Harvard Law Review Harvard Law Review Harvard Law Review


Trademark Injury in Law and Fact: A Standing Defense to Modern Infringement

Trademark law and the law of standing have grown apart. Trademark law has expanded to recognize infringement in the absence of concrete harm to trademark owners. Meanwhile, the law of standing has contracted to reject claims by plaintiffs whose injuries are not concrete. This Note argues that under current Article III scrutiny, these trademark claims of abstract harm should not survive.

Taken seriously, standing doctrine should limit the excesses of trademark law by imposing the harm requirement that trademark needs and lacks. Part I describes the concept of injury in modern trademark infringement. Liability is predicated on a likelihood of consumer confusion, even when there is no reason to believe that confusion has produced any harm. Part II explains the harm requirement in current standing doctrine. Last Term, in TransUnion LLC v. Ramirez,1×1. 141 S. Ct. 2190 (2021). the Supreme Court held that a legal violation is not an injury in fact unless it entails concrete harm.2×2. See id. at 2204. To be concrete, the injury must have a “close historical or common-law analogue.”3×3. Id. Part III argues that the modern conception of likely confusion flunks the concrete-harm test because trademark law traditionally required business harm to producers. In the absence of business harm, the injury that trademark owners sustain from others’ use of their marks is closer to a nonjusticiable moral injury. When trademark owners do not establish business harm, courts must dismiss their claims on standing grounds. Part IV assesses the strengths and weaknesses of the standing defense and concludes that it is both constitutionally mandated and practically worthwhile.

I. Modern Trademark Doctrine: Legal Injury Without Concrete Harm

In trademark law, courts rely on the statutory likelihood-of-confusion standard as proof positive of infringement.4×4. See 15 U.S.C. § 1125(a)(1)(A); Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 780 (1992) (Stevens, J., concurring in the judgment) (“The federal courts are in agreement that . . . the test for liability is likelihood of confusion . . . .”). Scholars broadly agree that liability for trademark infringement stretches beyond legitimate claims of injury.5×5. E.g., Robert G. Bone, Taking the Confusion Out of “Likelihood of Confusion”: Toward a More Sensible Approach to Trademark Infringement, 106 Nw. U. L. Rev. 1307, 1310 (2012) (criticizing the overbreadth of modern trademark liability, which “fails to appreciate the central importance of harm to the infringement analysis”); Mark A. Lemley, The Modern Lanham Act and the Death of Common Sense, 108 Yale L.J. 1687, 1688 (1999) (“Courts protect trademark owners against uses that would not have been infringements even a few years ago . . . . [T]hese changes have loosed trademark law from its traditional economic moorings and have offered little of substance to replace them.”); Rebecca Tushnet, Registering Disagreement: Registration in Modern American Trademark Law, 130 Harv. L. Rev. 867, 869 (2017) (“Trademark scholars widely agree that our current system for evaluating what rights a trademark owner should have over others’ uses of their (or similar) marks is broken. Courts too readily find too many acts to be infringing even when they’re harmless or actually useful to consumers.”). This Part explains how liability exceeds harm.

Trademarks are words, symbols, and other devices that identify the source of products.6×6. 15 U.S.C. § 1127. By reserving the use of a trademark for the source that it designates, trademark rights help consumers more efficiently locate the products they want and protect producers’ investments in their brands.7×7. William M. Landes & Richard A. Posner, Trademark Law: An Economic Perspective, 30 J.L. & Econ. 265, 269–70 (1987). But these rights have costs. By granting producers exclusive control over their marks, trademark rights hurt competition. They prevent others from selling products with similar marks that consumers recognize as substitutes and prefer. As trademarks have become commodities valued in and of themselves, trademark rights have awarded owners monopolies over product markets that jeopardize social welfare.8×8. See Glynn S. Lunney, Jr., Trademark Monopolies, 48 Emory L.J. 367, 422–31 (1999). In addition to suppressing competition, trademark rights inhibit expression. Trademarks are a vocabulary, a system of words and symbols, that trademark owners deploy to stifle criticism of or even mere references to their products.9×9. See Rochelle Cooper Dreyfuss, Expressive Genericity: Trademarks as Language in the Pepsi Generation, 65 Notre Dame L. Rev. 397, 403 (1990). This speech control is especially restrictive in a culture where brands are pervasive. Whatever benefits trademark rights confer must be weighed against these costs to competition and expression for protection to be worthwhile.10×10. William McGeveran & Mark P. McKenna, Confusion Isn’t Everything, 89 Notre Dame L. Rev. 253, 277 (2013).

Modern trademark law has failed at striking this balance among policy goals because it does not take account of producer or consumer harm. Rather, the law permits the assertion of trademark rights at the expense of competition and expression, even when consumers and producers have not incurred losses whose redress might justify these costs. The test for trademark infringement is whether the defendant’s use of a mark is likely to cause consumer confusion.11×11. 15 U.S.C. § 1125(a)(1)(A). The likelihood-of-confusion test applies to federal and state law infringement claims. Restatement (Third) of Unfair Competition § 21 cmt. a (Am. L. Inst. 1995). Likelihood of confusion is assessed through a multifactor inquiry that considers such elements as the strength of the plaintiff’s mark, the similarity between the parties’ marks, the similarity between the parties’ products, the channels in which the products are sold, consumer sophistication, the defendant’s bad intent, and evidence of actual confusion.12×12. E.g., AMF Inc. v. Sleekcraft Boats, 599 F.2d 341, 348–49 (9th Cir. 1979). Courts have recognized increasing varieties of confusion. Traditionally, confusion was limited to “source confusion,” in which a defendant’s use of a similar mark leads consumers to believe that the plaintiff produced the defendant’s product.13×13. 4 J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition § 24:6 (5th ed.) (Westlaw) (last visited October 3, 2021). More recently, courts have expanded trademark rights to remedy “sponsorship or affiliation confusion,” confusion about the plaintiff’s connection with the defendant’s product; “initial-interest confusion,” confusion that arises but is cleared up before a purchase; and “post-sale confusion,” confusion after purchase not by the buyer but by other viewers of the product.14×14. See id. §§ 23:6–:8. “Today, the test of ‘likelihood of confusion’ encompasses any type of confusion,” Professor J. Thomas McCarthy writes in his leading treatise, summarizing the current state of affairs.15×15. Id. § 24:6.

The likelihood-of-confusion test is widely panned. The test is open-ended and subjective, producing uncertainty and expensive litigation.16×16. See William McGeveran, Rethinking Trademark Fair Use, 94 Iowa L. Rev. 49, 61–71 (2008). It is also a weak tool for assessing likely confusion. Courts consider but do not require evidence of actual confusion.17×17. E.g., Maker’s Mark Distillery, Inc. v. Diageo N. Am., Inc., 679 F.3d 410, 423 (6th Cir. 2012). When evidence is presented, meager figures may suffice, such as confusion in ten to fifteen percent of consumers.18×18. See 4 McCarthy, supra note 13, § 23:2; 6 id. § 32:188. Some of the factors that tend to be weighted most heavily, including the strength of the plaintiff’s mark and the defendant’s bad faith, are poor proxies for consumer confusion.19×19. See Bone, supra note 5, at 1343–47. The proliferation of actionable confusion theories has skewed the analysis in favor of trademark owners, widening rights and increasing the costs to competition and expression they impose.20×20. See Michael Grynberg, Trademark Litigation as Consumer Conflict, 83 N.Y.U. L. Rev. 60, 96–109 (2008).

