes. First, it misconstrued evidence of the parties’ attempt to capture residual goodwill in the abandoned “Moke” mark as evidence of the mark’s genericness. Residual goodwill often results from trademark abandonment but is irrelevant when addressing trademark distinctiveness. Second, the district court misunderstood Don Racine’s testimony, taking him to suggest that multiple unrelated companies manufactured “Mokes,” and that consumers use “Moke” to refer to a style of vehicle, not a particular model. But even if Racine’s testimony could serve as evidence of the relevant public’s perception of “Moke” as generic—and it couldn’t—Racine suggested no such things. 48 Without these two evidentiary mistakes, the record is clear and uncontroverted: “Moke” is a distinctive mark. The majority fails to fully correct these missteps when it determines that the record is insufficient to decide the question and remands for a do-over. A proper view of the evidence, shorn of conceptual confusion, shows no support for the notion that “Moke” is generic. So I would hold that “Moke” is an inherently distinctive term protectable under the trademark laws and reverse the district court’s finding on this issue. I respectfully dissent. I. BACKGROUND The original Moke—a brand of low-speed, open-air vehicle produced by British Motor Company (“BMC”) starting in the 1960s—garnered a small but dedicated following in the United States as the car became loved as an eye-catching beach buggy. Then BMC stopped producing the Moke in 1993, and the brand fell into commercial disuse in this country. But the domestic love for the vehicle survived thanks to a small group of dedicated Moke fanatics. As the nostalgia market boomed, the parties here sought to capitalize on it by bringing the Moke brand back to life with newly produced open-air vehicles under the old name. This suit arises out of their fight over who was the first to reuse the trademark; the parties each sought to claim the Moke trademark and sued each other for trademark infringement. Yet for all the quarrels between them, this suit comes before us because of an issue on which both parties are united. Both parties agreed before the district court, and continue to agree here, that “Moke” is protectable under the trademark laws as an inherently 49 distinctive trademark—that is, by the word’s “intrinsic nature” it “serves to identify a particular source of a product.” Two Pesos, Inc., 505 U.S. at 768. This contrasts with a term that is generic, and thus unprotectable, because it “names a ‘class’ of goods or services, rather than any particular feature or exemplification of the class.” USPTO v. Booking.com B.V., 591 U.S. 549, 556 (2020). Despite the parties’ agreement, the district court—after the long, winding journey laid out in the majority opinion—found that “Moke” was generic. 1 Moke America, 671 F. Supp. 3d at 691. Key to this determination was its conclusion that both parties “employed the MOKE mark in a manner that explicitly acknowledged and commercially exploited consumers’ association of the MOKE mark with . . . BMC and its ostensible corporate successors,” rather than with themselves. Id. at 693 (emphasis in original). This The district court first found that the trademark was descriptive—one category shy of generic. Moke Am. LLC v. Am. Custom Golf Cars, Inc., 671 F. Supp. 3d 670, 685 (E.D. Va. 2024). Descriptive trademarks “describe a function, use, characteristic, size, or intended purpose of the product” such as “After Tan post-tanning lotion” and “5 Minute glue.” Sara Lee Corp. v. Kayser-Roth Corp., 81 F.3d 455, 464 (4th Cir. 1996). The district court’s ultimate genericness finding is not affected by its intermediate determination, but it’s worth pointing out that this intermediate determination of descriptiveness also rested on a conceptual error. At a hearing, evidence arose that Mini Mania, Moke America’s predecessor, used the “Moke” brand to signal the compatibility of various aftermarket car parts that Mini Mania sold. J.A. 592. The district court understood this to be evidence that “Moke” was a descriptive trademark. Moke America, 671 F. Supp. 3d at 685. That was a mistake. A phone accessory company might specify that its phone cases are designed for Apple iPhones, a shoe store might advise that its shoelaces fit best on Converse shoes, and an autobody repair shop might warn that only some of the motor oil it sells is optimal for Ford engines. These uses do not turn Apple, Converse, or Ford into descriptive marks; such marks are still inherently distinctive, even though they are colloquially used to describe another product. In any case, this conclusion says nothing about the separate question of whether Mini Mania’s use of “Moke” when selling parts established priority rights in the mark—a fact-intensive matter that should be left for the district court to address on remand. 