Unit.StartTD2M209 A Book Body enter text 2025-09-25 337 Part II. Federal Dilution Law 2 enter text Chap09 Trademark Dilution - Federal, State, and International Law, Second Edition td-book 0 main0009 Chapter 9. The Meaning of Dilution Under Federal Dilution Law 1DocumentChapter 9. The Meaning of Dilution Under Federal Dilution LawThis chapter is current through May 2025.David WelkowitzWhittier Law SchoolDisclaimer: The opinions expressed here are solely those of the author and should not be construed as representing the opinions, directly or indirectly, of the U.S. Copyright Office.I.Introduction339II.The TDRA Definition and Traditional Categories ofDilution340A.The Federal Definition of Dilution3401.The Current Statutory Definition3402.The Original FTDA Definition of Dilution342B.The Legislative History of the TDRA Definition ofDilution3431.The 1988 Dilution Bill3432.The Federal Trademark Dilution Act of 19953443.The Trademark Dilution Revision Act of 2006345C.The Categories and Mechanism of Dilution3461.The Categories of Dilution3462.The Mechanism of Dilution3483.“Actual” Dilution vs. “Likelihood” of Dilution349D.Determining Likelihood of Dilution by Blurring3511.The Elements of Blurring Under the TDRA: Distinctiveness, Association, and Likelihood ofHarm3512.Analysis of the TDRA Definition and the Statutory Guiding Factors3543.Tests for Likelihood of Dilution by Blurring Priorto the TDRA Are of Limited Use After the TDRA3584.Judicial Applications of the TDRA Definition andFactors3605.The Importance of Similarity and its Role in theAnalysis of Blurring3736.Summary: Judicial Application of the Factors ofBlurring377E.Determining Dilution by Tarnishment3781.Defining Tarnishment3782.Conduct Constituting Tarnishment: JudicialEvaluations380a.Sexually Oriented and Other UnwholesomeUses of a Famous Mark380i.The TDRA Definitional Requirements381ii.Presumption of Tarnishment in SexuallyOriented Cases381iii.Unwholesome Uses of Famous Marks inGeneral (Including Sexually OrientedUses): Case Law Favors Plaintiffs384b.Poor Quality Goods or Services386c.Subjecting the Mark to Ridicule or Criticism390F.Dilution by Confusion3961.Discussion and Commentary3962.Confusion Usually Viewed as Tantamount toDilution398G.Dilution by Genericide400H.Dilution and Internet Domain Names4021.“Cybersquatting” and the Challenge forDilution4022.Development of Per Se Dilution Test for InternetCases4043.The TDRA Undermines the Rationale for aPer Se Test4094-A. Cybersquatting and Dilution After the TDRA4.Domain Names and Critical Commentary4105.Descriptive Uses of Famous Marks in DomainNames4146.Use of Marks in Metatags4157.Hyperlinks and Tarnishment416I.Contributory Dilution417III.The Required Level of Similarity to DemonstrateDilution419A.The Requisite Level of Similarity419B.Illustrating Similarity: Examples from the Case Law422C.Trade Dress and Similarity425IV.The Need for a Trademark Use428V.Proving Dilution: Case Law and Suggested Approaches433A.Introduction433B.Case Law Discussing Evidence of Dilution434C.Dilution Surveys: Characteristics and Pitfalls4401.Designing a Survey: Determining What toMeasure4402.Designing and Introducing a Survey: SomeLitigation Pitfalls443VI.Dilution Protection for Product Configurations: A Difficult Dilution Problem445A.The Issues445B.Tests for Dilution Do Not Limit Protection for Product Configurations4471.Product Configurations and the Test for Dilution by Blurring4472.Tarnishment of Product Configurations448C.Functionality, Fame, and Distinctiveness as Gatekeepers Between Dilution and Patent4491.Functionality as Barrier to “Dilution Patent”4492.Fame and Distinctiveness as Gatekeepers ofDilution Protection450a.Demonstrating Secondary Meaning in a Product Configuration451b.A “Famous” Product Configuration Mark453D.The Possible Clash Between the TDRA and the PatentClause454[Author Note: As indicated in many of the citing references in this and other chapters, the Federal Circuit reversed a jury verdict (and the district court’s decision to uphold that verdict) in favor of Apple, Inc. and against Samsung Electronics, Inc., on Apple’s claims of trade dress dilution involving its iPhone. However, the Federal Circuit did not discuss the merits of the dilution claims, as its reversal was on the grounds that the trade dress in question was functional. Because there are so few cases discussing trade dress dilution under the TDRA, I have chosen to keep the discussions of the district court opinion for any guidance they may provide. However, readers should take care to note that the precedential value of the decision is inhibited by the reversal. (The Supreme Court later reversed the Federal Circuit’s patent holding on the issue of whether damages should be awarded for the whole product or just a component part.)]I. IntroductionThe meaning of dilution is obviously a central issue for any claim made under the Federal Trademark Dilution Act (FTDA), and the amended version, the Trademark Dilution Revision Act (TDRA). As discussed earlier in this text, dilution is a difficult term to define with great precision. Indeed, it has been called an “elusive” concept by courts.The seminal article by Frank Schechter referred to dilution as “preserv[ing] the uniqueness of a trademark.” Schechter further described the injury caused by dilution as “the gradual whittling away or dispersion of the identity and hold upon the public mind of the mark or name by its use upon non-competing goods.” His concern was protecting those marks that had a unique reputation with the public, which associated those marks with an identifiable level of quality. His description of the “whittling away” of the mark’s selling power remains the most consistent feature of dilution theory. However, the task of defining dilution more precisely first fell to state and federal courts, interpreting state statutes that simply provide a claim when one shows “likelihood of dilution” or, more precisely, “likelihood of dilution of the distinctive quality” of the mark, without defining the term “dilution.” Unlike those state laws, from the outset the FTDA contained a definition of dilution, although the TDRA significantly amended the original definition. However, the mere existence of a definition has not made the application of the statute a simple process. Courts did not agree on how the original FTDA definition translated into a judicially usable test. And it remains to be seen whether the TDRA’s definition will provide additional clarity.This chapter explores the use (and nonuse and perhaps even misuse) of Schechter’s theory by the FTDA and the TDRA to create a cause of action for dilution. Our exploration also includes the tests developed by courts and statutory drafters to apply the dilution concept to actual cases.II. The TDRA Definition and TraditionalCategories of DilutionA.The Federal Definition of Dilution1.The Current Statutory DefinitionSection 43(c)(1) of the Lanham Act gives a cause of action to the owner of a famous mark against a person who uses a mark that is likely to cause “dilution by blurring” or “dilution by tarnishment.” These two categories of dilution are defined in Sections 43(c)(2)(B) and (C) as follows:(B) For purposes of paragraph (1), “dilution by blurring” is association arising from the similarity between a mark or trade name and a famous mark that impairs the distinctiveness of the famous mark. In determining whether a mark or trade name is likely to cause dilution by blurring, the court may consider all relevant factors, including the following:(i) The degree of similarity between the mark or trade name and the famous mark.(ii) The degree of inherent or acquired distinctiveness of the famous mark.(iii) The extent to which the owner of the famous mark is engaging in substantially exclusive use of the mark.(iv) The degree of recognition of the famous mark.(v) Whether the user of the mark or trade name intended to create an association with the famous mark.(vi) Any actual association between the mark or trade name and the famous mark.(C) For purposes of paragraph (1), “dilution by tarnishment” is association arising from the similarity between a mark or trade name and a famous mark that harms the reputation of the famous mark.In addition, Section 43(c)(1) amplifies these definitions by permitting dilution claims regardless of the presence or absence of competition between the parties, and regardless of the absence of confusion between the famous mark and the second user’s mark. Furthermore, Section 43(c)(1) makes clear that no showing of “economic injury” is needed in order to obtain an injunction against a likelihood of dilution.The TDRA clearly recognizes only two forms of dilution—blurring and tarnishment. By doing so, it apparently rejects other forms of dilution, such as genericide. Whether the TDRA truly adopts Schechter’s conception of dilution is another matter. Schechter did not seem to view dilution as a balancing of factors: his premise appeared to be that any use of another’s coined, distinctive mark would cause dilution. The TDRA does not contemplate that broad a concept of dilution. Schechter also assumed that dilution would apply to cases of non-competing goods—presumably, he believed that ordinary infringement would cover competing-goods situations. And Schechter did not address the issue of tarnishment. On the other hand, Schechter’s concept was that other uses lessen the uniqueness and marketing power (the distinctiveness, in other words) of a mark. The TDRA definition incorporates those ideas in the definition of blurring and the factors used to analyze it.2.The Original FTDA Definition of DilutionAs enacted in 1995, the FTDA contained a very different definition of dilution than the current version. The original definition was as follows:The term “dilution” means the lessening of the capacity of a famous mark to identify and distinguish goods or services, regardless of the presence or absence of—(1) competition between the owner of the famous mark and other parties, or(2) likelihood of confusion, mistake, or deception.On its face, this definition leaned in the direction of Schechter’s “whittling away” concept, though it did not literally adopt it. Apart from its inclusion of competing-goods cases, which Schechter seemed not to include, this definition went beyond the idea of a mark losing its marketing power, because the mark was now less unique, to the idea of a mark losing its capacity to act as a trademark altogether. The statute also incorporated the concept of injury to the distinctiveness of a mark, but in an awkward fashion. Originally, Section 43(c)(1) permitted an injunction against second users of famous marks if the use “cause[d] dilution of the distinctive quality of the mark.” Substituting the original definition of dilution in this sentence created a certain amount of tension between the definition, which focuses on the mark’s source-identifying power, and the operative language, which focused on the mark’s distinctiveness (which is, of course, related to its source-identifying ability).Obviously, this definition does not give as much guidance as the current definition; it lacks any guiding factors, for example, which are included in the current definition of blurring. Moreover, by focusing on the diminution of the mark’s source-identifying power, the definition arguably did not include a claim for tarnishment, which does not necessarily attack the capacity of a mark to act as a mark. On the other hand, this definition more clearly contemplated a possible claim for genericide (that is, turning a famous mark into a generic one), although the only court to confront the issue indicated otherwise.B.The Legislative History of the TDRA Definition of Dilution1.The 1988 Dilution BillThe FTDA, which was passed in 1995, was nearly added to the Lanham Act in 1988, as part of the Trademark Law Revision Act of 1988. The 1988 dilution bill (indeed, the entire 1988 reform) had its origins in a 1987 proposal by the Trademark Review Commission of the United States Trademark Association to reform the nation’s trademark laws. A review of those proposals is instructive for interpreting the FTDA.The definition of dilution proposed by the Review Commission is essentially the same as the one in the FTDA. However, the Review Commission did not intend that tarnishment of a trademark be a part of the cause of action for dilution. The Review Commission viewed tarnishment as a “separate form of legal wrong,” and proposed an amendment to Section 43(a) of the Lanham Act to remedy that problem. The original bill introduced in Congress (S. 1883) contained both the dilution and tarnishment provisions. The bill also did not specifically address the issue of genericide (the destruction of a mark’s source-identifying ability, rendering it generic) and at least one witness believed that the bill did not cover this problem.The bill that finally emerged from the Senate committee, and, shortly thereafter, from the Senate, contained some significant changes in the definition of dilution from the original bill. Most important, the separate tarnishment section had been eliminated. The committee report contains this explanation:It is the committee’s contention that tarnishment and disparagement, … should continue to be decided on a case-by-case basis, and that the amendments it made to the legislation regarding [this] issue[] should not be regarded as either limiting or extending applicable decisional law.Thus, it appears that the Senate did not intend to include tarnishment in the definition of dilution. That would have left trademark-disparagement claims to be decided under a likelihood-of-confusion rationale.In addition, the definition of dilution contained in the bill that passed the Senate is not exactly the same as the USTA Review Commission proposal (or the original FTDA). The definition of dilution in S. 1883 was as follows:The term “dilution” means the material reduction of the distinctive quality of a famous mark through use of the mark by another person, regardless of [competition or absence of confusion].Unlike the USTA proposal, which focused on protecting source-identifying properties, the S. 1883 definition clearly focuses on “the distinctive quality” of the famous mark. Although this difference probably has little effect on a possible tarnishment claim, it could affect a “blurring” claim.2.The Federal Trademark Dilution Act of 1995The 1988 dilution provisions ultimately failed, having been deleted from the final bill in the House-Senate conference committee. They were revived in 1995 as a stand-alone amendment to the Lanham Act, which became the Federal Trademark Dilution Act of 1995. Although the definition of dilution in the 1995 version of the FTDA does not differ materially from the definition in the 1987 USTA proposal, the legislative intent apparently had changed. The House committee report on H.R. 1295, which became the FTDA, states that “[t]he purpose of H.R. 1295 is to protect famous trademarks from subsequent uses that blur the distinctiveness of the mark or tarnish or disparage it, even in the absence of a likelihood of confusion.” Another portion of the report reiterates that conclusion:The definition [of dilution] is designed to encompass all forms of dilution recognized by the courts, including dilution by blurring, by tarnishment and disparagement, and by diminishment.At least in the view of the House committee, the FTDA definition, though essentially unchanged from the USTA’s 1987 proposal (and not vastly different from the 1988 Senate bill), encompassed all recognized forms of dilution (perhaps even genericide, or “diminishment”). When Congress passed the Trademark Amendments Act of 1999, it reiterated its view that dilution under the 1995 version of the FTDA encompassed both blurring and tarnishment.3.The Trademark Dilution Revision Act of 2006As seen in this and other chapters, Congress’s vision of a uniformly interpreted federal dilution law proved as elusive as a definition of dilution. But it took the Supreme Court to push Congress to change the statute. Following the Supreme Court’s 2003 decision in Moseley v. V Secret Catalogue, Inc., where the Court interpreted the FTDA to require a showing of actual dilution and not just likelihood of dilution, the International Trademark Association (INTA) appointed a committee to propose revisions to the statute. This committee made a comprehensive proposal to revise the FTDA. This proposal was introduced in the House of Representatives as the Trademark Dilution Revision Act of 2005. Among the proposed revisions was a division of dilution into two categories, dilution by blurring and dilution by tarnishment. The proposed revision included definitions of those categories, as well as a list of factors to use when