The most fundamental criticism of the test pertains to the concept of likely confusion itself. Notably absent from likely confusion is whether that confusion causes harm. This absence is a matter of blackletter law. “All that must be proven to establish liability and the need for an injunction against infringement is the likelihood of confusion,” McCarthy’s treatise states.21×21. 5 McCarthy, supra note 13, § 30:2.50. “In effect, injury to the trademark owner is presumed.”22×22. Id. As many have recognized, this presumption of harm from likely confusion is unwarranted. Professor Robert Bone explains: “[T]he likelihood of confusion test suffers from a normative gap. It focuses exclusively on the probability of confusion when it should also consider confusion-related harm and the reasons for redressing that harm.”23×23. Bone, supra note 5, at 1309. Given the many infirmities of the test, courts are prone to discerning infringement when no harm is done.24×24. See Rebecca Tushnet, What’s the Harm of Trademark Infringement?, 49 Akron L. Rev. 627, 627 (2016).

The temptation to bring claims of harmless confusion is exacerbated by the structure of trademark lawsuits, which trademark owners control. The Lanham Act,25×25. Ch. 540, 60 Stat. 427 (1946) (codified as amended in scattered sections of 15 U.S.C.). the federal trademark statute, deputizes trademark holders to file lawsuits on behalf of themselves and consumers.26×26. See 15 U.S.C. § 1114(1). This setup benefits consumers in that trademark owners have better information and more resources to discover misconduct and bring claims.27×27. Laura R. Bradford, Trademark Law and Agency Costs, 55 IDEA 193, 197 (2015). But it also means that trademark owners get to decide for consumers which uses of their marks are worth disputing. Self-interest invites producers to perceive consumer confusion in the absence of real harm.28×28. See Grynberg, supra note 20, at 72. As Professor Mark McKenna writes: “Mark owners are able to characterize almost every conceivable use of their trademarks in consumer confusion terms, and because courts have simply equated confusion with harm, they have lacked the tools to resist novel confusion claims.”29×29. Mark P. McKenna, Testing Modern Trademark Law’s Theory of Harm, 95 Iowa L. Rev. 63, 67 (2009). In the name of consumer confusion, McDonald’s prevented a dentist’s office from continuing to call itself “McDental,”30×30. McDonald’s Corp. v. Druck & Gerner, DDS., P.C., 814 F. Supp. 1127, 1139 (N.D.N.Y. 1993). the Motion Picture Association of America halted a designer’s attempt to register the trademark “Rated R Sportswear” for clothing,31×31. Motion Picture Ass’n of Am. Inc. v. Respect Sportswear Inc., 83 U.S.P.Q.2d 1555, 1564 (T.T.A.B. 2007). and Elvis Presley’s estate got a King-themed dive bar to change its name from “Velvet Elvis” to “Velvet Melvin.”32×32. Kaitlyn Tiffany, Why Celebrities Try to Trademark Their Catchphrases and Baby Names, Vox (Apr. 19, 2019, 2:40 PM), https://www.vox.com/the-goods/2019/4/19/18507920/celebrity-trademark-history-baby-names-taylor-swift [https://perma.cc/Q2WN-GQDM]; see also Elvis Presley Enters., Inc. v. Capece, 141 F.3d 188, 207 (5th Cir. 1998). The malleability of likely confusion makes it an ideal framing device that trademark owners can mold to almost any case they may wish to bring.

II. Modern Standing Doctrine: Legal Injury Is Not Factual Injury

Whereas trademark law registers harm from “confusion in the air,”33×33. Rebecca Tushnet, Intellectual Property as a Public Interest Mechanism, in The Oxford Handbook of Intellectual Property Law 95, 106 (Rochelle Dreyfuss & Justine Pila eds., 2018). the law of standing requires harm on the ground. Under Article III, courts have grown increasingly skeptical of congressional definitions of injury, insisting that plaintiffs show concrete harm independent of a statutory violation. This Part explains the doctrine of standing, focusing on the concrete-harm requirement that the Court emphasized in its recent TransUnion decision.

To bring a claim in federal court, plaintiffs must have standing: they must prove their personal stake in the matter, by demonstrating that the defendant’s conduct caused them harm.34×34. TransUnion LLC v. Ramirez, 141 S. Ct. 2190, 2203 (2021). As Justice Scalia put it, the standing inquiry boils down to asking plaintiffs, “What’s it to you?”35×35. Id. (quoting Antonin Scalia, The Doctrine of Standing as an Essential Element of the Separation of Powers, 17 Suffolk U. L. Rev. 881, 882 (1983)). To answer that question to the satisfaction of federal courts, allegedly aggrieved plaintiffs must establish (1) an “injury in fact,” namely, a harm that is concrete, particularized, and actual or imminent, not conjectural or hypothetical; (2) the injury was likely caused by the defendant; and (3) the injury would likely be redressed by judicial relief.36×36. Lujan v. Defs. of Wildlife, 504 U.S. 555, 560 (1992) (internal quotation marks omitted) (quoting Whitmore v. Arkansas, 495 U.S. 149, 155 (1990)); see id. at 560–61. Plaintiffs must show standing for each claim they make and for each form of relief they seek, according to the evidentiary standards applicable at the relevant stage of litigation.37×37. TransUnion, 141 S. Ct. at 2208.

Standing means to protect the separation of powers in two primary ways. First, it carries out the Article III restriction that federal courts resolve only “Cases” or “Controversies.”38×38. Id. at 2203 (quoting U.S. Const. art. III, § 2). When plaintiffs have a personal stake, federal courts avoid adjudicating abstract disputes and limit their purview to “real controversy with real impact on real persons.”39×39. Id. (quoting Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2103 (2019)). Second, standing intends to prevent courts from infringing on the Executive’s Article II authority.40×40. See Lujan, 504 U.S. at 576–77; Allen v. Wright, 468 U.S. 737, 760–61 (1984); Cass R. Sunstein, What’s Standing After Lujan? Of Citizen Suits, “Injuries,” and Article III, 91 Mich. L. Rev. 163, 213 (1992) (“[M]any of the recent standing cases might be thought to be Article II cases masquerading under the guise of Article III; we may even say that the Article II tail is wagging the Article III dog.”). If unharmed private parties could sue defendants for legal violations of their choosing, then they could use courts to enforce the law according to their personal agendas, rather than the public interest.41×41. See TransUnion, 141 S. Ct. at 2207. Article II reserves these enforcement decisions for the executive branch.42×42. See id.

In seeking to restrain judicial power and to preserve executive power, the Court has increasingly scrutinized and rejected congressional definitions of injury.43×43. This Note applies rather than critiques the Court’s standing doctrine, though critiques are abundant. As many have rightly complained, judicial scrutiny of congressional action raises its own separation of powers concerns. E.g., id. at 2221 (Thomas, J., dissenting) (“In the name of protecting the separation of powers, this Court has relieved the legislature of its power to create and define rights.” (citation omitted) (citing id. at 2203, 2207 (majority opinion))); William A. Fletcher, The Structure of Standing, 98 Yale L.J. 221, 233 (1988) (“[S]uperimposing an ‘injury in fact’ test upon an inquiry into the meaning of a statute is a way for the Court to enlarge its powers at the expense of Congress.”); Sunstein, supra note 40, at 234 (criticizing the injury-in-fact requirement for “undermin[ing] Congress’ power to create property rights where they had not existed before”). Recent standing decisions have emphasized the need for factual injury, and its distinction from legal injury.44×44. See William Baude, Standing in the Shadow of Congress, 2016 Sup. Ct. Rev. 197, 215; Rachel Bayefsky, Constitutional Injury and Tangibility, 59 Wm. & Mary L. Rev. 2285, 2295–300 (2018). Although ostensibly derived from Article III, the concept of factual injury did not appear in the Court’s opinions until 1970.45×45. TransUnion, 141 S. Ct. at 2219 (Thomas, J., dissenting). Up to that point, legal injury — a legal violation and a cause of action — conferred the right to sue.46×46. Id. Then, in Association of Data Processing Service Organizations v. Camp,47×47. 397 U.S. 150 (1970). the Court introduced the term “injury in fact,” which became an additional means by which plaintiffs could establish standing.48×48. See TransUnion, 141 S. Ct. at 2219 (Thomas, J., dissenting). Injury in fact served to “broaden[] the categories of injury that may be alleged in support of standing,” since plaintiffs could sue based on either a legal or a factual injury.49×49. Sierra Club v. Morton, 405 U.S. 727, 738 (1972). Yet legal injury alone still sufficed: “[T]he actual or threatened injury required by Art. III may exist solely by virtue of ‘statutes creating legal rights, the invasion of which creates standing.’”50×50. Havens Realty Corp. v. Coleman, 455 U.S. 363, 373 (1982) (quoting Warth v. Seldin, 422 U.S. 490, 500 (1975)). In the landmark 1992 decision Lujan v. Defenders of Wildlife,51×51. 504 U.S. 555 (1992). however, the Court made clear that factual injury was a separate prerequisite. Congress could create new sources of standing by “elevating to the status of legally cognizable injuries concrete, de facto injuries that were previously inadequate in law.”52×52. Id. at 578. In a concurrence to Lujan, Justice Kennedy expounded a broader view of Congress’s authority to create standing. He maintained that “Congress has the power to define injuries and articulate chains of causation that will give rise to a case or controversy where none existed before.” Id. at 580 (Kennedy, J., concurring in part and concurring in the judgment). The Court continued to quote this language with approval in Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1549 (2016), but TransUnion cited only the Lujan majority’s approach, 141 S. Ct. at 2205. But factual injury was an “irreducible constitutional minimum.”53×53. Lujan, 504 U.S. at 560.