1 50 exploitation of residual goodwill around BMC’s vehicles, the district court found, ran contrary to trademark law’s public policy goals. Id. at 695–96. The district court also thought the testimony of Don Racine—the CEO of Moke America’s predecessor in interest, Mini Mania, and the only witness called by either party during the bench trial— indicated that multiple unrelated companies produced “Mokes,” and that “Moke” was used as a term for a “style of vehicle.” Id. at 688, 698–99. Between the parties’ exploitative use of “Moke” and Racine’s testimony, the district court believed MOKE functioned “not as an indication of source, but as a talisman of a particular style,” making it “quintessential genericism.” Id. at 694. II. DISCUSSION This case is a dispute over the ownership of the “Moke” mark. Each party—Moke America on the one hand and Moke International on the other—seeks to claim “Moke” as their own and hold the other liable for trademark infringement. To ultimately prevail, each party must show two things: (1) that the “Moke” mark was distinctive, not generic; and (2) that they were the first to use it in commerce. See B & B Hardware, Inc. v. Hargis Indus., Inc., 575 U.S. 138, 142 (2015). Set aside the second, temporal-priority requirement; the district court should decide that on remand. Our only question is whether the parties produced evidence that “Moke” satisfied the first requirement. On the record before us, the parties have proven the “Moke” mark’s distinctiveness. A mark is distinctive—as opposed to generic—when it can “distinguish [a party’s] goods, including a unique product, from those manufactured or sold by others” and “indicate the source of the goods, even if that source is unknown.” 15 U.S.C. § 1127. One can satisfy 51 this element by showing that the mark is inherently distinctive, which comes in three flavors: “suggestive,” which means the mark “connotes a characteristic of a product”; “arbitrary,” which means the mark is “unrelated to the product” but “well-known in a different context”; and “fanciful,” which means the mark is “newly invented” and means nothing. Ashley Furniture Indus., Inc. v. SanGiacomo N.A., Ltd., 187 F.3d 363, 369 (4th Cir. 1999). Showing that a mark is inherently distinctive ordinarily requires little evidence—all it requires is some reason to think the term is distinctive by its “intrinsic nature.” Two Pesos, Inc. 505 U.S. at 768. And the parties had a good reason. As they explained in their post-trial briefs, “the word MOKE has no meaning in the context of automotive products,” making it either arbitrary or fanciful. 2 J.A. 913; see also J.A. 887, 892. By itself, that evidence is enough to show inherent distinctiveness. Because the term doesn’t mean anything, it is naturally understood to “distinguish a particular artisan’s goods from those of others.” B & B Hardware, 575 U.S. at 142. This conclusion tracks common sense. Without referencing anything but the word itself, it should be obvious that “Moke” is distinctive and therefore capable of “distinguish[ing]” and “indicat[ing] the source of goods. § 1127. Indeed, that is precisely why it falls into the inherently distinctive category—the distinctiveness is inherent in the “Moke” is a slang term in British English for a donkey. See Moke America, 671 F. Supp. 3d at 692. But since “trademark law is territorial,” Abitron Austria GmbH v. Hetronic Int’l, Inc., 600 U.S. 412, 426 (2023), the meaning of a slang term from outside the country likely should not bear on the analysis unless there is some reason to think that domestic consumers understand the slang’s meaning. So although the exact flavor of inherent distinctiveness doesn’t matter here, I take Moke America to be correct in its assertion that the “Moke” mark means nothing and is thereby fanciful, not arbitrary or suggestive. 2 52 word. Because, in America, “Moke” is a nonsensical word, it “almost automatically tell[s] a customer [it] refer[s] to a brand.” See Qualitex Co. v. Jacobson Prods. Co., 514 U.S. 159, 162 (1995). That should have been the ballgame. As this Court has recognized—and as the majority reiterates above—inherently distinctive marks are valid and protectable “without the holder having to make any other showing.” See OBX-Stock, Inc. v. Bicast, Inc., 558 F.3d 334, 340 (4th Cir. 2009); see also Majority Op. at 6. So by submitting the word and explaining its meaninglessness, the parties did all they needed to do to demonstrate distinctiveness. Yet rather than accept that the parties had met their burden, the district court erroneously requested more evidence that “Moke” was not a generic mark. On appeal, the parties jointly contend that the district