analyzing dilution by blurring. The proposal also reversed the Moseley decision by allowing relief upon a showing of “likelihood” of dilution by blurring or tarnishment. An amended version of the original proposal quickly passed the House. A year later, the Senate passed a somewhat altered version of the House proposal, but without changing the House-passed definitions of dilution. Finally, in September 2006, the House acceded to the Senate version and enacted the Trademark Dilution Revision Act of 2006 (TDRA). Section 43(c)(2) of the TDRA retains the two categories of dilution and the suggested factors contained in the original House bill. It also contains definitions of the two categories of dilution which are quoted in Section II.A.1. above.The creation of two specific categories of dilution appears to reflect a problem with the original definition of dilution that was pointed out by the Supreme Court in Moseley: that the definition may not have included tarnishment, despite Congress’s desire to do so. The amended statute includes a definition of tarnishment and makes it clear that tarnishment is actionable as dilution.C.The Categories and Mechanism of Dilution1.The Categories of DilutionThe TDRA recognizes just two categories of dilution: blurring and tarnishment. Unlike the 1964 Model State Trademark Bill and the original FTDA, the TDRA defines each category separately. The definition of blurring seems to be fairly consistent with the general pre-TDRA understanding of the category. It aims at uses that lessen the distinctiveness of a famous mark. Furthermore, one of the factors listed under the definition of blurring—whether the plaintiff was the “exclusive use[r] of the mark”—reinforces the link to Schechter’s idea of dilution as harm to uniqueness.Case law prior to the TDRA discussed other possible forms of dilution, although only one—genericide—could actually be categorized by name. The others were hinted at, but in a manner so vague as to make nomenclature difficult. Whether any of these other forms of dilution are actionable under the TDRA is questionable. Genericide normally refers to a use of a mark as a generic term, rather than as a mark. The definition of blurring requires that the second use conjure up an “association” with the famous mark; use of the mark as a generic term may not do that. In addition, the definition requires that the association “impair[] the distinctiveness” of the mark. One might fit a total loss of source identification into that definition, although it is certainly a less-than-perfect fit. Any “other” forms of dilution would have to fit within the definitions of blurring or tarnishment in order to be actionable under federal law.2.The Mechanism of DilutionDilution has been variously described in this text as “whittling away” at the distinctiveness of the mark (blurring), and as an attack on the reputation, and thus marketing value, of the mark (tarnishment). However, the reason that the concept is “elusive” may be that these descriptions mask what dilution really is about.One major category of dilution is blurring. “Blurring” is a very indefinite way to categorize dilution. A dictionary definition of “blur” is “[t]o lessen the perception of; to dim.” This may be an adequate description of dilution by blurring (“dimming” of the unique source-identifying properties of a mark, or lessening the perception of a mark as a source identifier), but it does not indicate how that would occur. Does any second use of a famous mark tend to “dim” its unique source-identifying properties? If not, how are those uses that cause blurring separated from those that do not? The TDRA attempts to do this with six guiding factors. But, as discussed later in this chapter, these factors are, at best, indirect evidence of possible dilution.Another way to describe dilution by blurring is a process by which a mark comes to signify more than one source, where it previously signified a unique source. This process is implicitly recognized by the TDRA in the factor of the degree of exclusivity of use by the famous mark owner. But this, too, does not easily allow us to separate diluting from non-diluting uses.One proposed method is to employ the economic idea of increased search cost. Dilution is said to result when the use of a second mark causes the consumer to have to search his or her mind harder to determine which user a particular use of the mark refers to. This approach is appealing because, in theory, one might measure the increased “mental search time,” thus giving empirical evidence of dilution. However, the search cost concept also has been subject to some trenchant criticism.The TDRA definition recognizes the problem, and defines dilution as a process that diminishes the distinctiveness of a mark and that results from an association between a famous mark and a second mark.The category of dilution by tarnishment gives a bit more guidance because it reasonably describes a set of circumstances that can be measured by facts—whether association with an unsavory trait or product is causing (or is likely to cause) harm to the “aura” of the mark. The primary problem with tarnishment is defining the associations that are deemed sufficiently “unsavory.” The TDRA appears to eschew such an attempt, defining tarnishment in terms of the ultimate effect on the reputation of the famous mark. Even so, measuring the likely effect of one or several potentially tarnishing uses on a mark is an uncertain matter.3.“Actual” Dilution vs. “Likelihood” of DilutionOne approach to separating diluting uses from non-diluting uses of a mark would be to require the plaintiff to show that the second use is actually harming the distinctiveness (or reputation) of the plaintiff’s mark. Such a showing would insure that dilution protection is not being predicated on a mere possibility of dilution. However, requiring actual dilution under the TDRA would require overcoming its statutory language and history.The earliest state laws, most of which were patterned after either the Massachusetts law or the 1964 Model State Trademark Bill, permitted an injunction based on a “likelihood of dilution.” Similarly, the TDRA permits an injunction if a second mark is “likely to cause dilution by blurring or dilution by tarnishment.” (The TDRA adds that no showing “of actual economic injury” is required to obtain an injunction.) Moreover, the “likely to cause” language was drafted specifically to overturn the Supreme Court’s interpretation of the original FTDA that actual dilution was required. Thus, it is difficult to see how one could require actual dilution as a prerequisite to liability under the TDRA.The issue resolved by the Supreme Court centered on the original language of Section 43(c)(1), which stated:The owner of a famous mark shall be entitled … to an injunction against another person’s commercial use in commerce of a mark or trade name, if such use … causes dilution of the distinctive quality of the mark … .Specifically, the debate focused on the “causes dilution” language, which, it should be noted, had since been copied by a majority of states in enacting their dilution laws. The courts of appeals split on whether this language meant that actual dilution was necessary. On one side were courts that rejected the actual dilution argument, citing the difficulties of proof that would result and the perceived intent of Congress to import the then-prevalent likelihood-of-dilution standard from the states. Other circuits adopted the actual dilution requirement, citing the plain meaning of the statutory language, and the specter of overprotecting trademarks with a looser standard.The Supreme Court resolved the split, in 2003, in Moseley v. V Secret Catalogue, Inc., holding unanimously that actual dilution was required. The Court concluded that the “text [of the FTDA] unambiguously requires a showing of actual dilution, rather than a likelihood of dilution.” The Moseley decision made proving dilution claims difficult and, three years after it was decided, the TDRA replaced the “causes dilution” language with “likely to cause dilution.” Although the TDRA resolves the issue as a matter of federal law, many states still have statutes containing the “causes dilution” language. Federal courts applying those state laws appear to be using the actual-dilution standard.The most sensible circumstance under the TDRA in which it may be appropriate to require actual dilution is a close case where the facts do not point clearly to a likelihood of dilution (particularly if the use has existed for a while). One might expect evidence of actual dilution in some such cases. This would not change the standard; the absence of actual dilution, under the circumstances, would simply be a factor indicating that dilution is unlikely to occur.D.Determining Likelihood of Dilution by Blurring1.The Elements of Blurring Under the TDRA: Distinctiveness, Association, and Likelihood of HarmThe “elusive” nature of dilution—particularly dilution by blurring—has made it difficult for courts to devise an easily articulable test for dilution. Much of the problem may be traced to two sources: (1) statutory ambiguities, and (2) a concern about overprotection of trademarks. But it also reflects a philosophical uneasiness with the implications of dilution protection. The successive attempts in the original FTDA and the TDRA to define dilution can be seen as efforts to overcome that philosophical skepticism by reassuring courts that dilution is cabined in a principled and definable manner.The current federal definition of dilution by blurring “is association arising from the similarity between a mark or trade name and a famous mark that impairs the distinctiveness of the famous mark.” This definition focuses on three elements: association, similarity, and distinctiveness. The end result of dilution, according to this definition, is harm to the famous mark’s distinctiveness. This harm must be the result of an “association” between the famous mark and the second user’s mark. That leads to the following critical points:•“Likelihood” means more than a possibility of dilution.•Dilution requires more than mere association.Beginning with the first point, on the surface this is very much the same as is the case in an ordinary infringement situation. Likelihood of confusion is a larger quantum than mere possibility. In other words, to return to dilution, a plausible narrative of dilution is not sufficient to impose liability. Although this should be obvious from the language of Section 43(c)(1), it is an important point to emphasize because the usual description of dilution may lead courts to find likely dilution when only a possibility exists. Dilution is generally described as an incremental process—the so-called “death of a thousand cuts.” Thus, it may be tempting to see virtually any similar use that triggers an association as “likely” to lead to dilution. But that is not how the statute is written.As to the second point, it is equally important to understand that the definition of dilution requires that the association cause harm to the distinctiveness of the mark. Thus, it requires more than just an association. From a practical standpoint, it means that the factors do not contain a direct means of measuring likely dilution, because none of them directly measures any kind of harm to the distinctiveness of the mark.Section 43(c)(1) makes clear that dilution does not depend on demonstrating likely economic harm to the famous mark owner; dilution may occur “regardless of the presence or absence of … actual economic injury.” This seems to be less in line with Schechter’s original idea that the marketing value of a trademark was the primary asset worth protecting. However, a loss of distinctiveness presumably would lead to a loss of marketing power—and economic injury.Finally, if one contrasts the TDRA definition of blurring with the original FTDA definition of dilution, one sees an interesting difference. The original definition focused on harm to the source identifying power of a famous mark, whereas the TDRA focuses on a loss of “distinctiveness.” While not identical, a loss of distinctiveness would cause a loss of source-identifying power, so the difference probably will not alter the outcome of many blurring cases. However, the TDRA goes further and ties the loss of distinctiveness to a process that involves association and similarity. This begs an important question. What do we mean by “distinctiveness”? Does it mean (as Schechter posited) uniqueness, or something else? And what is the relationship between an association and loss of distinctiveness by the famous mark—in other words, what level of association is relevant and what other factors must we consider when deciding whether dilution is likely to occur?The Supreme Court addressed this issue prior to the TDRA in Moseley v. V Secret Catalogue, Inc. There, the Court stated that “at least where the marks at issue are not identical,” the mere fact that the second use leads to a “mental association” between the second mark and the famous mark is not enough to demonstrate blurring. Using the fact pattern of the Ringling case (where the state of Utah used the slogan “greatest snow on earth”) as an example, the Court went on to say:[E]ven though Utah drivers may be reminded of the circus when they see a license plate referring to the “greatest snow on earth,” it by no means follows that they will associate “the greatest show on earth” with skiing or snow sports, or associate it less strongly or exclusively with the circus.Thus, the Court appeared to view blurring as a situation where, after exposure to the second use, when the consumer subsequently sees the famous mark, he or she thinks of the second user’s product or service (or at least is less likely to think of the famous mark owner’s products or services). However, the Court did not give any indication of how one might measure such a change in perception, beyond pointing to surveys. Although the Court’s holding dealt with the existence of actual dilution (the standard required under the pre-TDRA statute), its framework is equally applicable to likelihood of dilution. By “mental association,” one assumes that the Court was referring to the second mark calling to mind the famous mark. The Court’s point was that such calling to mind does not necessarily affect how one subsequently views the famous mark. More particularly, it does not mean that one who has seen the second mark will now view the famous mark as referring to anything other than the famous mark owner. If the famous mark still points uniquely to its owner, then, regardless of whether the second mark reminds us of the famous mark, the distinctiveness of the famous mark has not been diminished.This understanding is consistent with the TDRA definition of dilution by blurring. The definition does not just refer to association between the famous mark and a second mark. It refers to an association that harms the famous mark’s distinctiveness. Thus, an association, while a necessary element of dilution, is not by itself sufficient. The association must cause (or be likely to cause) a change in consumers’ perception of the famous mark.However, that still leaves one to consider how to separate those uses that are likely to lead to a loss of distinctiveness from those that are not. If the second mark does not even call to mind the famous mark—if there is no “association” between them—then no dilution will occur. But for those marks that do conjure up a famous mark, there is the problem of separating the diluting ones from the non-diluting ones. In the background, we should keep in mind the often-expressed concern that making proof of dilution too easy will overprotect trademarks (leading perhaps to the mark being a “property right in gross”). Separating the diluting from the non-diluting requires a closer look at the relevant factors that may indicate likely dilution and at the courts’ implementation of the statutory definition.2.Analysis of the TDRA Definition and the Statutory Guiding FactorsAs amended by the TDRA, the federal dilution statute now contains a separate definition of “dilution by blurring”:“dilution by blurring” is association arising from the similarity between a mark or trade name and a famous mark that impairs the distinctiveness of the famous mark.In addition to the basic definition, the TDRA states that courts may “consider all relevant factors” and then lists six non-exclusive factors to assist courts in determining likelihood of dilution by blurring:(i) The degree of similarity between the mark or trade name and the famous mark.(ii) The degree of inherent or acquired distinctiveness of the famous mark.(iii) The extent to which the owner of the famous mark is engaging in substantially exclusive use of the mark.