Over the past five years, the Court has widened the divide between factual and legal injury by sharpening the concrete-harm requirement. This process began with the 2016 decision Spokeo, Inc. v. Robins.54×54. 136 S. Ct. 1540. Plaintiff Thomas Robins alleged that consumer reporting agency Spokeo had disseminated false information about him, violating the Fair Credit Reporting Act55×55. 15 U.S.C. §§ 1681–1681x. (FCRA).56×56. Spokeo, 136 S. Ct. at 1546. Regardless of the legal injury, the Court held that Robins needed to demonstrate concrete factual injury to establish standing.57×57. Id. at 1548. Before Spokeo, the Court had treated “concrete” and “particularized” as rough synonyms that referred to the plaintiff suffering in an individual way.58×58. Bayefsky, supra note 44, at 2298–99. In Spokeo, however, the Court specified that concreteness was an “independent requirement” meaning “‘real,’ and not ‘abstract.’”59×59. Spokeo, 136 S. Ct. at 1548. Robins’s injury was particularized — it was his personal statutory rights that were violated — but that was no longer sufficient.60×60. See id. To determine whether the harm was concrete, “both history and the judgment of Congress play important roles.”61×61. Id. at 1549. For critiques of Spokeo’s vagueness, see, for example, Baude, supra note 44, at 224–27; and Bayefsky, supra note 44, at 2317–21. This mealymouthed test urged both scrutiny of and deference to congressional definitions of injury. It was “instructive” to consider whether the injury had a “close relationship to a harm that has traditionally been regarded as providing a basis for a lawsuit in English or American courts.”62×62. Spokeo, 136 S. Ct. at 1549. But it was “also instructive” to consider congressional judgment “because Congress is well positioned to identify intangible harms that meet minimum Article III requirements.”63×63. Id. Nevertheless, the Court cautioned that a plaintiff does not “automatically satisf[y] the injury-in-fact requirement whenever a statute grants a person a statutory right and purports to authorize that person to sue to vindicate that right.”64×64. Id. Ultimately, the Spokeo Court did not decide whether Robins had established concrete injury but instead remanded the case to the Ninth Circuit, which determined that he had.65×65. Robins v. Spokeo, Inc., 867 F.3d 1108, 1118 (9th Cir. 2017). The suggestion that a statutory violation might not confer standing was just a warning.

This past Term, in TransUnion v. Ramirez, the Court carried out Spokeo’s threat. It rejected standing for the majority of class members in an FCRA lawsuit because they had established only statutory violations without concrete injury.66×66. TransUnion LLC v. Ramirez, 141 S. Ct. 2190, 2200 (2021). In TransUnion, plaintiff Sergio Ramirez sued the credit reporting agency on behalf of a putative class of over 8,000 consumers who alleged that TransUnion had wrongly labeled them as potential terrorists.67×67. See id. at 2200–02. Writing for the Court, Justice Kavanaugh held that the consumers did not sustain concrete injury just because TransUnion had mischaracterized them in its internal credit files.68×68. Id. at 2210. Concrete injury required something more, such as dissemination of the false reports to retailers.69×69. Id. at 2212.

TransUnion severed the relationship between legal injury and factual injury that Lujan and Spokeo had strained. The opinion largely dispensed with any remaining deference to congressional judgments of injury. Although courts needed to pay “due respect” to legislative enactments,70×70. Id. at 2204. “Congress’s creation of a statutory prohibition or obligation and a cause of action does not relieve courts of their responsibility to independently decide whether a plaintiff has suffered a concrete harm under Article III.”71×71. Id. at 2205. Stated simply, “an injury in law is not an injury in fact.”72×72. Id. The opinion was scattered with one-liners from the lower courts after Spokeo hammering home this same idea.73×73. See id. at 2203 (quoting Casillas v. Madison Ave. Assocs., Inc., 926 F.3d 329, 333 (7th Cir. 2019)); id. at 2205 (quoting Hagy v. Demers & Adams, 882 F.3d 616, 622 (6th Cir. 2018); Trichell v. Midland Credit Mgmt., Inc., 964 F.3d 990, 999 n.2 (11th Cir. 2020)). Most memorably, Justice Kavanaugh opened and closed the opinion with his own catchphrase: “No concrete harm, no standing.”74×74. Id. at 2200, 2214.

Notably, in its concrete-harm test, the Court reiterated only the historical prong of the test from Spokeo. The congressional-judgment prong was conspicuously absent. To establish concrete harm, plaintiffs needed to identify a “close historical or common-law analogue for their asserted injury,” though not “an exact duplicate.”75×75. Id. at 2204. The Court did not define the contours of the relevant history or common law, but it reiterated Spokeo’s search for a harm closely related to one “traditionally recognized as providing a basis for a lawsuit in American courts.”76×76. Id. at 2200 (citing Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1549 (2016)). Most obviously, concrete injury included tangible harm, such as physical or monetary injury.77×77. Id. at 2204. But intangible harm with the requisite traditional pedigree also qualified, such as “reputational harms, disclosure of private information, and intrusion upon seclusion,” as well as constitutional injury.78×78. Id. The Court conceded that the test “can be difficult to apply in some cases” but maintained that it was nevertheless “essential” to preserving the separation of powers.79×79. Id. at 2207.

In applying the test to the facts of TransUnion, the Court articulated a vision of “concrete harm” that sought discernible consequences. The Court required dissemination of plaintiffs’ credit reports because dissemination was an essential element of defamation, the traditional harm against which the Court assessed their injury.80×80. Id. at 2209. But dissemination also made the harm concrete in a material sense. Rephrasing the concrete-harm test for this context, Justice Kavanaugh asked, “‘if inaccurate information falls into’ a consumer’s credit file, ‘does it make a sound?’”81×81. Id. (quoting Owner-Operator Indep. Drivers Ass’n v. U.S. Dep’t of Transp., 879 F.3d 339, 344 (D.C. Cir. 2018)). For harm to be concrete, it needed to produce a perceptible change in the world outside itself.

Along these lines, the Court held that an unrealized risk of harm was not concrete harm in claims for damages. For injunctive relief, a “sufficiently imminent and substantial” risk of harm sufficed.82×82. Id. at 2210. For the damages that the TransUnion plaintiffs sought, however, present harm was needed.83×83. Id. at 2210–11. Justice Kavanaugh illustrated the point with an analogy:

Suppose that a woman drives home from work a quarter mile ahead of a reckless driver who is dangerously swerving across lanes. The reckless driver has exposed the woman to a risk of future harm, but the risk does not materialize and the woman makes it home safely. As counsel for TransUnion stated, that would ordinarily be cause for celebration, not a lawsuit.84×84. Id. at 2211.