court erred in placing upon them the burden of proving non-genericness when neither party raised the genericness of “Moke” as a defense to infringement. The parties are wrong about their burden, and the majority is right to reject their argument. All agree that a party claiming ownership in an infringement suit bears the burden of submitting the evidence that shows the mark’s distinctiveness. See, e.g., Emergency One, Inc. v. Am. Fire Eagle Engine Co., 332 F.3d 264, 269 (4th Cir. 2003) (Emergency One II). But since all marks that are not generic are distinctive, evidence of non-genericness is evidence of distinctiveness. See Booking.com, 591 U.S. at 553; see also Reese Publ’g Co. v. Hampton Int’l Commc’ns, Inc., 620 F.2d 7, 11 (2d Cir. 1980). The two are flip sides of the same coin. Thus, the burden of showing non-genericness—i.e., 53 distinctiveness—always rests with the party claiming ownership, no matter what the other party does in defense. The parties’ insistence to the contrary fails to understand that the additional evidence required to satisfy a burden of persuasion depends on the state of the evidence already in the record. This is just what it means to show something by a preponderance: The party must prove its claim is more likely true than false given the evidence before the court. True, additional evidence of distinctiveness will usually only be required when the alleged infringer raises the asserted mark’s genericness as a defense. See, e.g., Ale House Mgmt., Inc. v. Raleigh Ale House, Inc., 205 F.3d 137, 140 (4th Cir. 2000). In such cases, the genericness defense stands as an obstacle that must be surmounted through additional evidence produced by the party asserting ownership. But a genericness defense is not the only obstacle that might warrant more evidence of distinctiveness. The question before a factfinder is simply whether the evidence tilts one way or the other. So if the distinctiveness evidence proffered by the party asserting ownership were itself insufficient, dubious, or contradictory, the record would fail to satisfy the burden, even in the absence of any defense. In such cases, a factfinder would be well within its rights to ask for more. As a trivial example, imagine a party that seeks to prove the word “truck” is fanciful—but then offers as the only evidence a dictionary that contains “truck” as a common entry. It would defy common sense to think the factfinder is bound to find “truck” distinctive simply because the other party failed to plead genericness as a defense. The problem here is that the district court mistakenly believed the evidence submitted by the parties was contradictory and thus required additional evidence. 54 Specifically, it thought the record contained two pieces of evidence “on the other side of the ledger” that outweighed the facially apparent inherent distinctiveness of “Moke”: (1) the parties’ own usage of the “Moke” mark, which directed consumers “towards the Moke vehicles manufactured in the twentieth century by BMC and others, not the vehicles produced by either Plaintiff or Defendants”; and (2) Don Racine’s testimony, which the district court thought contained references to multiple unrelated companies’ products as “Mokes” and “indicate[d] that consumers in the American market understood and continue to understand MOKE as a style.” Moke America, 671 F. Supp. 3d at 693–95, 698–99. But neither of these pieces of evidence, properly understood, weighs against finding “Moke” distinctive. As to the first, the parties’ use of “Moke” is precisely what we would expect from their fully lawful efforts to take up an abandoned mark with residual goodwill. As to the second, Racine’s testimony did not support the idea that consumers use “Moke” as a generic term, both because Racine is not a member of the relevant consuming public, and because Racine’s testimony consistently used “Moke” as a specific brand. And without evidence to the contrary, the inherent distinctiveness of “Moke” should have carried the day. See Interprofession du Gruyere v. U.S. Dairy Export Council, 61 F.4th 407, 426 n.8 (4th Cir. 2023); Glover v. Ampak, Inc., 74 F.3d 57, 60 (4th Cir. 1996). I turn now to explaining these two mistakes in more detail. A. Trademark Law Permits Exploiting an Abandoned Mark The district court mistakenly thought the parties’ own usage of “Moke” cut against them because it mistook the permissible effects of trademark abandonment with the 55 symptoms of trademark genericness. To understand, we must back up and start from the foundations of trademark law. Trademark law exists to support two aims: First, it “protects the goodwill represented by particular marks”; and second, it “allows ‘consumers readily to recognize products and their source,’ preventing ‘consumer