(iv) The degree of recognition of the famous mark.(v) Whether the user of the mark or trade name intended to create an association with the famous mark.(vi) Any actual association between the mark or trade name and the famous mark.By including a number of factors, and by stating that they are non-exclusive, it appears that the drafters rejected the “any similar use equals dilution” approach to the problem. They also appear to have rejected the assumption that confusion is highly relevant to dilution, as neither confusion nor similarity of products is a listed factor. Nothing prevents a court from considering confusion as a relevant factor. However, given the prevalence of judicial uses of confusion in dilution cases prior to the TDRA, its omission from the statutory list must be considered significant.If the goal of the TDRA was to simplify the analysis and/or make the distinction between diluting and non-diluting uses easier to discern, then it is unclear whether the Act will prove to be successful. The statute makes no effort to prioritize the factors, leaving their application to a case-by-case analysis highly dependent on context. Although this is consistent with the framework of confusion analysis, it makes advance prediction by potential users of their possible liability especially difficult. Perhaps, as more cases are decided, a de facto hierarchy of factors will appear.The first factor, similarity, is obviously highly relevant to the analysis. Without sufficient similarity, there will be no association between the two marks and, hence, no possible dilution. But what degree of similarity should be relevant, how do we measure similarity, and what should be the effect of a lack of similarity on the analysis? Although similarity has long played a central role in the analysis of blurring, recent cases have made its role less clear.At first glance, the role of the next factor, degree of inherent or acquired distinctiveness, is uncertain. After all, a famous mark must be highly distinctive, at least in acquired distinctiveness. But, if this factor is a measure of uniqueness—particularly if the famous mark is highly inherently (or conceptually) distinctive—then its role is clearer. If a famous mark is highly unique, particularly if it is very strong conceptually (i.e., it is unusual and neither descriptive nor even suggestive) then another’s use of that mark will be more likely to distract those who view the famous mark and will be more likely to cause consumers to recall both the famous mark owner and the second mark owner when they hear or see the mark.The third factor—substantially exclusive use—primarily indicates the mark’s level of uniqueness. A less than unique mark might be famous, but it would be less likely to be diluted by another similar mark, especially a non-competing one. A consumer would be less likely to think of a particular offending user upon seeing the non-unique mark when there are already other users of the mark to distract the consumer’s thought process. Furthermore, a consumer may be less likely to make an association between the offending mark and any particular owner of the mark, unless the context points to a particular owner. If there is no association, then there can be no dilution.Although the fourth factor is relevant, the degree of recognition seems redundant as a stand-alone factor, in light of the threshold requirement that a mark be famous to be eligible for dilution protection. Moreover, the relevance of this factor is tempered by an empirical question: Is a mark more likely to be diluted if it is well recognized than if it is less well recognized? On the other hand, a highly recognized mark may be more likely to trigger an initial association between the offending mark and the famous mark.The factor of intent to create an association appears to be a measure of “predatory intent.” In a confusion context, this would probably be seen as a measure of bad faith. However, an intent to create an association may not be a sign of bad faith. A similar factor is used in the analysis of likelihood of confusion, and its presence tends to be dispositive. However, its relevance to confusion has been questioned, so it is fair to ask whether it actually is relevant to likelihood of dilution. Instinctively, one might think that if a person actively attempts to create an association between the two marks, then it is more likely that there will be an association. However, this factor must be used with care, especially since many situations where this factor would favor the plaintiff involve a parody or a play on words, where courts are noticeably more reluctant to impose liability.The sixth factor—actual association by consumers between the two marks—was not expressly a part of prior tests, but it is a logical one. If dilution arises from an association between the two marks, then evidence of such association should be relevant. But its inclusion as a factor also indicates that it is not dispositive. That further suggests that dilution is more than a mental association, that it requires some change in the consumer’s perception of the famous mark (at the least, a perception that there is more than one user of the mark).Probably the most common evidence of an actual association would be a survey, asking, for example, what one thinks of when confronted with the offending mark. Even when such evidence exists, however, its probative value should not be overstated. A mental association is a first step in the dilution process, but it is not the end result. Dilution occurs when the mental association changes the consumer’s perception of the famous mark—i.e., when the diluting use distracts the consumer from the force of the famous mark. A second use that “calls to mind” the famous mark may have no effect on the famous mark’s unique ability to sell an image, or its ability to conjure up a unique association between the mark and a single owner. And none of the factors attempts to measure this end result of dilution.The other side of the actual-association factor is how to deal with its absence. Because the applicable standard is likelihood of dilution, not actual dilution, courts do not insist on evidence of actual association. Nevertheless, its absence should not be ignored. In particular, where one would expect some evidence of association, its absence should be seen as some indication that there is no association or that any association is too weak to find a likelihood of dilution.The key to the proper application of the new test is to remember that the factors do not define blurring; the definition defines blurring. Thus, to succeed on a blurring claim, a famous mark owner must show that the defendant’s use is likely to cause an “association … that impairs the distinctiveness of the famous mark.” Although the factors certainly are intended to serve as helpful guides, the test must be more than a mechanical application of the factors. That said, the factors obviously will play an important role in determining blurring. The keys to the efficacy of this new test will be how much weight is assigned to each factor, how courts determine the intent of the second user, and how the courts relate the factors back to the definition. Whether this new test actually demands more than a showing of a high degree of fame and sufficient similarity is difficult to say. When applying this test, courts should keep in mind that, although the TDRA overturned the requirement of actual dilution, it still requires a showing that dilution is likely, not just that it is possible. One might infer a possibility from fame and similarity, but does that make dilution likely? Perhaps this is where a combination of factors (iii) and (vi) will prove most useful.3.Tests for Likelihood of Dilution by Blurring Prior to the TDRA Are of Limited Use After the TDRAThe most prominent feature of the TDRA was its clear statement that likelihood of dilution was the operative test under the amended statute. This was intended to overrule the Supreme Court’s interpretation in Moseley of the language in the original FTDA, which the Court interpreted to require actual dilution in order to demonstrate a right to relief. However, in the seven years prior to the Supreme Court’s decision, several circuits applied a likelihood-of-dilution standard to FTDA cases. Because the TDRA permits courts to use “all relevant factors,” it is appropriate to inquire whether any of the pre-TDRA case law can be of assistance after the TDRA.Before the TDRA was enacted, federal courts struggled to find a serviceable test for dilution under the FTDA. In large part, this was because the statute itself provided little guidance apart from its broad definition of dilution. Prior to the Supreme Court’s Moseley decision, courts had divided roughly into two groups when measuring blurring. One group of courts used some form of multifactor test to determine the existence of blurring. Another group of courts, led by the Seventh Circuit, used a second test, focusing on only two factors: the similarity of the marks and the renown of the senior (famous) mark. Although these two tests predominated in the cases where some test was used, there were a number of cases in which courts evaluated blurring claims with little indication of how they reached their conclusion.It seems clear that the TDRA did not endorse what might be called the “holistic” approach of the latter group of courts. By adding factors to the statutory definition of dilution by blurring, the drafters of the statute obviously intended that courts adopt a more structured approach to dilution claims, somewhat akin to that used in a likelihood-of-confusion analysis. It is also questionable whether the Seventh Circuit’s approach, which relied almost exclusively on fame and similarity, is an appropriate test under the TDRA. Although both the degree of renown and degree of similarity are factors under the TDRA, it does not appear that they were intended to be the exclusive measures of likelihood of dilution. Moreover, recent cases have held that the TDRA does not require any particular level of similarity as a prerequisite to a finding of likely dilution. Therefore, relying on these two factors alone would extend dilution beyond its intended boundaries.Two other pre-TDRA tests relied on additional factors. One test, originally developed for use in cases applying New York law, relied on the following six factors: (1) similarity of the marks; (2) similarity of the products; (3) sophistication of the consumers; (4) predatory intent; (5) renown of the senior mark; and (6) renown of the junior mark. Quite obviously, this test looks very much like a test for likelihood of confusion. But the classic dilution case is quite different from a likelihood of confusion case—it assumes that the consumer is not confused, but that the mark has come to designate more than one source of goods or services, thus losing its unique identity. Although a number of early cases cited this test, it was subject to a great deal of criticism, even from the circuit that developed it. Moreover, its efficacy as a guide for the TDRA is doubtful. The drafters of the TDRA were aware of this test and the omission of similarity of products, sophistication of customers (which is a strange factor in dilution, given its non-reliance on confusion), and renown of the junior mark indicates that those factors were not considered especially relevant in the analysis.The other significant multifactor test was also developed by the Second Circuit. The court acknowledged the usefulness of several of its earlier factors but wished to include these additional factors:Those [additional factors] include actual confusion and likelihood of confusion, shared consumers and geographic isolation, the adjectival quality of the junior use, and the interrelated factors of duration of the junior use, harm to the junior user, and delay by the senior in bringing the action.But these factors, if anything, increased the link between dilution and confusion, factors which the TDRA conspicuously omitted. Thus, although this test was adopted by several other circuits before the Supreme Court’s decision in Moseley, it has not been used since the TDRA.Although some of the factors in the Mead Data and Nabisco cases are arguably relevant post-TDRA, courts should be careful when importing factors of which the TDRA drafters were aware and which were clearly omitted from the statute. This is particularly true of factors pointing directly at confusion.4.Judicial Applications of the TDRA Definition and FactorsAlthough some of the post-TDRA cases have come to their conclusions with rather cursory explanations, other cases have made an effort to use the TDRA factors.In an early post-TDRA case, a court found against a dilution claim, primarily on the basis of a lack of similarity and a concomitant lack of mental association. Although all of the factors except predatory intent favored the plaintiff, the court was careful to note that the distinctive feature of the plaintiff’s mark was the combination of “Century” and “21,” not “Century” alone. The lack of overall similarity, which is a product of the fact that the core of each mark, “Century,” is a common word and the fact that overall the marks were somewhat different, meant that no dilution was likely.The first appellate case to apply the TDRA’s blurring factors was Louis Vuitton Malletier, S.A. v. Haute Diggity Dog, LLC. As the court recognized, this case did not arise in the classic blurring context of a noncompetitor using the famous mark (or a nearly identical version) in an ordinary trademark manner for the defendant’s goods. Rather, the defendant used a parody of the plaintiff’s marks to denominate its goods. The defendant produced a line of dog toys that it called “Chewy Vuiton,” an obvious play on the “Louis Vuitton” mark used on high fashion accessories.As the Louis Vuitton court understood, applying the TDRA factors in a parody case requires careful attention to the underlying definition of dilution. A parody must create an association between itself and the famous mark in order to be successful. Therefore, in a parody case several factors that might ordinarily implicate blurring will probably point in the plaintiff’s favor: degree of similarity, intent to create association, and actual association. Moreover, because a successful parody depends on public recognition of the object of the parody, the other factors (degree of distinctiveness, exclusive use, and degree of recognition) are likely to point to the plaintiff as well. However, as the court noted, the existence of an association (even one intentionally created) does not necessarily mean that the distinctiveness of the famous mark has been, or is likely to be, harmed. And it is the likely damage to distinctiveness, not simply association, which is the key to the definition of dilution.Thus, the court stated that “because the famous mark is particularly strong and distinctive, it becomes more likely that a parody will not impair the distinctiveness of the mark.” In short, “parody is relevant to the overall question of” dilution by blurring. Because the defendant communicated both an association with the mark and a disassociation, the court found that the use would not impair the Louis Vuitton mark’s distinctiveness.Louis Vuitton also failed in a dilution claim against a competitor in relation to its Monogram Multicolore mark. The court cited two reasons for ruling against Louis Vuitton: a lack of sufficient similarity and a lack of evidence of actual dilution. The court’s reasoning as to dilution is instructive, even under a likelihood-of-dilution standard. As the court noted, Louis Vuitton’s evidence showed, at most, that consumers made a mental association between the defendant’s mark and the Monogram Multicolore mark. That, the court said, was not sufficient—it cited the Supreme Court’s statement that mental association alone does not show dilution. However, there was no evidence that the association caused any harm to the distinctiveness of the Monogram Multicolore mark. Not only were consumers able to distinguish the two, and properly attribute any detrimental associations to the defendant, but the evidence showed that the popularity of Louis Vuitton products had increased during the time that the defendant’s mark had been in use.Another case that denied a blurring claim was Nissan Motor Co., Ltd. v. Nissan Computer Corp., a long-running suit that began years before the TDRA. Nissan Motor Company objected to Nissan Computer Corporation’s registration of nissan.com and nissan.net domain names for its business. Nissan Motor claimed both infringement and dilution. After determining