What the Court was after was a collision between risk and reality. Until the crash, there was no material consequence to the woman and thus no concrete harm. One could imagine cases where an unrealized risk was far from a “cause for celebration” — where, for example, awareness of the risk caused distress. The Court acknowledged the possibility of such scenarios, in which “the exposure to the risk of future harm itself causes a separate concrete harm.”85×85. Id. But that was not the case here, where the woman was too far ahead of the danger to know it was there, and it was not the case in TransUnion, where the plaintiffs did not allege emotional injury.86×86. See id. at 2211 n.7. Precollision, and predissemination, the risks were not felt. They did not “make a sound.”

Concrete harm also required satisfactory evidence. The plaintiffs had shown that they had requested copies of their credit reports, which consumers generally did not do unless they were considering a transaction that would have triggered the release of their reports to third parties.87×87. Id. at 2212. But the Court was not persuaded of dissemination: the “inferences on which the argument rests are too weak,” in light of the burden of proof at trial and the availability of direct evidence had dissemination occurred.88×88. Id. Harm did not always need to be easy to measure or prove: “[T]here is a significant difference between (i) an actual harm that has occurred but is not readily quantifiable, . . . and (ii) a mere risk of future harm.”89×89. Id. at 2211. When evidence was available, however, the Court expected plaintiffs to have it.

After TransUnion, it is abundantly clear that a statute is only the starting point of the standing analysis. Regardless of legal injury, courts must independently determine whether plaintiffs suffered factual injury. Factual injury requires concrete harm, namely, harm that is closely related to a historical or common law injury. Concrete harm should “make a sound”: it should produce material change in the world. Plaintiffs must present evidence that concrete harm has occurred or, in the case of an injunction, imminently will.

III. Trademark Standing After TransUnion

The likelihood-of-confusion test does not require concrete harm, but Article III standing does. By recognizing trademark infringement in the absence of concrete harm, courts have favored plaintiffs that lack standing after TransUnion.

Justiciability based critiques of trademark law are rare. Typically, scholars have sought to introduce harm into the infringement analysis by altering the merits inquiry.90×90. See, e.g., Bone, supra note 5, at 1363 (arguing that courts should assess confusion-related harm); Mark A. Lemley & Mark McKenna, Irrelevant Confusion, 62 Stan. L. Rev. 413, 450 (2010) (arguing that courts should assess whether confusion is material); Rebecca Tushnet, Running the Gamut from A to B: Federal Trademark and False Advertising Law, 159 U. Pa. L. Rev. 1305, 1360 (2011) (same); McGeveran & McKenna, supra note 10, at 295 (proposing defenses that aim to narrow the scope of liability to harmful conduct). On those occasions when courts have been perturbed by the lack of harm from confusion, they have generally responded by denying remedies, not standing.91×91. E.g., Adidas Am., Inc. v. Skechers USA, Inc., 890 F.3d 747, 759 (9th Cir. 2018) (denying a preliminary injunction because there was no evidence of irreparable harm to the plaintiff); Herb Reed Enters., LLC v. Fla. Ent. Mgmt., Inc., 736 F.3d 1239, 1250–51 (9th Cir. 2013) (same). Those that have raised justiciability concerns have described them in passing or in other contexts. Several commentators have observed that the harms in some intellectual property cases do not constitute injury in fact, but none have focused on trademark’s basic likelihood-of-confusion injury.92×92. In the field of trademark, Professor Rebecca Tushnet has questioned “whether the sometimes outré theories of trademark harm we see can really survive current Article III scrutiny.” Rebecca Tushnet, Robinhood’s Newsletter Isn’t Commercial Advertising, Rebecca Tushnet’s 43(B)log (Aug. 31, 2021, 1:16 PM), https://tushnet.blogspot.com/2021/08/robinhoods-newsletter-isnt-commercial.html [https://perma.cc/96UQ-8F8X]. Professor Jeremy Sheff has rejected the justiciability of trademark claims premised on the alleged harm of lost control over reputation. Jeremy N. Sheff, Marks, Morals, and Markets, 65 Stan. L. Rev. 761, 799–800 (2013). Beyond trademark, scholars have spoken in broader terms about intellectual property violations that may not satisfy the injury-in-fact requirement. See Seth F. Kreimer, “Spooky Action at a Distance”: Intangible Injury in Fact in the Information Age, 18 U. Pa. J. Const. L. 745, 793 (2016) (“The intangible ‘harm’ suffered by the intellectual property owner whose rights are violated is a particularized injury only because the law makes it so.”); Daniel J. Solove & Danielle Keats Citron, Standing and Privacy Harms: A Critique of TransUnion v. Ramirez, 101 B.U. L. Rev. Online 62, 65 (2021) (“If Spokeo and TransUnion are carried to their logical conclusion, common and longstanding private rights of action for countless laws, including copyright law, might no longer be viable in federal court.”); Rebecca Tushnet (@rtushnet), Twitter (June 28, 2021, 10:45 AM), https://twitter.com/rtushnet/status/1409523253400514560 [https://perma.cc/A3F8-YKEM] (“Logically, TransUnion v. Ramirez should apply to a number of IP cases. Practically, I expect distinctions will be invented to further reify IP harms as real while privacy harms aren’t.”). Significantly, in TransUnion itself, Justice Thomas’s dissent signaled that the majority’s concrete-harm requirement threatened copyright and patent claims, in which infringement does not require proof of monetary loss.93×93. See TransUnion, 141 S. Ct. at 2217–18 (Thomas, J., dissenting).

This Part explains how TransUnion undermines claims of trademark infringement too. Under TransUnion, to be concrete, an alleged injury must have a “close historical or common-law analogue,” although not “an exact duplicate.”94×94. Id. at 2204 (majority opinion). The legal injury of likely confusion is not concrete because trademark injury traditionally meant business harm to trademark owners.95×95. This Note does not squarely address trademark dilution, but the outcome would be similar. Dilution refers to use of a mark that reduces the mark’s ability to distinguish source. 4 McCarthy, supra note 13, § 24:67. Presumably, trademark infringement and unfair competition claims are the nearest traditional analogues. As in infringement claims, then, dilution claims would require producers to establish that they sustained a loss.

A. Trademark Injury in Law: Likelihood of Confusion

Traditionally, trademark liability did not rest on consumer confusion alone. While consumer confusion was an element of some historical and common law claims, infringement depended on harm to producers in the form of lost sales or reputational damage. To be sure, trademark claims have firm historical grounding. Trademark lawsuits date back to at least the eighteenth century, with a few cases even earlier.96×96. See 1 McCarthy, supra note 13, § 5:2; Mark P. McKenna, The Normative Foundations of Trademark Law, 82 Notre Dame L. Rev. 1839, 1850–52 (2007). The Court described this foundation in the 1879 Trade-Mark Cases,97×97. 100 U.S. 82 (1879). which declared that the trademark right was “long recognized by the common law and the chancery courts of England and of this country, and by the statutes of some of the States[,] . . . was not created by the act of Congress, and does not now depend upon it for its enforcement.”98×98. Id. at 92. The question, however, is what injury to that right meant.