confusion.’” George & Co. LLC v. Imagination Ent. Ltd., 575 F.3d. 383, 392–93 (4th Cir. 2009) (quoting OBX-Stock, Inc., 558 F.3d at339); see also Robert G. Bone, Hunting Goodwill: A History of the Concept of Goodwill in Trademark Law, 86 B.U. L. Rev. 547, 549 (2005) (laying out trademark law’s twin aims and describing the tension between them). Seeing these aims, one might intuitively believe that the law does not grant trademark protection to a party seeking to capitalize on another’s effort in cultivating goodwill (which would contravene the first aim) by intentionally getting customers to recall the previous source of a product (which would contravene the second aim). For the mine run of lawsuits—where an undisputed trademark holder sues an alleged infringer—this intuition is correct. A party who uses a mark that “is likely to cause confusion . . . as to the origin, sponsorship, or approval of his or her goods, services, or commercial activities by another person” is liable to that other person. § 1125(a)(1)(A). 3 The alleged infringer’s intent is material to this likelihood-of-confusion analysis, as courts presume that an intentional attempt to confuse consumers will succeed. See Sara Lee This is the trademark infringement statute for unregistered trademarks. There is a separate but substantially similar trademark infringement statute for registered trademarks. See 15 U.S.C. § 1114. 3 56 Corp., 81 F.3d at 466 (citing Osem Food Indus. Ltd. v. Sherwood Foods, Inc., 917 F.2d 161, 165 (4th Cir. 1990)). So courts will find that businesses cannot intentionally draw on unearned goodwill by trying to use a mark substantially similar to another business’s active trademark. See id. But the intuition that the trademark laws prohibit capitalizing on another’s goodwill falls apart when the original trademark holder is no longer in the picture, such as when the original holder abandons the trademark. After a trademark is created and used, it does not belong to its owner for all time. By statute, a mark is “abandoned” if “its use [in the ordinary course of trade] has been discontinued with intent not to resume such use.” 15 U.S.C. § 1127. 4 Nothing else is required. And once a trademark is abandoned, it “returns to the public domain” and can “be appropriated for use by others in the marketplace in accordance with the basic rules of trademark priority.” George & Co., 575 F.3d at 400 (citation omitted); see also Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 17:2 (5th ed. 2024). Those basic rules tell us that the first to re-use the abandoned mark gains priority trademark rights, along with any residual goodwill the mark retains. George & Co., 575 F.3d at 400. The law thus freely permits the commercial One statutory quirk to note: The Lanham Act defines genericide as a second type of abandonment. “A mark shall be deemed to be ‘abandoned’ if . . . any course of the owner, including acts of omission as well as commission, causes the mark to become the generic name for the goods or services on or in connection with which it is used.” § 1127. Using “abandonment” to refer to two distinct concepts, however, engenders confusion. So I follow the general convention of using “abandonment” to refer only to a cessation of use without intent to reuse and reserve “genericide” for when a previously distinctive mark becomes generic over time. See Am. Online, Inc. v. AT&T Corp., 243 F.3d 812, 821 (4th Cir. 2001) (using “genericide”). 4 57 exploitation of an abandoned trademark’s residual goodwill from its prior association with the original holder. 5 So, in cases where a valuable mark is abandoned, there is often a scramble among businesses seeking to be the first to grab the equivalent of a $20 bill dropped on the sidewalk. The Second Circuit encountered this situation in Manhattan Industries, Inc. v. Sweater Bee by Banff, Ltd., 627 F.2d 628 (2d Cir. 1980). In that case, General Mills formally abandoned—apparently for tax purposes—its rights in the “Kimberly” trademark for high-quality women’s clothing on May 7, 1979. Id. at 629. Immediately, a “free-forall ensued.” Id. The plaintiff in the case began shipping women’s clothing with a “Kimberly” mark on May 9, just two days after General Mills’ abandonment; the defendant was one day behind, starting its “Kimberly” branded sales on May 10. Id. Crucially, nothing in the case suggests that anyone—plaintiff, defendant, or Second Circuit—thought it was illicit for either party to intentionally exploit the goodwill built by General Mills in the “Kimberly” brand. Nor does anything in the case suggest that a lingering association between “Kimberly” and General Mills stripped the mark of its distinctiveness as a source identifier. Instead, the Second Circuit correctly reiterated that “[w]hen General Mills abandoned its mark, [the plaintiff] and [the defendant] ‘were equally free to attempt to In cases materially distinct from ours where abandonment is contested, some courts look at the existence of residual goodwill not as a standalone inquiry, but as evidence supporting the original owner’s intent to resume use. See, e.g., Defiance Button Mach. Co. v. C & C Metal Prods. Corp., 759 F.2d 1053, 1060 (2d Cir. 1985). The reasoning is that a trademark owner is less likely to intentionally discontinue using a trademark with substantial residual goodwill. While perhaps sensible, such reasoning is irrelevant to this case since BMC claims no such intent. 