that the Nissan mark was famous, and had been famous when Nissan Computer registered the domain names in 1991, the court turned to the analysis of blurring. In an interesting and possibly significant comment, the court stated that the six-factor TDRA test “sets a different standard” than the pre-Moseley likelihood-of-dilution standard used by many circuits. The court also recognized that the blurring inquiry is distinct from the determination of fame. The court proceeded to review each of the TDRA factors. The nissan.net and nissan.com domain names were found to be identical to the Nissan mark while Nissan Computer Corp. logo was “similar.” On the other hand, Nissan was found not inherently distinctive—it is “a common Jewish surname,” the “seventh month of the Jewish calendar. … Moreover, to a Japanese speaker, ‘Nissan’ is an easily-recognized abbreviation of the generic words ‘Japanese Industrial Companies, Ltd.’” Interestingly, the court only found “a medium-high degree of acquired distinctiveness”—one would expect more of a famous mark, especially one that is not inherently distinctive. The factor of exclusive use favored neither side because many other companies, including some “larger companies” and some with whom Nissan Motor had a name-sharing agreement, used Nissan in their names. The intent factor was also found to be neutral.The crucial factor was evidence of association between Nissan Computer and Nissan Motor. The court analyzed this factor in light of the evolution of Internet usage. In the court’s view, the widespread use of search engines made it unlikely that users automatically assume that “trademark.com” would lead to the proper Web site for the owner of a famous mark. Most significantly, the court noted that Nissan Motor failed to produce any evidence of association or blurring. Although the TDRA does not require actual dilution, the court indicated that without some evidence, an inference of likelihood of blurring (or of association) would be speculative. It chided Nissan Motor for failing to conduct even an association survey, and further indicated that it would have been desirable to have had a survey that might have demonstrated that consumers were likely to have their association with Nissan Motor blurred after seeing the Nissan Computer marks. Notably, too, the court did not presume dilution simply because the nissan.com domain name was “identical” to the Nissan mark.Finally, the court discussed two other factors not listed in the TDRA that bolstered its conclusion of no dilution. First, the passage of the Anticybersquatting Consumer Protection Act in 1999 relieved the pressure on courts to use dilution as a means of combating cybersquatting. Second, Nissan Motor’s “delay in bringing suit,” while not a bar, indicated that Nissan Motor did not believe that dilution was likely to occur. Thus, the court denied relief under a dilution-by-blurring theory.And for those who thought that the case of Victoria’s Secret against Victor’s Little Secret ended with the Supreme Court’s decision in 2003, think again. The Court did not order dismissal of the case; it remanded to the Sixth Circuit for application of the actual dilution standard. Although it took more than three years for the TDRA to reverse the Court’s decision about actual dilution, the Sixth Circuit did not act on the remand until after the TDRA was enacted. It then sent the case back to the district court, which applied the now-applicable TDRA. Applying the six factors, the court found that all of them favored Victoria’s Secret. Nevertheless, the court found no likelihood of dilution by blurring. It reasoned that there was no evidence that the distinctiveness of the Victoria’s Secret mark had been harmed (or was likely to be harmed); it noted that the association evidence—primarily the affidavit of an Army colonel who brought the Moseleys’ store to the attention of Victoria’s Secret—did not show any harm to the distinctiveness of the mark. Rather, it indicated possible harm to the mark’s reputation, which would not be blurring, but tarnishment. The court cited with agreement the Supreme Court’s admonishment that a mental association does not necessarily mean that dilution is going to occur. Thus, the court did not simply mechanically apply the factors—it related them carefully to the definition of dilution by blurring.Another district court, granting a motion to dismiss a dilution claim, also cited the fact that association does not necessarily lead to blurring.A dilution claim by jeans manufacturer Levi Strauss against Abercrombie & Fitch originally failed as well, but that decision was reversed and remanded for reconsideration. As in the Louis Vuitton case discussed above, the district court found the marks of the two parties (consisting of stitching patterns on the back pocket of the jeans) to be insufficiently similar to be likely to cause dilution. In addition, the court found that other blurring factors did not support Levi’s claim. In particular, it found that the defendant did not intend to create an association with Levi’s mark, and it found Levi’s survey evidence unsatisfactory as evidence of actual association and dilution. The court’s analysis of the factor of substantially exclusive use was interesting. It pointed out that, although the particular stitching used by Levi’s may be unique, there were many other brands using back-pocket stitching, and that this had not diluted the uniqueness of the public’s recognition of the Levi’s stitching. Thus, the court found this factor to be “neutral”; it seemed to be saying (although it did not say this in fact) that Levi’s argument came close to arguing that the problem was back-pocket stitching in general, not the particular stitching.However, on appeal, the Ninth Circuit reversed. Although, in most respects, the court of appeals did not disturb the district court’s findings, it decided to reexamine whether the “identical or nearly identical” standard it had promulgated for determining sufficient similarity under the FTDA survived the TDRA. It decided that the standard no longer applied to dilution by blurring under the TDRA and remanded for further consideration using the proper standard.In Gap, Inc. v. G.A.P. Adventures, Inc., the court rejected a dilution claim, despite the fact that it found in favor of the plaintiff on a confusion claim, and despite the fact that the court found the blurring factors to favor the plaintiff. The court stated that “‘[t]he fact that people “associate” the accused mark with the famous mark does not itself prove the likelihood of dilution by blurring.’” It then held that the plaintiff “has not proved any injury to its trademarks,” and that an injunction against infringement would obviate any likelihood of dilution in the future.The G.A.P. Adventures rationale was cited by the Trademark Trial and Appeal Board in Rolex Watch U.S.A., Inc. v. AFP Imaging Corp., where the Board rejected an opposition by Rolex to the registration of Roll-X for “x-ray tables for medical and dental use.” After establishing that the Rolex mark was famous, the Board turned to dilution. Echoing G.A.P. Adventures, it stated:“we must determine not only whether there is an ‘association’ arising from the similarity of the marks, but whether such association is likely to ‘impair’ the distinctiveness of the famous mark.”Here, the Board found that the second through fourth factors (degree of distinctiveness, exclusivity of use, and degree of recognition of the famous mark) favored Rolex. However, the other three factors (similarity, intent, actual association) did not. (There was a survey introduced to show actual association, but the Board found that its recognition percentage was relatively low.) Similarity may be the most significant factor in this case. Although they are pronounced the same, the Board found “a different appearance, meaning and commercial impression” between the marks.In Virginia Polytechnic Institute & State University v. Hokie Real Estate, Inc., the court refused to issue a preliminary injunction on blurring grounds on behalf of Virginia Tech (which uses the name Hokies for its athletic teams and alumni) against a real estate agency using Hokie in its name. In particular, the court emphasized that the third, fifth, and sixth factors (exclusivity, intent to create association, and actual association) weighed against dilution. With regard to the intent factor, the court separated a use designating the university from a use that designated one as an alumnus. And it found no evidence of actual association.Courts sometimes dismiss allegations of dilution on the pleadings for reasons apart from lack of fame. In Mintz v. Subaru of America, Inc., the Ninth Circuit upheld the dismissal of a complaint including allegations of dilution. The court stated that the clear dissimilarity between the marks precluded a dilution claim, which is based on an association arising from similarity.In Swatch S.A. v. Beehive Wholesale, LLC, the court granted summary judgment against a blurring claim based on its findings of minimal similarity, lack of intent to cause association, and lack of evidence of actual association.In Hershey Co. v. Promotion in Motion, Inc., the court dismissed a dilution claim after a bench trial. Analyzing the factors, it found that defendant’s SWISSKISS mark would not be deemed sufficiently similar to the Hershey’s Kisses mark to be likely to cause dilution. It also found that Hershey did not demonstrate substantially exclusive use of the Kisses mark on chocolate. It found in favor of Hershey on the factor of public recognition, but in favor of defendant on the factor of intent to create association. As to actual association, the court was skeptical of Hershey’s survey methodology, and also believed that the level of association found (about one-third of respondents) was somewhat low compared with other cases. Notably, the court acknowledged that association itself was not sufficient—the association must harm the distinctiveness of the mark.At both the district court level and the court of appeals level, Hugunin v. Land O’ Lakes, Inc., a suit by a fishing tackle company against a dairy company, is noteworthy for its discussion of association and dilution. The defendant counterclaimed for dilution. On a motion for summary judgment, while finding issues of fact, especially regarding the factors of intent to create association and actual association, the district court specifically stated that a jury was not required to find blurring even if it found association; association does not necessarily mean that the distinctiveness of the famous mark is likely to be impaired. On appeal, the Seventh Circuit went a step further and dismissed the dilution claim, asserting: [W]e can’t see how the [defendant] company could be hurt by the use of the same name by a seller just of fishing tackle. The products of the two companies are too different, and the sale of fishing tackle is not so humble a business as the sale of hot dogs by street vendors. And so it is beyond unlikely that someone dissatisfied with LAND O LAKES fishing tackle would take revenge on the dairy company by not buying any of its products, or that a customer would have difficulty identifying LandO’ Lakes’ dairy product because he had seen the LAND O LAKES mark used on Hugunin’s fishing tackle.In Louis Vuitton Malletier, S.A. v. My Other Bag, Inc., the court also dismissed a dilution claim, finding that mere association was insufficient to support dilution liability. Defendant, My Other Bag (MOB), made a line of canvas tote bags. On one side of each bag were the words “My Other Bag . . .”, and on the other side of the bag was a drawing of a handbag recognizably made by a famous maker such as Louis Vuitton. The court dismissed Louis Vuitton’s dilution claim, stating:Significantly, it is not enough to show—as Louis Vuitton indisputably can—that members of the public are likely to “associate” the defendant’s mark with the plaintiff’s mark (or that the defendant promotes such association).The court went on to apply the Fourth Circuit’s opinion in Haute Diggity Dog, another case involving a parodic use of a Louis Vuitton mark, insisting that Louis Vuitton make a demonstration (which it failed to do) that MOB’s use of a drawing of its bag would harm the distinctiveness of the Louis Vuitton mark.These cases correctly apply the concept of dilution set forth by the Supreme Court in its Moseley opinion, which is supported by the language of Section 43(c). While association is certainly necessary, it is not a sufficient condition for dilution.However, it is important to note that, at least at the pleading stage, it may be difficult to demonstrate that a parody should be deemed, as a matter of law, to eliminate the possibility of dilution. Thus, in DC Comics v. Mad Engine, Inc., the defendant sold T-shirts with a Superman-like shield (five-sided yellow and red shield), with the word “Dad” in place of the usual “S.” The defendant claimed it was a parody and thus not diluting, but the court held that, at the pleading stage, it could not determine whether consumers would view defendant’s use as a parody, and therefore denied a motion to dismiss.Starbucks Corp. v. Wolfe’s Borough Coffee, Inc. was a very long-running case (it made three trips to the court of appeals) that began before the TDRA and continued after it. This was a suit by Starbucks against a small coffee retailer that sells coffee under the names Mister Charbucks and Charbucks Blend. The latest opinion (Starbucks VI) would seem to have ended the case in favor of the defendant and, combined with the previous opinion (Starbucks IV), gives us some additional guidance as to application of the blurring factors. In a significant aspect of Starbucks IV, the court held that “substantial similarity” was no longer a threshold requirement of dilution; to the court, the inclusion of “degree of similarity” as a statutory factor meant that no threshold was intended by the drafters. (As in the Levi Strauss case, the court of appeals in Starbucks IV remanded after reconsidering the proper standard for evaluating similarity under the TDRA.) On the other hand, Starbucks IV upheld the district court’s conclusion that similarity should be analyzed by comparing the marks as presented in the marketplace. Thus, the significant differences in the packaging between Starbucks coffee and the defendant’s Mister Charbucks coffee, and the use of “Mister” or “Blend” with the “Charbucks” name, were relevant to the analysis of similarity. The court of appeals upheld the finding of only “minimal” similarity, but held that minimal similarity did not disqualify a case from dilution. The court of appeals also commented on two other factors: intent to cause an association and evidence of actual association. As to the former, the court held that bad faith was not a necessary element of intent; an intent to associate, even if not in bad faith, would favor a finding of dilution.As to the actual association factor, the district court thought that actual association in the case favored the defendant, based on the evidence submitted. However, the court of appeals indicated that the plaintiff’s survey, which showed that 30.5 percent of respondents thought of Starbucks first when told of the name Charbucks, indicated the possibility of actual association. The court of appeals also pointed out that the district court’s emphasis on a lack of confusion was error because dilution does not require confusion. Finally, the court stated that the factor of intent to cause association did not require “bad faith.” The case was remanded for a second time.On remand, the district court continued to adhere to its original conclusion of no likelihood of dilution. As in the G.A.P. Adventures and Rolex cases discussed above, the court emphasized that there must be a likelihood of harm to the distinctiveness of the famous mark—association is not enough. As to the factors, the court noted that although Starbucks and Charbucks may be similar, the similarity was offset by the defendant’s use of “Mister,” “Mr.,” or “Blend” before Charbucks, resulting in a “minimal degree” of similarity. On the issue of intent to associate, the court found in favor of the plaintiff, based on the defendant’s own admissions. Although the plaintiff submitted an association survey (discussed in the Second Circuit’s opinion), the court found that its 30.5 percent association only minimally favored the plaintiff. Finally, the court alluded to the general principle that dilution statutes should be applied with care, so as not to overly protect trademarks. Thus, although the factors of distinctiveness, recognition, exclusivity, and intent clearly favored Starbucks, the lack of similarity and