In nineteenth-century England, although standards differed by court, producer harm was always necessary. In courts of law, consumer confusion was also important. Violations were “a type of fraud and deceit practiced by the defendant on the public, to the injury of the trademark owner.”99×99. 1 McCarthy, supra note 13, § 2:1; see also Jamieson & Co. v. Jamieson (1898) 15 RPC 169 (AC) at 191 (UK); Sykes v. Sykes (1824) 107 Eng. Rep. 834, 834; 3 B. & C. 541, 541. In courts of equity, however, decisions enjoined infringement without proof of consumer confusion.100×100. See, e.g., Hall v. Barrows (1863) 46 Eng. Rep. 873, 876; 4 DE G.J. & S. 150, 156; Edelsten v. Edelsten (1863) 46 Eng. Rep. 72, 78; 1 DE G.J. & S. 185, 199–200; Millington v. Fox (1838) 40 Eng. Rep. 956, 961–62; 3 My. & Cr. 338, 352. Trademark harm was viewed “primarily as an injury to the plaintiff’s property right embodied in the trademark.”101×101. 1 McCarthy, supra note 13, § 2:1; see also Frank I. Schechter, The Historical Foundations of the Law Relating to Trade-Marks 152 (1925); McKenna, supra note 96, at 1854. As one 1878 decision stressed: “The Court interferes solely for the purpose of protecting the owner of a trade or business from a fraudulent invasion of that business by somebody else. It does not interfere to prevent the world outside from being misled into anything.”102×102. Levy v. Walker (1878) 10 Ch D 436 at 448 (Eng.).

Similarly, early American courts viewed trademark infringement as an injury to producers.103×103. See, e.g., Hanover Star Milling Co. v. Metcalf, 240 U.S. 403, 412 (1916) (“The redress that is accorded in trade-mark cases is based upon the party’s right to be protected in the good-will of a trade or business.”); Trade-Mark Cases, 100 U.S. at 92 (describing the trademark right as “a property right” belonging to the trademark owner). Some scholars link early American trademark courts to the English courts of equity because of the primacy of producer harm in American trademark law. See Schechter, supra note 101, at 143–44; McKenna, supra note 96, at 1860. Consumer confusion formed part of the harm in some cases,104×104. E.g., Avery & Sons v. Meikle & Co., 81 Ky. 73, 84 (1883) (“The object of trade-mark law is to prevent one person from selling his goods as those of another, to the injury of the latter and of the public.”); Boardman v. Meriden Britannia Co., 35 Conn. 402, 414 (1868) (“[T]he violation of property in trade-marks works a two fold injury; the appropriator suffers, in failing to receive that remuneration for his labors to which he is justly entitled, and the public in being deceived, and induced to purchase articles manufactured by one man, under the belief that they are the production of another.”). but others rejected the significance of confusion and declined to find infringement in spite of it when plaintiffs did not show that their own legitimate interests had been harmed.105×105. E.g., William R. Warner & Co. v. Eli Lilly & Co., 265 U.S. 526, 528 (1924) (“The use of a similar name by another to truthfully describe his own product does not constitute a legal or moral wrong, even if its effect be to cause the public to mistake the origin or ownership of the product.”); Canal Co. v. Clark, 80 U.S. 311, 327–28 (1871) (same); United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90, 101–04 (1918) (denying infringement despite potential consumer confusion because the defendant used the plaintiff’s mark in an area remote from where the plaintiff operated); Weener v. Brayton, 25 N.E. 46, 47 (Mass. 1890) (“The jurisdiction of a court of equity . . . is founded, not upon the imposition upon the public thus practiced, but on the wrongful invasion of the right of property therein which has been acquired by others. A remedy is afforded only to the owner of the right of property in such trade-marks, on account of the injury which is thus done to him.”). Treatment was similar in the related field of unfair competition, which governed the infringement of unregistered trademarks, initially called “trade names.”106×106. See Christine Haight Farley, The Lost Unfair Competition Law, 110 Trademark Rep. 739, 751–53 (2020). Like trademark claims, these unfair competition claims focused on harm to producers.107×107. See McKenna, supra note 96, at 1863; Oliver R. Mitchell, Unfair Competition, 10 Harv. L. Rev. 275, 281 (1896). Some courts added consumer confusion to their conception of the wrong,108×108. E.g., Goldwyn Pictures Corp. v. Goldwyn, 296 F. 391, 401–02 (2d Cir. 1924); Alff v. Radam, 14 S.W. 164, 165 (Tex. 1890). while others did not.109×109. E.g., Am. Washboard Co. v. Saginaw Mfg. Co., 103 F. 281, 285 (6th Cir. 1900); Fonotipia Ltd. v. Bradley, 171 F. 951, 963 (C.C.E.D.N.Y. 1909). The Court described this variation in the common law in a 2004 trademark decision, KP Permanent Make-Up, Inc. v. Lasting Impression I, Inc.110×110. 543 U.S. 111 (2004). :

[S]ome unfair competition cases would stress that use of a term by another in conducting its trade went too far in sowing confusion, and would either enjoin the use or order the defendant to include a disclaimer. But the common law of unfair competition also tolerated some degree of confusion from a descriptive use of words contained in another person’s trademark. While these cases are consistent with taking account of the likelihood of consumer confusion as one consideration in deciding whether a use is fair, they do not stand for the proposition that an assessment of confusion alone may be dispositive.111×111. Id. at 119–20 (citations omitted).

The Court highlighted the indeterminacy of confusion at common law to support its holding that some degree of confusion did not preclude a fair use defense to trademark infringement.112×112. Id. at 121–22. In the context of fair use, then, the Court has already imported the common law into current doctrine to conclude that consumer confusion does not compel liability. After TransUnion, that common law reasoning has broader implications: trademark injury did not then and cannot now rest on likelihood of confusion alone.

When defendants’ conduct caused only consumer confusion but no loss to producers, trademark claims failed.113×113. McKenna, supra note 96, at 1866–71; see also, e.g., United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90, 101–04 (1918) (denying infringement despite potential consumer confusion because the plaintiff would suffer no damage to its current business); N.Y. & R. Cement Co. v. Coplay Cement Co., 44 F. 277, 278–79 (C.C.E.D. Pa. 1890) (rejecting an unfair competition claim despite likely consumer deception because there was no evidence of particularized damage to the plaintiff). As the Seventh Circuit explained in 1912: “That the public is deceived may be evidence of the fact that the original proprietor’s rights are being invaded. If, however, the rights of the original proprietor are in no wise interfered with, the deception of the public is no concern of a court of chancery.”114×114. Borden Ice Cream Co. v. Borden’s Condensed Milk Co., 201 F. 510, 513 (7th Cir. 1912). Consumer confusion was sometimes necessary but never sufficient as a basis for trademark liability.115×115. E.g., Am. Washboard Co. v. Saginaw Mfg. Co., 103 F. 281, 285 (6th Cir. 1900) (“[I]t is an important factor that the public are deceived, but it is only where this deception induces the public to buy the goods as those of complainant that a private right of action arises.”); Avery & Sons v. Meikle & Co., 81 Ky. 73, 100 (1883) (“While we think fraud upon the public may enter into the ground of jurisdiction, it of itself, in our judgment, is not sufficient unless some probable or possible injury to the plaintiff is shown.”); Hall v. Barrows (1863) 46 Eng. Rep. 873, 877; 4 DE G.J. & S. 150, 159 (“Imposition on the public is indeed necessary for the Plaintiff’s title, but in this way only, that it is the test of the invasion by the Defendant of the Plaintiff’s right of property . . . .”). Interpreting this history, scholars have disagreed on the overall dominance of producer interests over consumer interests, but no one disputes that producer injury was a prerequisite.116×116. Compare McKenna, supra note 96, at 1848 (“‘[T]raditional’ American trademark law was unapologetically producer-oriented.”), with 1 McCarthy, supra note 13, § 5:2 (maintaining that trademark law historically sought to protect both consumers and producers), and Robert G. Bone, Hunting Goodwill: A History of the Concept of Goodwill in Trademark Law, 86 B.U. L. Rev. 547, 560–61 (2006) (same). To have standing, trademark owners must demonstrate an actual or sufficiently imminent injury of their own.