5 58 capture the mark to their own use.’” Id. at 630 (quoting Sutton Cosmetics (P.R.) v. Lander Co., 455 F.2d 285, 288 (2d Cir. 1972)). To be sure, trademark abandonment sits uneasily alongside trademark law’s twin aims. An abandoned trademark’s value comes both from the efforts of the original owner and the possibility that the public still attributes marked products to the original source. So exploiting that value looks very much like stealing goodwill and intentionally inducing source confusion—behavior that we might expect trademark law to prohibit. For this exact reason, many scholars have argued that trademark law should account for residual goodwill and potential source confusion when assessing whether a mark has been abandoned. See, e.g., Michael S. Denniston, Residual Good Will in Unused Marks—The Case Against Abandonment, 90 Trademark Rep. 615 (2000); David S. Ruder, New Strategies for Owners of Discontinued Brands, 3 Nw. J. Tech. & Intell. Prop. 61, 63–64 (2004); Jerome Gilson & Anne Gilson LaLonde, The Zombie Trademark: A Windfall and a Pitfall, 98 Trademark Rep. 1280, 1300 (2008); Jake Linford, Valuing Residual Goodwill After Trademark Forfeiture, 93 Notre Dame L. Rev. 811, 816–17 (2018). Law review articles, however, do not bind us. Rather, the chorus of academic voices proposing a change in the law only confirms what the existing law is: Once a trademark is abandoned through the discontinuation of use and lack of intent to resume use, any value in the abandoned trademark from residual goodwill or attribution to the original source is up for grabs—a fully legal boon for the first party to re-use the mark. Trademark abandonment is like a starter gun; as soon as it goes off, anyone can immediately move to pick up the mark. As with the “Kimberly” mark in Manhattan Industries, “a mark may be 59 seized immediately and the person doing so may establish priority of use and ownership under the basic rules of trademark priority.” Emergency One II, 332 F.3d at 268 (cleaned up). The upshot is that a party seeking to claim ownership in an abandoned mark and sue another for infringement must show the same two things as any trademark claimant: distinctiveness and temporal priority. If he shows them, then the re-user of the mark stands in the same position as the first user of the mark. 6 The district court missed this. Rather than treat the parties as if they were seeking to claim a novel trademark, the district court thought that because “Moke” had previously been used, it needed to consider “commercial context”—namely, that “Moke” was an abandoned mark, and that the parties intentionally sought to capitalize on its residual goodwill created by BMC. Moke America, 671 F. Supp. 3d at 693; see also id. at 694 (“[T]he parties draw on an association between the mark and another company’s products, seeking to exploit that association to sell their own vehicles. In that effort, the parties implicitly disclaim that the MOKE mark identifies their products and their products alone.”). The district court then held the parties’ attempt to exploit this residual goodwill against them. Because it thought it would contravene trademark law’s twin aims to permit the parties to benefit from the efforts of another, the district court required each party to produce “evidence that consumers associate “Moke” with their particular products.” Id. There is one narrow caveat. At times, a court may use its equitable powers to require an abandon mark’s re-user to avoid deceptive misrepresentations that attempt by falsehoods to “pass off” the source of the new product as the original trademark owner. See Restatement (Third) of Unfair Competition § 30 cmt. a. But even in that small subset, subsequent users own the trademark and are otherwise “free to use the abandoned designation.” Id. 6 60 at 695 (emphasis added); see also id. at 695–96 (“In sum, the Court finds that bestowing trademark status on the MOKE mark would disserve the policy rationales undergirding trademark protections, irrespective of which party ultimately won the mark.”). But the district court’s understandable intuition that the parties