the weak evidence of association led to the conclusion of no likelihood of blurring.In Starbucks VI, the Second Circuit upheld the findings of the district court. The court’s opinion focused on three of the blurring factors: degree of similarity (factor 1), intent to create association (factor 5), and actual association (factor 6). These are also the factors that relate to both marks, not just to the famous mark. Upholding the district court’s finding of minimal similarity, the court reiterated its conclusion from Starbucks IV that the addition of “Mister” and “blend” to the Charbucks mark justified the district court’s conclusion that the marks were only minimally similar. As to the intent factor, the court rejected Starbucks’ argument that an intent to create association raises a presumption of dilution. Consistent with its view that each factor enjoys independent status, the court noted that a presumption “would effectively merge the intent to associate and the actual association factors, by making the former determinative of the latter.” Finally, the court examined Starbucks’ evidence of actual association, a telephone survey. The court first upheld the district court’s determination that the survey was flawed because it failed to present respondents with the Charbucks mark together with the “Mister” or “blend” additions, with which it was consistently associated on the defendant’s website and in its store. Next, it upheld the finding that the survey only showed “minimal” association. Because of the presentation issues, and because when asked what company or store might sell a product called Charbucks only a small percentage of the respondents said Starbucks, the court of appeals affirmed. However, it noted that if Starbucks had offered a properly conducted survey into evidence, it might have ruled otherwise on this factor.The court of appeals also conducted its own balancing. It found that the three factors relating to the strength of the mark and exclusivity of use (which were largely uncontested in this case) were less compelling than the three factors bearing directly on the definition of dilution—similarity, intent to associate, and actual association. It stated that the first three may demonstrate “susceptibility to dilution,” but do not show the crucial elements of association arising from similarity. As the court noted, “[u]ltimately what tips the balance in this case is that Starbucks bore the burden of showing that it was entitled to injunctive relief.” That, plus the weak evidence of actual association, allowed the court to affirm. But it suggests that the case was a close one, and that proper evidence of association could have changed the outcome. One missing aspect of the discussion was the final part of the definition of blurring—association “that impairs the distinctiveness of the famous mark.” It appears that the court assumed that sufficient similarity combined with sufficient association would mean that harm to the famous mark’s distinctiveness would be inevitable. But that would seem to make the last phrase almost superfluous, since the definition already requires an “association arising from the similarity between” the two marks. The difficulty of separating the blurring factors and the factors of confusion is illustrated in New York City Triathlon, LLC v. NYC Triathlon Club, Inc. As it began the analysis of blurring, the court stated that “[the blurring] factors have been discussed at length in connection with Plaintiffs’ §43(a) claim.” The court proceeded to reiterate much of the discussion of likelihood of confusion, though applying it to the blurring factors. The two marks differed only slightly (the word “club” in the defendant’s mark), the plaintiff had used the mark exclusively for 10 years, the court seemingly assumed “a high degree of acquired distinctiveness” from the fact of secondary meaning, there was an apparent bad-faith change of name by the defendant, and an incident of confusion, which the court interpreted as demonstrating “actual association,” even though it could be deemed isolated. But there was no effort to determine whether the mark’s association with the plaintiff was harmed—which is the definition of dilution—when in fact the existence of confusion suggests that the association with the plaintiff remained strong.A contrasting view of the blurring analysis is evident in Miss Universe, L.P., LLLP v. Villegas, a suit by the owner of the Miss USA mark against the operator of a Miss Asia USA pageant. In contrast to the New York City Triathlon case, here the court began by emphasizing the dissimilarities between the two marks—namely, the “Asia” in Miss Asia USA. Regarding distinctiveness, this court emphasized a relative lack of inherent distinctiveness, a matter essentially ignored in the following year in New York City Triathlon. The parties agreed that the plaintiff had substantially exclusive use of the mark and that the mark was highly recognized. Regarding intent, the court carefully distinguished “bad faith” from “intent to associate.” Here, it found no intent on the defendant’s part to cause an association. Finally, the only evidence of actual association, some emails asking about a possible association, was dismissed by the court as “isolated inquiries.” Overall, the lack of evidence convinced the court that there was no likelihood of dilution.Other cases have found likelihood of dilution under the TDRA.One of the earlier cases to find dilution after a review of the factors is Nike, Inc. v. Nikepal International, Inc. This was a suit by the clothing manufacturer Nike against Nikepal, which “provides services and products to analytical, environmental, and scientific laboratories.” After finding the Nike mark was famous before Nikepal’s first use of its mark, the court turned to dilution by blurring, analyzing each of the six factors in turn. It found that each of the factors favored Nike: the marks were deemed “nearly identical,” Nike is an inherently distinctive mark, Nike’s use of the Nike mark was “substantially exclusive,” the Nike mark is widely recognized, the defendant intended to create an association between the marks, and, based on a survey, the court found that people exhibited a high degree of association between Nikepal and Nike. Thus, the court concluded that dilution was “likely.” However, the court never related its analysis to the definition of dilution by blurring—harm to the distinctiveness of the famous mark—instead relying on the factors alone.In Gianni Versace, S.P.A. v. Versace 19.69 Abbigliamento Sportivo SRL, the court found that defendant’s use of a mark containing the Versace name caused both dilution by blurring and tarnishment. In the blurring analysis, the court found all of the factors to favor plaintiff. In particular, the court found the marks to be very similar, because “Versace” was the dominant part of both marks—despite defendant’s claim that it used “Versace” with other words. It found that plaintiff’s actions to police its marks indicated exclusivity, The court further found evidence of association and knowledge of that association by defendant.In Hershey Co. v. Art Van Furniture, Inc., the court found that the defendant’s “couch wrapper” takeoff on Hershey’s candy bar wrapper was likely to dilute the Hershey packaging mark. In addition to finding exclusive use, the court found the uses highly similar and found an intent to cause an association. Interestingly, however, the court conceded that “whether such … association has actually been made is unclear.” Given the importance of the association—without it, by definition there is no dilution—this casts doubt on the court’s conclusion.In Visa International Service Ass’n v. JSL Corp., the district court found a likelihood of dilution of the plaintiff’s Visa mark by the defendant’s evisa.com Web site. The court found that four of the six factors—similarity, inherent and acquired distinctiveness, degree of recognition, and actual association (measured by a survey showing that 73 percent of respondents thought of another brand when shown the evisa mark)—favored the plaintiff. Although the court found neither evidence of intent to create an association nor evidence of exclusive use, it concluded that the plaintiff had made a sufficient showing of likelihood of dilution to warrant summary judgment.On appeal, defendant JSL claimed that its eVisa mark used the word “visa” in a manner that mirrors the actual meaning of the word, and that this precluded any blurring claim. Although it acknowledged that such a use can make a blurring claim weaker, the Ninth Circuit affirmed the conclusion of likelihood of blurring in this case. The court emphasized two key distinctions between this case and others. First, JSL was using the word eVISA as a trademark, not simply as a common English word. Second, Visa International’s Visa mark did not use the word in a manner that strongly evoked its common English meaning. The plaintiff’s use, therefore, allowed its mark to be considered very distinctive (as well as famous). The court also noted that the two marks were “effectively identical,” citing case law to the effect that identical marks lead to a presumption of dilution, and, in an important gloss on the TDRA factors, stated that empirical evidence of likely dilution is not needed. The court concluded that the defendant’s trademark use of the word was deemed likely to diminish the distinctiveness of the trademark value of the Visa word mark.Like Visa International, the case of Rolex Watch U.S.A., Inc. v. Rolex Deli Corp. involved an unauthorized use of an identical mark (Rolex). Relying largely on the identity of the marks, the court found dilution by blurring. In RE/MAX, LLC v. Shenzhen Remax Co., a magistrate judge found likely dilution by blurring, based in large part on the identity of the marks, as well as RE/MAX’s active policing of the mark, which the court found supported exclusivity of use. These factors were found to overcome a lack of intent to associate or evidence of actual association. In Juul Labs, Inc. v. Unincorporated Associations Identified in Schedule A, the court determined that using a counterfeit mark constituted dilution “as a matter of law.” In H-D U.S.A., LLC v. Square Wear, LLC, the court issued a preliminary injunction on blurring (and tarnishment) grounds based on defendant’s unauthorized sale of merchandise with plaintiff’s Harley-Davidson marks. In Hallmark Licensing LLC v. Dickens, Inc., a magistrate judge found blurring through the unauthorized distribution of outdated counter cards, though without analyzing how this could harm the distinctiveness of the mark. In Patagonia, Inc. v. TP Apparel, LLC, the court found dilution by blurring (in a default situation) where defendant used parodies of plaintiff’s Patagonia mark, without any real examination of harm to the distinctiveness of the mark (though this may be the result of accepting plaintiff’s allegations in its complaint).In Lorillard Tobacco Co. v. California Imports, Ltd., the court found dilution by blurring based on the high degree of similarity between the marks (Newport and Newprot), an inference of intent to trade on plaintiff’s reputation based on the similarity of the products, and evidence that the defendants were concerned about possible association between the two marks.In a case involving trade dress dilution, the District of Oregon found that the design of certain Skechers sneakers diluted both the adidas “Three Stripes” mark and the design of adidas’s “Stan Smith” line of sneakers. (Stan Smith, for the uninitiated, was a famous U.S. tennis player in the 1970s.) The court found that the “striking similarity” between Skechers’ uses of stripes and the adidas mark, as well as the similarities between the look of the Skechers sneaker and that of adidas, indicated that Skechers intended to create an association with the adidas marks. In addition, the court found that a confusion survey could also be used to show association between the marks. On appeal, the Ninth Circuit upheld the District Court’s ruling on dilution with a rather cursory analysis that largely relied on its analysis of confusion. It is perhaps revealing that the Ninth Circuit noted the “substantial overlap” between the statutory dilution factors and the confusion factors used in the Ninth Circuit.In another trade dress case, the Western District of Kentucky found that the green and yellow colors of John Deere farm equipment was diluted by defendant’s use of very similar colors on its farm equipment. The court found that all of the factors favored a finding of dilution. In particular, it found a high degree of similarity between the uses, that Deere was essentially the exclusive user of those colors on farm equipment, that there was a high degree of recognition of Deere’s marks, that there was an intent to create an association, and that there was evidence of actual association (the latter based on surveys showing 38 percent and 43 percent association). The court did not reassess the findings concerning the factors in light of the definition of dilution by blurring (i.e., likely harm to the distinctiveness of the mark), resting its decision on the strength of the factors analysis.In a somewhat unique case, a district court granted summary judgment to Louis Vuitton on a dilution claim against Hyundai, based on the latter’s use of a basketball with a Louis Vuitton-like mark on it in a Hyundai commercial. The court found that all six factors favored blurring. However, unlike the G.A.P. Adventures case discussed above, the court made no attempt to ascertain whether there was any real likelihood of harm to the distinctiveness of the mark. Hyundai essentially conceded that it used the Vuitton mark to raise Hyundai’s brand status. However, it is unclear how this would impair Vuitton’s distinctiveness; Hyundai was counting on this distinctiveness and continued brand awareness to raise its own image. Thus, the case appears to be one of free riding, not impairment of distinctiveness. But free riding is not one of the defined categories of dilution under the TDRA.The TTAB has upheld several post-TDRA dilution claims. The first such decision was National Pork Board v. Supreme Lobster & Seafood Co., where the TTAB found that the factors of degree of similarity (the marks were “highly similar”), degree of distinctiveness, exclusivity of use, degree of recognition, and intent to create association all favored the opposer. With regard to degree of distinctiveness, the TTAB primarily examined inherent distinctiveness, finding the mark The Other White Meat to be presumptively inherently distinctive, due to its registration without showing secondary meaning, and finding the mark to be suggestive. As to degree of recognition, the TTAB cited numerous pieces of evidence, including a recognition study that showed a very high degree of recognition—greater than almost all other slogans. With regard to intent to create an association, the TTAB was skeptical of the applicant’s testimony that it did not consider opposer’s slogan when creating its own slogan (The Other Red Meat, used for salmon), given the high degree of recognition of opposer’s slogan. On the other hand, the TTAB did not find bad faith, concluding that applicant may have believed that creating an association was permissible. Nevertheless, this factor counted in favor of dilution. Interestingly, the TTAB found the factor of actual association to be neutral, even in the face of a survey (which the TTAB found valid) that showed possible association. The TTAB took a narrow view of “actual” association, stating that such a finding was not possible where applicant had not yet used its slogan in commerce. Although the TTAB upheld the dilution claim, it did not specifically explain whether its analysis led to the conclusion that there was a likelihood of a change in the perception of The Other White Meat slogan.In Nike, Inc. v. Maher, Nike opposed an application to register Just Jesu It for clothing, based on Nike’s Just Do It mark. After finding Nike’s mark to be famous, the Board turned to dilution by blurring. The second through fourth factors (distinctiveness, exclusive use, and recognition) easily favored Nike. As to similarity, the Board noted that it had found the marks similar for confusion purposes, and that it believed the applicant’s mark would “conjure up” the Nike mark. There was no evidence of intent to create an association (mere knowledge of the Nike mark was not enough), nor any evidence of actual association (the mark was not yet in use). The TTAB then considered three other non-listed factors: the size of applicant’s business, parody, and relatedness of goods. It indicated that business size was not truly relevant to whether