This requirement of producer loss coheres with several principles of standing doctrine. In TransUnion, the Court evinced skepticism of conjectural, risk-based theories of harm and sought discernible, material consequences supported by direct evidence. Likelihood of confusion does not match this profile. By definition, likelihood of confusion is a matter of probability. No actual confusion is required. Rather, the multifactor test relies on proxies for confusion — on circumstantial evidence like the kind TransUnion rejected. As scholars have protested, likelihood of confusion encompasses cases of so-called “confusion in the air,” which has no material effect on consumers.117×117. Tushnet, supra note 33, at 106. This abstract confusion is not the sort that “makes a sound.” The trademark defendant may have created a risk of harm, but until that risk ripens into an actual or imminent injury, such as a mistaken purchase, the harm is not concrete. Although trademark plaintiffs seem unlikely to agree, in the Court’s estimation, their exposure to unrealized risk should give rise to celebration, not litigation.118×118. Cf. TransUnion LLC v. Ramirez, 141 S. Ct. 2190, 2211 (2021). A producer-loss requirement also fulfills the constitutional mandate that plaintiffs establish a “personal stake” in a matter.119×119. Id. at 2203. A bare likelihood-of-confusion injury implicates consumers, not trademark owners, who are the only parties to the case. It is difficult to imagine that consumers who are confused about the connection between McDental and McDonald’s would punish the restaurant as a result.

This conclusion — a conservative one in many respects — accords with the spirit of TransUnion. Justice Kavanaugh warned that standing was not an “open-ended invitation for federal courts to loosen Article III based on contemporary, evolving beliefs about what kinds of suits should be heard in federal courts.”120×120. Id. at 2204. TransUnion targeted legislative expansion, while the expansion in trademark law has largely come from modern judicial interpretation, but the same concerns about Article III boundaries should restrict both types of overgrowth. In the face of sweeping allegations of affiliation, initial-interest, and post-sale confusion, the plaintiff-loss requirement imposes discipline. By restricting standing to those claims in which producers can establish an injury, courts would depart from the current doctrinal approach, but they would do so in the service of judicial restraint.

B. Trademark Injury in Fact: Business Harm

To establish injury in fact, producers must demonstrate a loss of their own. Under TransUnion’s concrete-harm test, historically and at common law, trademark injury consisted of business harm to producers. Variously described as a “property right” or a right to “goodwill,”121×121. Courts and scholars have long debated whether trademark rights are properly classified as property rights or something else. Compare, e.g., Trade-Mark Cases, 100 U.S. 82, 92 (1879) (describing the trademark right as a “property right”), with Hanover Star Milling Co. v. Metcalf, 240 U.S. 403, 414 (1916) (describing the trademark right “as merely a protection for the good-will, and not the subject of property except in connection with an existing business”), and 1 McCarthy, supra note 13, § 2:10 (“While a trademark can be categorized as a kind of ‘property’ right, such a characterization often creates more confusion than clarity.”). This Note does not attempt to settle that debate but rather focuses on the scope of protection that trademark rights conferred, which is more consistent across cases. trademark rights protected the use of the mark in the plaintiff’s current business. Producers had no claim to the free-floating word or symbol. To warrant protection, the mark required a nexus to the products that the plaintiff was selling.122×122. See, e.g., United Drug Co. v. Theodore Rectanus Co., 248 U.S. 90, 97 (1918); Avery & Sons v. Meikle & Co., 81 Ky. 73, 86 (1883); Weston v. Ketcham, 51 How. Pr. 455, 456 (N.Y. Super. Ct. 1876); Leather Cloth Co. v. Am. Leather Cloth Co. (1863) 46 Eng. Rep. 868, 868; 4 DE G.J. & S. 137, 137. As the Supreme Judicial Court of Massachusetts held in 1890: “The right to a trade-mark cannot exist as a mere abstract right independent of and disconnected from a business. It is not property as distinct from, but only as incident to, the business.”123×123. Weener v. Brayton, 25 N.E. 46, 47 (Mass. 1890). The business-nexus requirement was pervasive and well understood. In the 1916 decision Hanover Star Milling Co. v. Metcalf,124×124. 240 U.S. 403. the Supreme Court observed that “[t]he same rule prevails generally in this country, and is recognized in the decisions of this court already cited.”125×125. Id. at 414 (citing cases). Mere use of a trademark did not invade a producer’s rights because trademark rights were limited to protecting the plaintiff’s current business. Only those uses that hurt that business were infringement.

In determining business harm, courts accepted varied forms of proof. Evidence of lost sales or reputational harm was typical.126×126. See, e.g., William R. Warner & Co. v. Eli Lilly & Co., 265 U.S. 526, 530 (1924) (“[T]he testimony discloses many instances of passing off by retail druggists of petitioner’s preparation when respondent’s preparation was called for.”); Boardman v. Meriden Britannia Co., 35 Conn. 402, 418 (1868) (recognizing damage to the plaintiff because “the finding of the committee is explicit, ‘that said respondents have sold large quantities of their spoons so put up in place of spoons manufactured by the petitioners’”); Sykes v. Sykes (1824) 107 Eng. Rep. 834, 834; 3 B. & C. 541–42 (finding infringement when “evidence supported the declaration” that “plaintiff was prevented from selling a great quantity of [its goods], and greatly injured in reputation, the articles so manufactured and sold by the defendants being greatly inferior to those manufactured by the plaintiff”). At times, courts demanded evidence that the plaintiff would incur particularized harm from the defendant’s use. For example, in cases where a defen-dant used a mark that could be mistaken for the marks of several producers, not just the plaintiff’s, courts required the plaintiff to show that its injuries were not the same as those of other producers.127×127. E.g., N.Y. & R. Cement Co. v. Coplay Cement Co., 44 F. 277, 279 (C.C.E.D. Pa. 1890) (declining to enjoin a defendant’s use of a name that several producers adopted because, even if the plaintiff suffered injury, the producers were “all injured alike”); Jamieson & Co. v. Jamieson (1898) 15 RPC 169 (AC), at 182 (UK) (“[I]t will not do for [the plaintiffs] to show that the Defendant is doing that which may be mistaken for theirs, because other Jamiesons’ goods may do that. It must be that he is passing his goods off as theirs, which is a very difficult matter to prove under these circumstances.”). When no such evidence was presented, these courts denied redress.128×128. E.g., Coplay, 44 F. at 279; Jamieson, 15 RPC at 194. In other cases, courts inferred business injury from likelihood of consumer confusion.129×129. E.g., Newby v. Or. Cent. Ry. Co., 18 F. Cas. 38, 40 (C.C.D. Or. 1869) (“Any act which produces confusion or uncertainty concerning this name is well calculated to injuriously affect the identity and business of a corporation. And as a matter of fact, in some degree at least, the natural and necessary consequence of the wrongful appropriation of a corporate name, is to injure the business and rights of the corporation by destroying or confusing its identity.”); Blofeld v. Payne (1833) 110 Eng. Rep. 509, 509; 4 B. & Ad. 410, 410 (affirming an infringement ruling because “the plaintiff was entitled to some damages for the invasion of his right by the fraud of the defendants, though he did not prove that their [products] were inferior, or that he had sustained any specific damage”). Yet trademark plaintiffs cannot surmise from these decisions that likelihood of confusion generally suffices to establish business harm. For one thing, these cases are data points, not rules. Other cases deemed confusion inadequate and required more.130×130. See supra pp. 678–80. More fundamentally, the modern relevance of these cases is critically limited by historical restrictions nonexistent in trademark law today. Before the twentieth century, trademark rights were confined to protecting against source confusion between direct competitors.131×131. See 4 McCarthy, supra note 13, § 24:2; McKenna, supra note 96, at 1900. In that context, it stood to reason that if consumers were confused, they would buy the defendant’s product when they meant to buy the plaintiff’s, and the plaintiff would lose a sale. Confusion and lost sales were one and the same.132×132. Lemley & McKenna, supra note 90, at 423.