should be prohibited from reaping the benefits created by BMC cannot support a conclusion that Moke is generic. By statute, trademark law allows the parties in this case to use the “Moke” mark and capitalize on BMC’s residual goodwill. Thus, the fact that “Moke” was previously trademarked and then abandoned changes nothing about the burden placed upon each party to show distinctiveness. Whether or not this is prudent policy, it is the law that Congress has chosen and that this court has consistently recognized. See § 1127; George & Co., 585 F.3d at 400-01; Emergency One, Inc. v. Am. FireEagle, Ltd., 228 F.3d 531, 535 (4th Cir. 2000) (Emergency One I). Nor does it matter that some consumers may indeed erroneously associate the parties’ products with BMC, a company that no longer makes Mokes. This just reflects the value remaining in the trademark. As explained, a protectable mark need only indicate a product’s source. And indicating the source is not the same thing as indicating a particular, known source. So long as consumers would think “Moke” was a source identifier, the mark is distinctive, even if they have never heard of such a brand before— or, crucially, even if they associate the brand with the wrong company. 7 The trademark If a trademark’s association with the wrong company was evidence of the trademark’s genericness, an infringer could get away with its infringement by simply being very successful and inducing lots of source confusion. But it should be clear why that sort of bootstrapping shouldn’t work. 7 61 statute makes this explicit by defining a trademark as “any word, name, symbol, or device, or any combination thereof . . . used by a person . . . to indicate the source of the goods, even if that source is unknown.” § 1127 (emphasis added). In other words, consumers need only understand that a “Moke” label plastered on a cardboard box indicates who the box came from, not what the box contains. See Rudolph Int’l., Inc. v. Realys, Inc., 482 F.3d 1195, 1198 (9th Cir. 2007) (explaining the “who-are-you/what-are-you” test). “Moke” must be able to inform consumers that a box emblazoned with “Moke” comes from the same place as every other box bearing the same “Moke” label—nothing more. 8 Here, the parties did not need to proffer empirical evidence that consumers associate “Moke” with either Moke America or Moke International, or any particular business at all. They just needed evidence that consumers would see “Moke” as a brand. B. Racine’s Testimony Was Not Evidence of Genericness Beyond this, the district court had only one other piece of evidence it thought weighed against finding “Moke” distinctive: Don Racine’s testimony, which the district court took to be evidence that “the MOKE mark is perceived in the marketplace” as a generic term for a class of car. Moke America, 671 F. Supp. 3d at 698. But the district court, possibly influenced by the background policy intuitions already discussed, erred in Not counting intentional mark duplication, in which case products bearing identical marks would come from different sources due to active infringement. We don’t consider such misuse of the mark when defining what it means to have a legally protectable mark in the first place. 8 62 multiple ways when interpreting Racine’s testimony. Racine’s testimony does not support genericness, leaving the record devoid of evidence that “Moke” is anything but distinctive. For starters, it’s unlikely that Racine’s testimony is the sort of evidence that can even support a genericness finding. “[A] generic term is one that members of the relevant public understand as identifying the type, rather than the source, of a product.” Gruyere, 61 F.4th at 414 (emphasis added). “For ordinary consumer goods”—including cars like Mokes that are sold to the public—“it is a term’s meaning to consumers, not to professionals in the trade, that is the test of genericness.” 1 McCarthy, supra, § 12:4. That is, we care only about what the car-buying public thinks. As the founder of a business that distributed Moke parts for decades, Racine is not a member of the car-buying public; rather, his understanding of the term “Moke” is likely akin to a professional’s. That Racine was also a “prolific consumer of all things MOKE,” Moke America, 671 F. Supp. 3d at 698, does not change the fact that his view is colored by his professional industry knowledge. So his understanding of “Moke” should not be taken as reflecting the relevant car-buying public’s perception of “Moke,” which is the only perception that matters. See Glover, 74 F.3d at 59–60 (finding the testimony of various knife distributors, knife experts, and knifemaking company employees to be completely irrelevant in determining the meaning of “White Tail,” a pocketknife brand, to the relevant