dilution could occur. It rejected a parody claim, noting that applicant did not intend to comment on opposer’s goods or services. And it viewed relatedness of goods as favoring dilution—thus implying that confusion may play a role in dilution. On balance, the Board held that the factors favored a finding of likely dilution.In UMG Recordings, Inc. v. Mattel, Inc., the owner of the Motown mark opposed Mattel’s application to register Motown Metal for toys. In its discussion of blurring, the TTAB found that the first four factors favored the opposer, while the last two (intent and actual association) lacked any supporting evidence. The key factors appeared to be similarity and exclusivity of use. In a manner not unlike a confusion analysis, the TTAB looked at “appearance, sound, connotation and commercial impression” in its similarity analysis. Motown Metal was found sufficiently similar to conjure up Motown. With regard to exclusivity, the major stumbling block was the frequent use of Motown to refer to the city of Detroit. The TTAB found such use “insignificant,” and concluded that the opposer had established “substantially exclusive” use of the mark. On balance, the TTAB concluded that there was a likelihood of dilution.And in Research in Motion Ltd. v. Defining Presence Marketing Group, Inc., the TTAB upheld a dilution claim by the maker of BlackBerry mobile phones against an application to register CrackBerry as a mark for a variety of services, including electronic and mobile phone related services. Regarding the dilution factors, the applicant virtually conceded the main factors, including similarity, relying instead on a parody claim. However, the Board rejected the parody claim. Because CrackBerry would be used as a trademark, it could not claim to be covered by the fair use exclusion. Moreover, because CrackBerry had become a popular nickname for a BlackBerry phone, the kind of separation between the famous mark and the parody necessary to obviate dilution was not found. Therefore, the Board found that likelihood of dilution had been shown.In each of the Nike, UMG, and Research in Motion decisions, it should be noted that the TTAB also found likelihood of confusion, which is often the case in dilution cases.Another TTAB decision, Chanel, Inc. v. Makarczyk, found dilution by blurring without ever reaching the issue of confusion. After determining that the Chanel mark was famous, the Board turned to the multifactor analysis of blurring. The application for registration was for the identical mark—Chanel—so there was no question that this favored a finding of dilution. Interestingly, however, the Board did not invoke any presumption of dilution from the identity of the marks. In light of the Board’s findings in connection with the analysis of fame, there was little question that the factors of degree of inherent or acquired distinctiveness and recognition (factors 2 and 4) would favor Chanel. As to factor 3, degree of exclusivity of use, the Board noted that the evidence showed no other users (apart from one user licensed by Chanel under stringent controls), and an aggressive policing policy on the part of Chanel. Factor five, intent to create an association, was also found to favor Chanel. Applicant’s business model evidently involved using famous luxury brands—including Chanel—as the names of his apartment buildings, and his websites indicated that the purpose was to connect his apartments with luxury represented by the brands. There was no evidence of actual association, so this factor was neutral. The Board then analyzed whether there was a likelihood of impairment of the distinctiveness of the mark. Analytically, this was proper—the definition of dilution by blurring is not mere association; it is harm to the distinctiveness of the mark resulting from the association. However, the Board’s treatment of this aspect of dilution leaves much to be desired and illustrates the difficulty of defining dilution. The Board cited a conclusory affidavit from a Chanel official, asserting that registration of another Chanel mark would dilute the brand’s distinctiveness. The Board then asserted that many luxury brands (though not Chanel) had licensed their marks for use in connection with hotels and interior design features. Based on these pieces of evidence, the Board found a likelihood of impairment, and concluded that there was a likelihood of dilution.The impairment analysis in Chanel is somewhat odd. The Board did not discuss how the fact that luxury brands license their marks for various ventures leads to a conclusion of impairment of distinctiveness. It would appear that the Board’s discussion is more relevant to a finding of likelihood of confusion than impairment of distinctiveness. (If consumers are confused about whether the applicant licensed the use of the Chanel name for his buildings, that would be evidence of the continued distinctiveness of the mark, not evidence that the mark now signifies two different sources.)In New York Yankees Partnership v. IET Products and Services, Inc., the TTAB sustained an opposition to two applications solely on the grounds of dilution by blurring (without even considering confusion). Both marks were parodic takeoffs on marks used by the New York Yankees baseball team: the “TopHat logo” (a top hat on a baseball bat surrounded by a baseball with the Yankees’ name across it), and The House That Ruth Built, a reference to the original Yankee Stadium. One of the proposed marks was a top hat similar to the Yankees’ one, but with a syringe substituting for the baseball bat, and a prohibition sign across it. The two marks are pictured below:Fig. 9-1: The Yankees’ TopHat logo and the applicant’s parody logo The other proposed mark, following the same theme, was The House That Juice Built (an obvious reference to performance-enhancing drugs). In each case, the Yankees claimed dilution by blurring; the Board never discussed possible dilution by tarnishment. After establishing the prerequisite elements of dilution—famous marks, use in commerce, and use by the applicant after the marks became famous—the Board turned to the issue of dilution. Given the fact that the point of the marks was to poke fun at the Yankees, it is not surprising that the Board found virtually all of the factors to favor the Yankees. However, certain aspects of its analysis deserve closer examination. The Board began with the TopHat logo. On the issue of similarity, the Board conceded several items of dissimilarity: the syringe, the lack of a baseball background (which the Board somewhat misdescribed as a “circle”), and the lack of the word “Yankees,” as well as the existence of a prohibition sign on the applicant’s mark. As for intent to create an association, the applicant conceded that its purpose was to create a parody, which can only succeed if there is an association.The most critical issue, however, was largely overlooked in the Board’s opinion: whether the association was likely to harm the distinctiveness of the Yankees’ marks. Here, the Board’s reasoning was unfortunate. In a prior opinion, the Board had indicated support for the Fourth Circuit’s approach to parodies as trademarks; namely, that a successful parody could negate harm to distinctiveness. However, the Board now has decided to reject this possibility:We take this opportunity to modify our prior suggestion in Research in Motion that an alleged parody should be considered as part of our dilution analysis even when parody does not provide a safe harbor for a defendant. We now choose to not consider the parody defense as part of the assessment of the dilution claim, because, as discussed supra, a mark that identifies source—which it must for registration—will not concurrently qualify for a statutory exclusion to a dilution claim. Stated another way, when an applicant’s mark is registrable, because it is being used in commerce to indicate source, such use is not a noncommercial use or fair use. Thus, given the circumstances generally presented by opposition and cancellation proceedings based on allegations of dilution, we find it virtually impossible to conceive of a situation where a parody defense to a dilution claim can succeed in a case before the Board.There are two significant problems with the Board’s statement. First, it seemingly conflates the fair use and noncommercial use exclusions with non-diluting uses. The point of the exclusions is that they prevent liability even if dilution would otherwise be likely to occur. Thus, they do not necessarily address the threshold question of likelihood of dilution. The point of the Fourth Circuit’s Haute Diggity Dog opinion was that parody may prevent dilution from occurring—it was not intended to create a “defense.” Second, even to the extent that the Board did not conflate likely dilution with exclusions from liability, its rejection of the premise of the Fourth Circuit’s opinion, combined with a lack of discussion of how the parody would harm the Yankees’ mark’s distinctiveness, ignores a crucial part of the definition of dilution by blurring. As is also the case with confusion, the formalism of the factor structure cannot trump the core inquiry, which in dilution is harm to the famous mark’s distinctiveness. As the Fourth Circuit pointed out in its Haute Diggity Dog opinion, rather than harming a famous mark’s distinctiveness, a successful parody may reinforce its distinctiveness. Indeed, the parody depends for its value on the continued distinctiveness of the mark. Thus, it would seem that for a successful parody the logical dilution claim (if any exists) is for tarnishment, not blurring.Having found blurring with respect to the TopHat logo, the Board turned to THE HOUSE THAT JUICE BUILT mark. Similar to the TopHat logo mark, the Board found all factors except actual association and exclusivity of use to favor the Yankees, with the other two being neutral. The Board then applied the same reasoning from the TopHat logo mark to THE HOUSE THAT JUICE BUILT mark and found dilution by blurring. It appears that the Board’s analysis presumes harm from association or intent to cause association. However, the Board does not cite any empirical evidence to support that conclusion. (The Board’s opinion in this regard may be contrasted with the opinion of the Southern District of New York in Louis Vuitton Malletier, S.A. v. My Other Bag, Inc., discussed above, which emphasized the need for more than just association, and which applied the Fourth Circuit’s parody analysis.) In a non-precedential opinion, the TTAB found dilution by blurring in an opposition to an intent-to-use application that incorporated the design of the Empire State Building in a logo for NYC Beer. Opposer owned registered trademarks in a design drawing of the Empire State Building and asserted that the applicant’s proposed logo was likely to cause dilution. After finding the design to be famous, the Board examined the six statutory factors. With regard to similarity, although the applicant’s logo included other design features and words, the TTAB found that it “prominently features” the Empire State Building (and applicant admitted that it was used to evoke New York City). Thus, the differences “do not serve to diminish the similarity of the building design therein to Opposer’s design mark.” Distinctiveness and level of recognition (factors two and four) favored the opposer as well—not surprising, given the finding of fame. The third factor, substantially exclusive use, also favored opposer. Unlike the New York Yankees Partnership case, there was no evidence of intent-to-cause association. And, given that the mark had not yet been used, there was no evidence of actual association. Nevertheless, the Board found sufficient evidence of likelihood of dilution. Although at the outset of its analysis the Board emphasized the need not just for association but for harm—likelihood of impairment of distinctiveness—the Board did not attempt to explain how the four factors that favored opposer led to the conclusion that harm was likely to occur. It would appear that fame and similarity were deemed sufficient, at least with a very recognizable mark that was exclusively used by one party.A somewhat similar situation occurred in Nike, Inc. v. Capital E Finance Co., also a non-precedential opinion, where the Board allowed a dilution claim by Nike against the application for the mark Just Did It (an obvious play on Nike’s “Just Do It” mark). The Board followed the six-factor analysis for blurring, finding the first four factors favored likely blurring, but found the other two factors—intent to create association and actual association—neutral (noting that the applicant had not yet used the mark). The Board then discussed the requirement that the applicant’s mark be likely to impair the distinctiveness of the opposer’s mark, but found this satisfied because “Applicant’s mark would be used in direct competition with Opposer.” It is unclear why that would necessarily harm the mark’s distinctiveness, especially in light of the Board’s later finding that there was a likelihood of confusion. One thing is clear: A play on a famous mark will not fare well before the TTAB.5.The Importance of Similarity and its Role in the Analysis of BlurringThe concept of similarity is central to an analysis of dilution. Dilution requires, in the first instance, that the second user’s mark conjure up the famous mark. If the marks are not sufficiently similar, that mental association will not occur. Moreover, it is important that the second mark triggers an association with the famous mark. It is not sufficient for the second mark to trigger an association with the type of products or services sold under the first mark.The TDRA puts similarity at the center of any evaluation of likelihood of dilution. Neither the original state laws patterned after the 1964 Model Bill, nor the original FTDA (and similarly worded state laws) said anything about the required level of similarity. Courts interpreting those statutes generally required that the two marks be “substantially similar” for dilution to occur. Under the TDRA, both the definition of blurring and tarnishment invoke similarity. Obviously, similarity will play a critical role in dilution cases under federal law. Indeed, it may end up being the critical factor in many cases; the mental association needed to demonstrate likelihood of dilution could depend on the existence of sufficient similarity. However, the TDRA has complicated the issue of similarity and its place in the analysis of dilution, particularly dilution by blurring. Prior to the TDRA, courts (analyzing both federal and state law) viewed similarity as a threshold issue; if the marks were insufficiently similar, the analysis effectively ended. Moreover, courts generally required a high level of similarity—either “substantial” similarity or a similarity such that the marks were deemed “identical or nearly identical.” However, the TDRA’s use of the concept of similarity in the definition of blurring makes its role in the analysis less clear.The TDRA defines dilution by blurring as an “association arising from the similarity between a mark or trade name and a famous mark that impairs the distinctiveness of the famous mark.” It goes on to list six factors to help courts analyze blurring, the first of which is “[t]he degree of similarity between the mark or trade name and the famous mark.” Thus, the question arises: is similarity a threshold element, as suggested by the definition, or is it merely a factor, and, if the latter, is there still a required threshold of similarity? This question was addressed by the Second and Ninth Circuits.The Second Circuit addressed the issue in Starbucks Corp. v. Wolfe’s Borough Coffee, Inc. The court first noted that the TDRA does not expressly address the required level of similarity, but also noted that the TDRA does not require “substantial” similarity in the definition of dilution by blurring. Second, the court reasoned that because the first blurring factor looks only at the “degree of similarity,” without requiring any threshold, the other five factors would be rendered immaterial if there were a required level of similarity that was not met. Thus, the court jettisoned its existing framework, which required “substantial similarity” between the marks, in favor of making similarity only one of the factors to consider.The Ninth Circuit followed a similar approach in Levi Strauss & Co. v. Abercrombie & Fitch Trading Co. Until its decision in Levi Strauss, the Ninth Circuit had required marks to be “identical or nearly identical” in order to be considered sufficiently similar for dilution purposes. Although this formulation had not proven as formidable as it might appear, it