That equation was limited to source confusion between competitors, however. Other types of confusion did not translate to business harm. Early courts understood this limitation and ruled accordingly. In the 1890 case Weener v. Brayton,133×133. 25 N.E. 46 (Mass. 1890). the Supreme Judicial Court of Massachusetts rejected an infringement claim by a noncompetitor plaintiff, a union, because the defendant’s sale of cigars bearing the union’s label did not establish harm to the union, which did not market cigars.134×134. Id. at 46–47. Fraud stood for harm only when the plaintiff was “himself a manufacturer, dealer in, or owner of, the articles which were fraudulently represented . . . to be his”135×135. Id. at 48. :

In such case the fraud complained of would have a natural and inevitable tendency to lessen the sales, affect the reputation of the articles manufactured or dealt in, injure the business of the complainant, and would thus afford him a ground for relief by reason of the special and peculiar damage which he would sustain, or to which he might be exposed.136×136. Id.; accord Leather Cloth Co. v. Am. Leather Cloth Co. (1863) 46 Eng. Rep. 868, 870; 4 DE G.J. & S. 137, 141–42.

When the plaintiff did not market the goods at issue, that logic fell apart. Since the union officers and members did not market cigars, the defendant’s deceptive use of the union’s label “d[id] not show that any business which they pursue has been affected, or that they have sustained any definite loss or any injury except that which must be extremely remote and purely speculative.”137×137. Weener, 25 N.E. at 48. Nineteenth- and early-twentieth-century cases framed their skepticism of injuries to noncompetitors in terms like “remote” and “speculative” that are familiar from standing doctrine. In the 1912 case Borden Ice Cream Co. v. Borden’s Condensed Milk Co.,138×138. 201 F. 510 (7th Cir. 1912). the Seventh Circuit used this language to reject an unfair competition claim by a condensed milk manufacturer against an ice cream manufacturer. The plaintiff had argued that consumers’ confusion would lead them to buy the defendant’s ice cream instead of the plaintiff’s condensed milk, but the court deemed this prospect “too speculative and remote.”139×139. Id. at 515. Early decisions also voiced concerns about administrability and judicial restraint that resonate with the concerns limiting standing. In the 1900 case American Washboard Co. v. Saginaw Mfg. Co.,140×140. 103 F. 281 (6th Cir. 1900). the Sixth Circuit rejected an unfair competition claim by a noncompetitor, citing concerns about “open[ing] a Pandora’s box of litigation” and doubting that “courts have the power to suppress such trade” when the plaintiff suffered no injury.141×141. Id. at 286; see also N.Y. & R. Cement Co. v. Coplay Cement Co., 44 F. 277, 279 (C.C.E.D. Pa. 1890) (rejecting a trademark claim where the plaintiff did not establish particularized injury based on enforcement concerns since “[w]hen an injury is a public one it should be prosecuted as a public wrong”).

Standing restrictions echo these historical restraints and foreclose a presumption of business harm from likelihood of confusion except in traditional cases. When parties do not compete, by definition, the defendant’s products do not substitute for the plaintiff’s, so a sale by the defendant does not come directly at the expense of a sale by the plaintiff. Under modern confusion theories, plaintiffs’ harm stories rely on a novel chain of assumptions rejected at common law. For a producer to suffer concrete harm from affiliation confusion, for example, consumers would need to: (1) actually, not just likely, be confused about the plaintiff’s affiliation with the defendant’s product; (2) dislike the defendant’s product; (3) connect their gripes about the defendant’s product to the plaintiff; and (4) withhold business that they otherwise would have transacted with the plaintiff or lower their estimation of the plaintiff. As trademark scholars point out, marketing research demonstrates that this sequence of events rarely occurs.142×142. See Lemley & McKenna, supra note 90, at 429; McKenna, supra note 29, at 115; Rebecca Tushnet, Gone in Sixty Milliseconds: Trademark Law and Cognitive Science, 86 Tex. L. Rev. 507, 543 (2008). If it does, plaintiffs should proffer some evidence of the intermediate steps, not jump to a conclusion of business harm from likelihood of confusion alone. TransUnion rejects standing when the “inferences on which the argument rests are too weak.”143×143. TransUnion LLC v. Ramirez, 141 S. Ct. 2190, 2212 (2021). Outside claims of source confusion between direct competitors, likelihood of confusion proves little beyond likelihood of confusion itself.

When called on to produce evidence of harm, trademark owners tend to object on the ground that their losses are difficult to measure. It is true that trademark owners may struggle to quantify the loss to sales or reputation attributable to defendants’ conduct.144×144. E.g., 5 McCarthy, supra note 13, § 30:2.50 (“It is notoriously difficult for the owner of a trademark to prove measurable damage caused by acts of infringement.”). But see Sandra Rierson, IP Remedies After eBay: Assessing the Impact on Trademark Law, 2 Akron Intell. Prop. J. 163, 174 (2008) (“[C]ourts should not accept without question the notion that the trademark holder’s goodwill is so ethereal and intangible that damage done to it via infringement is simply incalculable.”). Yet, as TransUnion recognized, plaintiffs do not need to calculate a loss to establish that some loss has materialized: “[T]here is a significant difference between (i) an actual harm that has occurred but is not readily quantifiable, . . . and (ii) a mere risk of future harm.”145×145. TransUnion, 141 S. Ct. at 2211; accord Tushnet, supra note 24, at 638. Proof difficulties may reduce the amount or quality of the evidence that courts accept in some cases, but they do not wipe out the need for proof wholesale.146×146. See Rierson, supra note 144, at 175 (arguing that valuation difficulties are “no excuse for treating all cases of trademark infringement as though they were created equal”). An absence of evidence is not evidence of absence, but it is not evidence of anything else either. If trademark owners have incurred a loss, or are on the verge of doing so, they must have some way of knowing, and they should present that knowledge to factfinders.

One loss that is not concrete is the commonly alleged injury of lost control over reputation. Trademark owners assert that they have suffered harm because defendants’ uses of their marks have wrested control from owners of their reputations. Modern courts are receptive to this argument.147×147. E.g., Kos Pharms., Inc. v. Andrx Corp., 369 F.3d 700, 726–27 (3d Cir. 2004). As Professor Jeremy Sheff explains, however, this loss is not justiciable harm: the argument that “‘present quality is no assurance of continued quality’ . . . seems to allow a plaintiff who suffered no injury to obtain a judgment against a defendant who may (or may not) injure him someday, ignoring standing and ripeness doctrines that would seem to be directly applicable.”148×148. Sheff, supra note 92, at 799–800 (footnotes omitted) (quoting AMF Inc. v. Sleekcraft Boats, 599 F.2d 341, 353 (9th Cir. 1979)). Lost control over reputation only opens up a risk of harm. That harm is not the “sufficiently imminent and substantial” one that TransUnion requires for an injunction, let alone the actual loss required for damages.149×149. McKenna, supra note 29, at 83. Prefiguring Justice Kavanaugh’s car crash analogy, Professor Rebecca Tushnet wrote: “Like the speeder who doesn’t get into an auto accident, the defendant might have caused a risk of harm, but the risk didn’t materialize.”150×150. Tushnet, supra note 24, at 639. Lost control does not produce concrete harm until some damage is done.

In the absence of actual or sufficiently imminent business harm, the harm that trademark owners incur from others’ use of their marks is probably closer to a moral injury. Like most people, brands want recognition, and they may perceive injustice when others adopt their trademarks without giving them the credit they think they deserve.151×151. See Sheff, supra note 92, at 764, 801; see also Tushnet, supra note 24, at 638–39 (“[C]onfusion harm . . . is in large measure a normative belief, not an empirical concept. Having conceived of confusion and dilution as wrongs, trademark owners insist that harm must exist in the real world as a thing that has effects on the trademark owner and not just as a category of actionable behavior.”). Courts regularly share these beliefs. Disapproval of defendants’ alleged free riding is evident on the face of many trademark rulings and likely motivates many more.152×152. See, e.g., Fonotipia Ltd. v. Bradley, 171 F. 951, 962 (C.C.E.D.N.Y. 1909) (“It is almost as if the court should be asked to enjoin individuals from theft, upon the ground that the criminal statutes did not make the taking of the particular kind of property in question larceny . . . .”); Avery & Sons v. Meikle & Co., 81 Ky. 73, 103 (1883) (likening trademark infringement to “stealing the livery of heaven to serve the devil in”). Whether or not these complaints are sympathetic, they are not legal injury under the Lanham Act,153×153. See Dastar Corp. v. Twentieth Century Fox Film Corp., 539 U.S. 23, 29 (2003). much less concrete injury under Article III. A defendant’s gain is not in itself a brand’s loss.