public of actual or potential purchasers of pocketknives). Even were we to accept Racine’s testimony as evidence of the relevant public’s understanding of “Moke,” the content of Racine’s testimony does not suggest “Moke” is generic. The district court thought Racine referred to vehicles produced by “an unnamed 63 Chinese company” as “Mokes,” suggesting that a Moke was not a particular brand, but a type of vehicle. Moke America, 671 F. Supp. 3d at 698. But Racine did no such thing. In response to a question about car kits that Racine sold, kits that were used to modify Mini Coopers to look like Mokes, Racine answered that “the Moke only came from China”— but then immediately reworded and clarified that “[t]he kit was only from China.” J.A. 541–42 (emphasis added). In context, Racine was, at most, calling the car kit from China a “Moke car kit.” But a third-party kit used to graft Moke aesthetics onto another car can be labelled a “Moke car kit” without implying that “Moke” is generic in the same way thatan electronics store selling an “Apple iPhone 12 port to USB-C converter” phone accessory does not imply that “Apple iPhone 12” is generic. The same logic applies to Racine’s discussion of Mokes made in Australia and Portugal. The district court stated that Racine “could not definitively testify as to whether the Mokes produced by Australia and Portugal were in fact manufactured by bona fide corporate successors to BMC,” yet referred to them all as “Mokes” anyway. Moke America, 671 F. Supp. 3d at 698. This again misconstrues the testimony. Racine was clear that Mokes manufactured in Australia and Portugal were BMC-made Mokes, and that “Moke” referred solely to BMC’s line of vehicles. When asked whether any company “before 2015 manufactured Mokes,” Racine responded that “BMC, whatever that parent company really is [named], 9 nobody other than them made them.” J.A. 520. Racine then The reason for Racine’s hesitation after naming BMC is that the name of the parent company that owned Moke changed repeatedly. As Racine later explained, “[t]he English were not good at maintaining businesses and [BMC] went through a number of hands (Continued) 9 64 clarified that “the Australian Moke . . . was a continuation of the exact same product, simply made in Australia rather than made in the UK,” and that “a version of BMC in Portugal made Mokes.” J.A. 521–22. In fact, far from being unclear, Racine was unequivocal in reserving the term “Moke” for either BMC’s vehicles or for his own company’s cars that were meant to mimic BMC’s Mokes, and his one statement that could be interpreted to support genericness was induced by the district court’s own confusion. 10 After Racine stated that “Moke was just another vehicle within [BMC’s] umbrella of the various cars they were making,” the district court interjected and asked, “It’s a style of vehicle; is that right?” J.A. 537–58. Racine politely replied, “You could say that, sure.” J.A. 538. The district court latched onto this “style” concession and used it repeatedly to support its conclusion that “Moke” is generic. See Moke America, 671 F. Supp. 3d at 688, 691, 693, 694, 701. The district court’s conclusion that Racine was using “Moke” as a generic “style” reflects a misunderstanding of how car naming conventions work. A car maker’s name can be a trademark, and a car model’s name can itself also be a trademark. 11 Stating that . . . BMC was one of the names . . . and then they became Jaguar Rover Triumph . . . [a]nd then it became Leyland.” J.A. 537. The hesitation was not out of any uncertainty as to whether Mokes were made by another manufacturer. For reasons given in Footnote 1, Racine’s description of third-party aftermarket parts compatible with Moke vehicles as “Moke parts” should not be interpreted as using “Moke” in a non-distinct manner. 10 This isn’t limited to cars. The phrase “Apple iPhone” consists of two trademarks: the maker, Apple, and the product line, iPhone. The phrase “McDonald’s Big Mac” consists of two trademarks: the maker, McDonald’s, and the product line, Big Mac. And so on. 11 65 Moke was a vehicle within BMC’s umbrella was merely saying that Moke was a model of BMC. Moke America’s counsel spotted the district court’s confusion immediately and tried to clarify by drawing a comparison. After giving the example that “Cadillac has a car that is called the Escalade,” counsel asked Racine “Is the Escalade a style or a brand of a type of car?” J.A. 539. Racine replied, “I guess it’s a brand.” J.A. 539. Counsel then asked whether “Moke” was a style or a brand of a type of car, and Racine again stated it was a brand, adding for clarity that “Moke is indeed unique to itself.” J.A. 539. It is thus abundantly clear that Racine used “Moke” to refer to the brand of cars made