had been consistently applied, and was even referenced by the Ninth Circuit, without apparent objection, in some post-TDRA opinions. However, after analyzing those post-TDRA cases, the court concluded that it had never squarely faced the issue of the continuing viability of its identical or nearly identical test, and decided to reexamine it.Like the Second Circuit, the Ninth Circuit noted that the TDRA constituted a wholesale revision of the dilution statute, rather than a minor amendment to change the operative standard. The amended statute now contained separate definitions of dilution by blurring and dilution by tarnishment, and, most important, contained several factors to guide the analysis of dilution by blurring. As the Second Circuit noted, the first of those factors is “the degree of similarity” between the marks, with no indication of any particular threshold. Although Abercrombie argued that the legislative history of the TDRA indicated that Congress did not intend to disturb prior case law on this issue, the court declined to use legislative history, citing what it felt was the clear language of the statute and the lack of clarity in the legislative history. Therefore, the court concluded that the “identical or nearly identical” standard would no longer govern claims alleging dilution by blurring. The court did note that similarity is a “necessary predicate for dilution analysis” and that “greater degrees of similarity manifestly are more likely to support a finding of dilution.”After some uncertainty, the TTAB accepted Starbucks and Levi Strauss. In Citigroup, Inc. v. Capital City Bank Group, Inc., the Board noted the Second Circuit’s analysis in Starbucks, but decided to adhere to its requirement that the marks must be “substantially similar.” It reasoned that the weight of any factor could vary with the facts and that the Second Circuit’s fear that emphasizing similarity would deemphasize the other factors was unwarranted. However, in Nike, Inc. v. Maher, the TTAB quoted extensively from Levi Strauss and appeared to approve its reasoning (though it could have made its intent clearer), and subsequent cases made clear that the TTAB no longer requires substantial similarity. In other circuits, the issue has yet to be resolved.Although the approach of the Second and Ninth Circuits is justifiable on the face of the statute, it should be interpreted with caution. Similarity is a critical element of dilution, and it would normally be inappropriate to permit a dilution claim to succeed when there is not a high degree of similarity. Giving extra weight to this factor is not inconsistent with the apparent intent of the drafters to make dilution a limited remedy. But there is a danger that courts will interpret the Second Circuit’s standard as permitting dilution recovery even when the marks are quite dissimilar.On the other hand, even after the TDRA, courts tend to regard the use of an identical mark by defendant to be presumptively diluting. Moreover, courts have used the existence of a high degree of similarity as evidence of intent to create association.In any event, it is clear from the cases discussed in Section II.D.4. above that similarity often plays a pivotal role in the analysis of blurring and courts should not lightly find that an offending mark sufficiently “conjures up” the famous mark.6.Summary: Judicial Application of the Factors of BlurringIn the absence of significant appellate guidance, any conclusions about the importance of various factors are necessarily somewhat speculative. However, from the cases discussed in Section II.D.4. above, we may draw a few tentative conclusions about the way courts are applying the TDRA blurring factors.It is clear that similarity of the marks at issue plays a critical role in the outcome of cases. In several cases denying liability, courts have cited a lack of similarity as a primary reason. Conversely, when courts find similarity they usually find likelihood of dilution. The TTAB cases discussed in Section II.D.4. above illustrate this well—the one case rejecting a dilution claim was the one where similarity was not sufficient to favor the opposer.As is the case with an analogous factor in the conventional confusion analysis, intent to create an association tends to greatly favor the plaintiff, except where the defendant can show a clearly successful parody, such as in the Chewy Vuiton case. But, as was evident in the case of the Hershey candy wrapper, unless a court can reasonably connect the intent to some actual association (or at least likelihood of actual association), then there is no predicate for a dilution claim. Moreover, it is worth remembering that the definition requires more than association—it requires an association that is likely to cause some harm to the mark’s distinctiveness.Obviously, any mark eligible for dilution protection must be famous, so a high degree of distinctiveness and recognition should be expected. Uniqueness has been cited in some of the cases, and this can come in two different contexts. The factor of exclusive usage is a measure of uniqueness. In addition, an inherently distinctive mark is more likely to be unique, and thus affected by an association, than a non-inherently distinctive mark. At least two cases have pointed to this latter factor as being important to the decision.Ultimately, two critical questions remain unanswered. First, there is a serious question of the degree to which some of the closest cases (such as those involving New York City Triathlon, America’s Team, and a Hershey candy bar wrapper) are influenced by a finding of confusion. In several cases, courts have referred back to their discussions of confusion in their analyses of the dilution claim. Second, there is still a question whether courts can, in a principled fashion, apply a test that requires more than just a famous mark and a very similar second use. The two forms of uniqueness and a greater insistence on at least evidence of association may be important if the test is to be applied appropriately.One final issue is the standard of appellate review of a district court’s evaluation of dilution by blurring. As is the case with respect to ordinary infringement, the circuits are not uniform on this issue. The Second Circuit “review[s] the [district] court’s determination as to each factor for clear error and its balancing of those factors de novo.” The Fourth Circuit has held that “likelihood of dilution, like likelihood of confusion, is a fact-intensive inquiry, so the appropriate standard of review is clear error.” Both of these holdings mirror those circuits’ review of likelihood of confusion. Thus, one would expect most circuits to follow that template and use the standard of review applied to ordinary infringement—which is not uniform.E.Determining Dilution by Tarnishment1.Defining TarnishmentAs discussed earlier in this book, tarnishment occurs when a trademark is associated with unwholesome or shoddy products or circumstances. Very often, but not exclusively, these associations involve sex or drugs. When a famous mark is associated with such unwelcome situations, the positive associations consumers originally had with the mark may become degraded by the new association of the mark with the unsavory activity. This may diminish the mark’s value as a marketing tool and reduce its selling power among consumers.However, the TDRA definition of tarnishment only indirectly relates to the marketing value of the trademark, focusing on the reputation, not distinctiveness, of the mark.The TDRA makes clear that claims for tarnishment are included in the federal dilution statute; the original FTDA language was not so clear. Moreover, unlike any previous dilution statutes, the TDRA actually defines dilution by tarnishment as an “association arising from the similarity between a mark or trade name and a famous mark that harms the reputation of the famous mark.” Potentially, this is broader than the existing case law, because it looks only at the result—harm to the reputation of the famous mark—and not to the situations causing such harm. Again, it does not focus on the distinctiveness of the famous mark but, rather, on its message. That is, even if the reputational harm is caused by something other than an association with unsavory conduct or items, and even if it does not diminish the uniqueness (or distinctiveness) of the famous mark, the TDRA definition seems to include such harm within the definition of tarnishment. This may place greater emphasis on the exclusions from liability as the limiting factors in a tarnishment case.Unlike under the TDRA’s definition of blurring, the statute provides no guiding factors for courts to consider. Thus, the courts are left to sort out the proper way to determine whether a particular use is likely to harm the famous mark’s reputation.2.Conduct Constituting Tarnishment: Judicial EvaluationsNow that it is clear that dilution by tarnishment is actionable under the TDRA, judicial evaluations should be the most helpful guides to this category of dilution. As indicated in the discussion of dilution under state law, courts have found tarnishment to occur in several different types of situations. These situations include: (1) association of the senior mark with unwholesome activity, such as drugs or pornography; (2) association of the senior mark with goods of poor quality; and (3) subjection of the mark to ridicule (as for example, by a parody or satire). Federal courts deciding FTDA claims prior to the TDRA appear to have been using state-law tarnishment claims as a guide to federal claims. In some ways, this is consistent with the TDRA because the state laws that they were following made “injury to business reputation” actionable. However, as discussed below, the existence of specific exclusions in the TDRA may circumscribe the application of tarnishment in ways that are not applicable under state laws. In addition, the TDRA’s rather broad definition of dilution by tarnishment, which focuses on harm to reputation rather than the activity causing that harm, may in some cases broaden the scope of tarnishment. Finally, courts seeking to regularize the analysis of tarnishment under the TDRA may impose new tests not previously used in tarnishment cases under federal law.a.Sexually Oriented and Other Unwholesome Uses of a Famous MarkOne of the most common situations in which courts invoke tarnishment is when the allegedly diluting mark is used in connection with conduct deemed unwholesome, particularly sexually oriented uses. The most common examples of sexually oriented uses of famous marks are on the Internet, either as domain names or on the Web sites themselves, although other such uses have been found tarnishing as well. Sexually oriented uses have also been the subject of a separate analysis by at least one court of appeals, as discussed below in Section II.E.2.a.ii. The panoply of case law relating to sexually oriented and other unwholesome uses will be discussed in Section II.E.2.a.iii. below.i.The TDRA Definitional RequirementsUnder the TDRA definition of dilution by tarnishment, a mark owner must show that the offending use is likely to harm the reputation of the famous mark. Although courts prior to, and even post-TDRA, generally have not required mark owners to submit evidence of potential harm to the mark in order to succeed, the specific definition in the TDRA, focusing on harm to reputation, suggests that some mechanism for linking association to harm ought to be required. That a use is particularly offensive should not alter this perspective. Although the statute specifically states that no economic harm need be shown to demonstrate dilution, it may be possible to show reputational harm in noneconomic terms, such as by a survey. Moreover, when evaluating the harm, courts should note that merely conjuring up the famous mark is not sufficient, nor is it sufficient that the consumer view the offending use with disfavor; the offending use must be likely to cause harm to the reputation of the famous mark—that is, it must be likely to change the consumer’s perception of the famous mark.Most courts have taken what appears to be a holistic approach to tarnishment. That is, they do not employ any particular analytical approach, instead relying on their own sensibilities and, if available, any evidence of harm to the reputation of the famous mark. Sexually oriented and drug-related uses will tend to offend the judges’ sensibilities and often lead to findings of tarnishment. However, one appellate court has adopted a formal approach in cases involving sexually oriented uses of famous marks, which will be discussed in Section II.E.2.a.ii below.ii.Presumption of Tarnishment in Sexually Oriented CasesThe long-running litigation between Victoria’s Secret and the owners of an adult novelty store named Victor’s Little Secret has already produced the Supreme Court’s only dilution opinion. Subsequently, it produced an appellate decision announcing a new approach to the issue of tarnishment in sexually oriented situations. In V Secret Catalogue, Inc. v. Moseley, the Sixth Circuit addressed the question of how to analyze tarnishment in such situations. The court held that in cases involving “sex-related products,” where there is “a clear semantic association between” the offending mark and the famous mark, a rebuttable presumption of tarnishment arises. This presumption, which the court likened to res ipsa loquitor, shifts the burden to the defendant to “com[e] forward with evidence that there is no likelihood or probability of tarnishment.”The court grounded its creation of the presumption on three bases. First, the court pointed to Section 25 of the Restatement (Third) of Unfair Competition (and which the court erroneously referred to as the “Restatement of Trademarks 3d”), to the effect that an “‘association’ between a famous mark and lewd or bawdy sexual activity disparages and defiles the famous mark.” Second, the court cited several cases in which sexually oriented uses of famous marks had been found tarnishing. However, the court’s statement that it found “no exceptions” to the sexual-use-as-tarnishment regime may have overlooked some cases. The third, and perhaps most crucial, piece of support was the change in the law via the TDRA from actual dilution to likelihood of dilution. The court opined that Congress intended that the new standard should lower the burden of proof for the plaintiff in a dilution case, and viewed its approach as consistent with that intent.Judge Gibbons’ concurring opinion muddies the analysis a bit by insisting that the “presumption” be called an “inference”—though one that shifts the burden of producing evidence—and that it not shift the burden of persuasion to the defendant. This is consistent with the Federal Rules of Evidence concept of presumptions, under which presumptions do not shift the burden of persuasion. Because Judge Gibbons provided the deciding vote on the panel, her statement ought to govern the outcome.Judge Moore issued a vigorous dissent. She disagreed with the majority’s conclusion that Congress intended to shift the burden of production of evidence—she viewed Congress as simply lessening the plaintiff’s burden by reducing the necessary showing from actual dilution to likelihood of dilution. She further pointed out that the Supreme Court and post-TDRA courts have agreed that mental association alone does not constitute dilution, whether measured by an actual-dilution or likelihood-of-dilution standard. The critical element, she maintained, is not whether the new mark is mentally associated with the famous one, but whether that association is likely to change one’s perception of the famous mark. Moreover, in the very case before the Sixth Circuit, the Supreme Court had pointed out that the critical piece of evidence of dilution—the affidavit of an Army colonel—contained no evidence that his view of the Victoria’s Secret mark had been affected by the Moseleys’ store. Instead, he harbored bad feelings toward the Moseleys. In short, she disputed the majority’s assertion that the new law permitted shifting the burden of production or persuasion to the defendant.Judge Moore also pointed out that the majority ignored the context of the famous mark. Victoria’s Secret flaunts sexuality as part of its advertising, which may decrease the likelihood that the use of a similar mark in a sexual context would tarnish the reputation of the mark.A district court in the Sixth Circuit has relied on this presumption to find dilution by tarnishment in a 2023 case.Commentary on the presumption. To Judge Moore’s cogent dissent one can add a few observations about the majority’s opinion. First, it is difficult to see how a defendant