IV. Assessing the Standing Defense

Taken seriously, standing doctrine has the potential to limit the excesses of trademark law by cutting off dubious claims early. This Part assesses the strengths and weaknesses of the standing defense and concludes that it is both necessary and useful.

When a plaintiff lacks standing, dismissal of its claim is constitutionally mandatory. The defense improves trademark doctrine too. The concrete-harm requirement responds directly to calls for reform by trademark scholars, who have urged courts to “force[] trademark owners to make their harm stories concrete rather than allowing them to rest on doctrinal shibboleths.”154×154. Tushnet, supra note 90, at 1372. Article III supplies that force. These myths of legal injury are precisely what TransUnion means to debunk.

In terms of efficiency, a standing defense could dispense with baseless claims faster, before litigation proceeds to the expensive merits and remedies stages. Earlier resolution would reduce litigation costs for defendants and limit chilling effects on competition and speech.155×155. See McGeveran, supra note 16, at 62–63. In the smallest cases, costs still average over a quarter of a million dollars.156×156. Id. at 62 n.59. The fact-intensive likelihood-of-confusion test is especially costly to litigate.157×157. Id. at 70. Even defenses such as fair use that intend to short-circuit the test regularly fail to produce quick resolutions in practice.158×158. Id. at 62. A standing defense holds the promise of reducing burdens on defendants and those fearing suit.

To be sure, the extent to which a standing defense would limit costs is unclear. As courts scrutinize claims more closely at the justiciability stage, the standing analysis begins to look more like the merits analysis, and the costs from the latter are liable to bleed into the former.159×159. See Richard H. Fallon Jr., The Linkage Between Justiciability and Remedies — And Their Connections to Substantive Rights, 92 Va. L. Rev. 633, 639 (2006) (“The penetration of merits judgments into justiciability determinations prominently occurs in standing analysis.”). Even so, there is reason to be optimistic. In recent lawsuits involving the analogous false advertising provision of the Lanham Act, multiple courts have rejected standing on motions to dismiss because plaintiffs failed to allege specific facts of harm.160×160. E.g., Allbirds, Inc. v. Giesswein Walkwaren AG, No. 19-cv-05638, 2020 WL 6826487, at *4 (N.D. Cal. June 4, 2020); ThermoLife Int’l LLC v. Am. Fitness Wholesalers LLC, No. CV-18-04189, 2019 WL 3840988, at *4 (D. Ariz. Aug. 15, 2019); Me. Springs, LLC v. Nestlé Waters N. Am., Inc., No. 14-cv-00321, 2015 WL 1241571, at *6 (D. Me. Mar. 18, 2015). This theory has succeeded since courts have already interpreted the false advertising provision, unlike the trademark provisions, to require harm.161×161. See Tushnet, supra note 90, at 1344–45; see also id. at 1360 (arguing that the trademark provisions should likewise be interpreted to incorporate a materiality requirement). After TransUnion, swift resolution on similar theories is plausible in the trademark context too.

A standing defense would also improve outcomes by preventing litigation from proceeding to the point that harm becomes difficult to challenge. Early resolution is of particular concern in light of the 2020 Trademark Modernization Act162×162. Pub. L. No. 116-260, §§ 221–228, 134 Stat. 1182, 2200–10 (2020) (to be codified in scattered sections of 15 U.S.C.). (TMA). In the TMA, Congress made it easier for plaintiffs to obtain injunctions by instituting a presumption of irreparable harm from likely confusion.163×163. See 15 U.S.C.A. § 1116(a) (West). At the same time, Congress made it more difficult for courts to test the validity of the injuries that trademark owners allege. In the past, some federal courts had not applied this presumption and had instead called on plaintiffs to prove injury or an imminent threat.164×164. E.g., Ferring Pharms., Inc. v. Watson Pharms., Inc., 765 F.3d 205, 217 (3d Cir. 2014). In so doing, these courts appended something of the harm requirement that the merits inquiry lacks, albeit only at the remedies stage.165×165. Tushnet, supra note 24, at 628. Today, once plaintiffs establish likelihood of confusion — or, for preliminary relief, a likelihood of a likelihood of confusion — an injunction usually issues.166×166. 5 McCarthy, supra note 13, § 30:2.50. Thus, a conclusion of harm follows from a probabilistic assessment of a misunderstanding whose damage is itself speculative. Standing may be one of the best chances that defendants now have to dispute harm before it becomes a foregone conclusion.

Despite these prospects, the standing defense is not a perfect solution. It pushes courts to constitutionalize questions that would better be resolved by statutory reinterpretation or reform.167×167. See Lemley & McKenna, supra note 90, at 449–50 (arguing that courts could incorporate a materiality requirement when interpreting the Lanham Act without statutory reform); Tushnet, supra note 90, at 1366–67 (same). Moreover, because the defense is a constitutional one, it applies only where Article III governs and so does not explicitly prevent trademark owners from litigating their claims in state courts. Although many states have adopted justiciability doctrines mirroring those in Article III, many others have not.168×168. See Helen Hershkoff, State Courts and the “Passive Virtues”: Rethinking the Judicial Function, 114 Harv. L. Rev. 1833, 1836–38 (2001). A shift to state courts would mark a drastic change in jurisdiction, since most trademark plaintiffs today bring Lanham Act claims in federal court,169×169. 6 McCarthy, supra note 13, § 32:1. but it would not improve their resolution in any predictable way. Given federal courts’ greater experience with the Lanham Act, there is some reason to fear that trademark law would further deteriorate.

Even in the federal courts, the success of a standing defense is far from certain. Modern courts are accustomed to treating likelihood of confusion as justiciable harm. If courts were to reverse that position now, several decades of trademark infringement rulings would seem wrongly decided. That said, in recent years, the Court has been willing to upend familiar trademark doctrine under constitutional scrutiny. In Matal v. Tam170×170. 137 S. Ct. 1744 (2017). and Iancu v. Brunetti,171×171. 139 S. Ct. 2294 (2019). the Court invalidated statutory bars to registering disparaging, immoral, and scandalous trademarks on First Amendment grounds.172×172. Id. at 2302; Tam, 137 S. Ct. at 1751. TransUnion is a watershed decision, so it is unsurprising that it would have dramatic consequences. But it would also be unsurprising if courts did not acknowledge them, since TransUnion bills itself as a mere application of Article III and longstanding precedent. Courts could apply the flexible “historical or common-law analogue but not exact duplicate” test for concrete harm to avoid sweeping change to current practice, by judging claims of likely confusion to be close enough.173×173. See Fallon, supra note 159, at 637 (“[C]ourts, and especially the Supreme Court, decide cases by seeking what they regard as an acceptable overall alignment of doctrines involving justiciability, substantive rights, and available remedies.”).

Still, if this textualist Court really means what it says, then trademark law requires an overhaul. The legal injury of likelihood of confusion bears none of the hallmarks and boasts none of the traditional pedigree of a bona fide injury in fact. Without proof of business harm, trademark claims must fail. In the words of Justice Kavanaugh, “No concrete harm, no standing.”174×174. TransUnion LLC v. Ramirez, 141 S. Ct. 2190, 2200, 2214 (2021).