by BMC, not to gesture at a generic style of car, like a sedan. So the district court erred in thinking Racine’s testimony weighed against finding distinctiveness. And without Racine’s testimony, there is nothing left to suggest that “Moke” was generic. C. The Majority Now that we’ve retraced the district court’s missteps, we can see how the majority opinion improperly leaves open the possibility for the same errors to be made on remand. The majority recognizes that a party usually satisfies its burden of showing nongenericness simply by demonstrating that a mark is inherently distinctive. But then the majority adds that the possibility a mark has become generic over time through usage—socalled “genericide”—might raise the parties’ burden to show evidence of non-genericness in a way that the possibility of genericness ab initio for a new mark does not. See Majority Op. at 36. The majority thus suggests that the district court might, on remand, require additional evidence to show distinctiveness. 66 But a party’s burden to show distinctiveness in the present does not change based on whether the mark was distinctive in the past. Cf. Lucky Brand Dungarees, Inc. v. Marcel Fashions Grp., Inc., 590 U.S. 405, 415 (2020) (“[I]n the trademark context,[] the enforceability of a mark and likelihood of confusion between marks often turns on extrinsic facts that change over time.”). A party seeking trademark protection today must answer the same questions, regardless of if the mark he seeks to protect is new or abandoned: Is the mark, today, distinctive because it is able to “indicate the source of goods”? 15 U.S.C. § 1127. Or instead, is the mark, today, generic because it merely “names a ‘class’ of goods or services”? Booking.com, 591 U.S. at 556. So long as the evidence submitted can answer these questions, nothing else is needed. In other words, when checking whether a term is generic, we don’t care how it is or became generic. We just care whether it is now. Genericide or genericness ab initio, it makes no difference. To wit, “escalator” used to be trademarked, but “coffee” never was. But nothing in trademark law distinguishes these words. See Am. Online, Inc., 243 F.3d at 820–21. That only the former was once a protected trademark is a fun trivia fact, but legally inconsequential. The majority’s concerns about genericide’s relevance might stem from the same understandable intuition that tripped up the district court: The evidence required to claim a mark’s ownership feels like it should be different when the mark has been previously used and then abandoned. Without rejecting this mistaken intuition, the majority remands this case to the district court for more factfinding because it believes that the record is insufficient to decide one way or the other. But the evidence needed to show distinctiveness doesn’t change based on whether the mark had a prior owner. And when 67 the record in this case is viewed properly, the evidence to support distinctiveness is sufficient. The parties proffered evidence that “Moke” is an inherently distinctive word. That is enough. * * * Stripped of Racine’s misunderstood testimony and of all legally inconsequential concerns about exploiting an abandoned mark’s residual goodwill, it becomes apparent that the evidentiary ledger here is one-sided. The evidence shows that “Moke” is inherently distinctive, and no evidence suggests it’s generic. The district court’s conclusion to the contrary depends on a misunderstanding of trademark abandonment law and an erroneous interpretation of Racine’s testimony. And although the majority does not affirm the district court, its decision to order a do-over suggests that the district court could make the same missteps again. That is unfortunate. I would instead follow established trademark law, reverse the district court, and hold that “Moke” is a distinctive trademark, remanding only for proceedings downstream of the distinctiveness determination. I therefore respectfully dissent. 12 While common sense isn’t binding precedent, I would ask readers, who are likely members of the relevant class of consumers, to ask themselves: Before reading these opinions, had you ever heard of a Moke? Did “Moke” strike you as a common word akin to “minivan” or “aspirin” when you first read it? Would you fully understand a sign that said, “No SUVs and trucks on the path, only compacts and Mokes allowed”? If you answered “no” to all three questions—as I suspect you did—query whether the majority’s decision not to reverse makes sense. 12 68

Some case metadata and case summaries were written with the help of AI, which can produce inaccuracies. You should read the full case before relying on it for legal research purposes.

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.