would rebut the presumption or inference. The most obvious method would be a survey, but a survey would pose practical problems. What sort of question could one pose to an audience that would distinguish between an association between the marks and a change in perceived reputation (especially when dilution is premised on often subtle changes in perception)? Asking outright whether the second mark changes one’s view of the famous mark might work, but it would seem likely that many respondents would not be able to separate the two concepts. Because the panel stated the burden extremely broadly—“[t]he inference must be overcome by evidence that rebuts the probability that some consumers will find the new mark both offensive and harmful to the reputation and the favorable symbolism of the famous mark”—it seems that such a burden could not be overcome by most surveys, since some consumers probably will find it harmful to the reputation of the famous mark, even if most do not.A second point raised by the opinion is the court’s implication that one need only show that “some consumers” would change their perceptions in order to support a tarnishment claim. That, of course, begs the question of how many of such consumers is sufficient. In likelihood-of-confusion cases, a relatively low level of confusion has been deemed sufficient. The issue is problematic in dilution cases. On the one hand, dilution is often described as a problem of many very small changes eventually leading to a large problem, which could support a fairly low level of change in order to support a dilution claim. On the other hand, dilution is not a protection against deception, and there is a danger that creating too low a barrier, particularly for tarnishment claims, could stifle many communicative uses which may not be automatically excluded by the statute. Moreover, if the barrier is set very low, then it will be virtually impossible for a defendant to overcome the presumption.A more fundamental issue with the presumption is whether it is legitimately grounded in fact. Although the court analogized the situation to res ipsa loquitor, it is by no means evident that sexually oriented uses of a famous mark necessarily tarnish an observer’s view of the mark. The court cited only case law holding such uses to be tarnishing, but did not cite (and the cases they cited similarly do not cite) empirical evidence to support the presumption. Indeed, there is empirical evidence indicating that sexually oriented uses may not tarnish a famous mark. Furthermore, a presumption that bans uses associating a famous mark with sex might be vulnerable to attack under the Supreme Court’s decisions in Matal v. Tam and Iancu v. Brunetti. Although in Tam the Justices did not fully agree on the First Amendment rationale for striking down a ban on disparaging remarks, both plurality opinions suggest that the presumption could be deemed viewpoint discrimination. In Brunetti, a majority agreed that the ban on registering “immoral or scandalous” marks constituted viewpoint discrimination, also suggesting that a presumption that automatically disfavors certain associations with sex could be deemed viewpoint discrimination.Another point about the majority’s opinion is its subtle mischaracterization of the Supreme Court’s opinion in the case. The majority states that “the Supreme Court held that ‘actual harm’ rather than merely the ‘likelihood of tarnishment’ is necessary.” In fact, the Court did not separate blurring and tarnishment in most of its discussion and never used the phrase “likelihood of tarnishment.” The Court’s main reference to tarnishment was an observation that tarnishment may not be within the definition of dilution in the original federal law (a proposition not raised by the defendant). When discussing the relationship between dilution and mental association, the Court commented that “‘[b]lurring’ is not a necessary consequence of mental association. (Nor, for that matter, is ‘tarnishing.’)” Thus, the Court indicated that it had not focused primarily on the tarnishment aspects of the claim. But, to the extent that it did, its opinion does not support a presumption of likely tarnishment from mental association.iii.Unwholesome Uses of Famous Marks in General (Including Sexually Oriented Uses): Case Law Favors PlaintiffsCertainly the most common situation for courts to invoke tarnishment is where the famous mark is used in connection with unwholesome activity, usually pornography. In cases under the FTDA, the most common circumstance was on the Internet, where marks have been used in connection with sexually oriented sites, either as domain names or on the sites themselves. In a number of such cases, the courts found tarnishment. Similarly, where the plaintiff’s famous mark is otherwise used by the defendant in an association with sexually oriented material, tarnishment has been found. However, not all such situations lead to a finding of dilution. In particular, Playboy Enterprises has lost suits because some courts apparently believe Playboy’s actions promoting sexually oriented material make it “tarnish-proof.” Similarly, a court refused to find tarnishment when a bar used the name Greatest Bar on Earth, citing sales of alcoholic beverages at the circuses featuring the mark Greatest Show on Earth.In V Secret Catalogue, Inc. v. Moseley, the Sixth Circuit upheld a finding of tarnishment based on an adult novelty store’s use of the name Victor’s Little Secret. The district court had cited evidence that the Army colonel who brought the matter to Victoria’s Secret’s attention was offended and that this would harm the reputation of Victoria’s Secret. Moreover, in what may be termed an ironic evidentiary move, the district court cited Victoria’s Secret’s statements that it carefully drew the line between “sexy and playful” and explicitly sexual in its advertising, whereas Victor’s Little Secret (the defendant’s store) sold adult sex toys in addition to its line of lingerie and other items. Most notably, the Sixth Circuit imposed a presumption of tarnishment where the unauthorized use is sexually oriented.In an even more ironic twist on sexually oriented marks being tarnished, the maker of Viagra obtained a preliminary injunction on the grounds of tarnishment against a defendant who sold advertising space on decommissioned military equipment. The defendant put the plaintiff’s “Viva Viagra” slogan on a decommissioned missile displayed in New York City and was proposing to exhibit it at various adult conventions where women would distribute condoms. The court, in a brief and conclusory fashion, decided that this was likely to tarnish the reputation of the plaintiff’s mark.In Lorillard Tobacco Co. v. California Imports, Ltd., the court found tarnishment based on the similarity of the marks (Newport and Newprot), the similarity of the products, and the fact that defendant’s product was associated with “spice,” or synthetic marijuana. In 3M Co. v. Christian Investments, LLC, the court, entering a default judgment, found that the use of plaintiff’s mark in a domain name that offered online gambling services constituted tarnishment.In a somewhat questionable decision, a federal district court found that a movie entitled “Dairy Queens” tarnished the Dairy Queen trademark. The movie was intended to be a satire of “dairy country” and would have portrayed it in an unflattering (and occasionally off-color) manner. The court rejected a First Amendment defense, rejecting the defendant’s argument that its commentary should have permitted the use of the title.Bulk e-mailers, or “spammers,” on the Internet also have been cited for tarnishment. In one case, a bulk e-mailer sent unsolicited e-mail to thousands of America Online (AOL) subscribers, many of which contained aol.com in the header, engendering numerous complaints to AOL. The court found that this tarnished AOL’s mark. In another case, a bulk e-mailer used a false return e-mail address indicating that the e-mail came from an AOL subscriber. This, too, was found to cause tarnishment. Although certainly reprehensible and probably tortious, such uses of the FTDA were at the outer limits of dilution claims. The bulk e-mail senders did not use AOL as a trademark, although they may cause an association of AOL with the unwelcome spam.When evaluating claims of tarnishment, one must also consider the noncommercial use exemption in Section 43(c)(3)(C); a sexually oriented parody can be considered a noncommercial use.b.Poor Quality Goods or ServicesMark owners have had mixed success attempting to claim tarnishment where the defendant used the mark in connection with goods or services that the mark owner believed were of poor quality and reflected badly on its brand. This category has some overlap with the unwholesome activity, especially where services are involved. In fact, mark owners have been more successful portraying alleged infringers’ services as tarnishing than portraying goods as tarnishing.In Gideon’s International, Inc. v. Gideon 300 Ministries, Inc., the plaintiff contended that the defendant, a Christian ministry, tarnished the Gideon mark by soliciting donations. The court upheld this claim, noting that the plaintiff’s members do not solicit donations. Similarly, in New York State Society of Certified Public Accountants v. Eric Louis Associates, Inc., the court credited a tarnishment claim, where the plaintiff alleged that the defendant’s for-profit activity would reflect badly on the plaintiff’s not-for-profit organization. However, the court was skeptical of another tarnishment claim, that the defendant’s method of doing business (as a “headhunter” firm) was viewed as “unscrupulous” and harmful to the plaintiff. In TotalEnergies Marketing Puerto Rico Corp. v. Rivera-Robles, the court granted a preliminary injunction for tarnishment where the evidence was that defendant covered up signage plaintiff’s mark and sold another brand of gasoline at its stations. And in New York City Triathlon, LLC v. NYC Triathlon Club, Inc., the court upheld a tarnishment claim based on the defendant’s poor reputation for providing services related to running a triathlon.In Hilton Worldwide, Inc. v. Global Advertising, Inc., also a default judgment, the court found that defendant’s telemarketing services, which used the Hilton trademarks and falsely represented a connection, were tarnishing. In Abbott Laboratories v. Revitalyte LLC, the court denied a motion for judgment on the pleadings, finding that plaintiff had made out a claim for tarnishment by alleging that defendant’s “advertisements [the offending use] feature references to heavy drinking by young adults, encourage excessive consumption of alcohol, and often feature other conduct that could be described as vulgar and crass, including the use of expletives and suggestive placement of the product.”In Gianni Versace, S.P.A. v. Versace 19.69 Abbigliamento Sportivo SRL, the court found dilution by tarnishment, citing evidence of the poor quality of defendant’s goods (the court also found dilution by blurring). This is an instance in which tarnishment was from poor quality goods, not services.Several other cases, although accepting the general proposition that poor quality goods or services can lead to tarnishment, have rejected such claims on the given facts. In Sony Computer Entertainment, Inc. v. Connectix Corp., the Ninth Circuit reversed a district court finding of tarnishment, which was based on poor-quality goods. The plaintiff claimed that the defendant’s device, which allowed computer users to play Sony PlayStation games without a dedicated console, gave users a bad experience and tarnished the PlayStation mark. The Ninth Circuit actually rejected two possible tarnishment claims. It found insufficient evidence of misattribution—that is, the defendant’s users attributing poor playability to Sony—and also insufficient evidence that the defendant’s product gave players a poor experience. Another example is Ace Technology Corp. v. GFM Corp., where the court rejected a tarnishment claim based on the defendant’s failure to use a patented process in the manufacture of licensed tools bearing the plaintiff’s mark. The court found no evidence that the defendant’s tools were, in fact, inferior, or that any customers had complained about their quality.In Nordstrom, Inc. v. Nomorerack, Inc., the court found that evidence of poor reviews of defendant’s service on various complaint websites was insufficient to demonstrate a reputation for poor service that was likely to cause dilution by tarnishment of plaintiff’s Nordstrom mark. And in ArcelorMittal USA, LLC v. Arillotta, where defendants defaulted, the court refused to find tarnishment where the main defendant was alleged to have fraudulently represented to third parties that it had a contract with plaintiff. Similarly, in Crossfit, Inc. v. Quinnie, another default case, the court (citing ArcelorMittal) refused to find tarnishment where plaintiff did not allege that the quality of defendant’s services (as opposed to other peripheral issues) was inferior.Similarly, in Starbucks Corp. v. Wolfe’s Borough Coffee, Inc., the Second Circuit affirmed the district court’s rejection of a tarnishment claim based on the defendant’s use of Mr. Charbucks as a coffee name, pointing out that the defendant’s coffee was of “very high quality.” In Lexington Management Corp. v. Lexington Capital Partners, the court refused to find tarnishment, despite the plaintiff’s claims that the defendant’s manner of doing business (aggressive “cold calling”) and allegations of regulatory investigations of the defendant caused tarnishment of the plaintiff’s business. And in Libbey Glass, Inc. v. Oneida, Ltd., the court denied both sides’ motions for summary judgment on the issue of tarnishment, finding an issue of fact as to the inferior quality of the defendant’s product. The mark at issue in this case was the design of the plaintiff’s product, not a logo or word mark.Calvin Klein Trademark Trust v. Wachner raised some interesting tarnishment issues in connection with poor quality goods. The court dismissed a tarnishment claim that was based on allegations that the defendant had made unauthorized distributions of authentic Calvin Klein merchandise. However, it permitted a trial on the plaintiff’s claim that the defendant had not followed prescribed packaging and design requirements. In a somewhat similar case, Four Seasons Hotels & Resorts B.V. v. Consorcio Barr, S.A., the court imposed liability under the FTDA and the Florida dilution statute where the defendant was licensed to use the Four Seasons mark for its hotel in Caracas, Venezuela, and did not operate the hotel according to the standards of a Four Seasons hotel. Although a suit against a licensee is not a suit against a wholly unauthorized user, such suits are often used when the licensee/franchisee continues to use the mark after the franchise agreement has been terminated. In a somewhat contrasting case, a court dismissed a dilution claim arising from the defendant’s unauthorized distribution of the plaintiff’s goods. The court acknowledged that the defendant’s distribution of expired goods “may cause reputational damage to [defendant],” but held that a diluting use must be one on non-competing goods.Even apart from a licensee or franchisee, the unauthorized use of the plaintiff’s mark on identical goods, particularly counterfeit goods, is a situation in which tarnishment claims are often successful. There are two variations on these cases: (1) unauthorized sales of actual branded goods (as in gray market situations) and (2) counterfeit goods. Both types of cases can rely on the statement in Moseley that the use of identical marks may be circumstantial evidence of actual dilution. Both rely on the underlying notion that the trademark owner’s loss of control threatens the quality-signal function of the trademark. Such uses do not directly attack the source-identifying function of the mark, since they use the mark on the same goods as the famous mark owner, and the public presumably assumes that the goods come from the famous mark owner. However, they do raise the possibility of harm to the trademark owner’s reputation, consistent with the TDRA’s definition of tarnishment. On the other hand, “unauthorized” sales are not necessarily sales of poor quality goods. Indeed, in Scott Fetzer Co. v. House of Vacuums, Inc., the court rejected a tarnishment claim based on alleged unauthorized sales and reconditioning of the plaintiff’s branded merchandise, noting that “[t]rademark law does not entitle markholders to control the aftermarket in marked products.” This sentiment was reiterated in another case, American Airlines, Inc. v. Despegar.com USA, Inc., where the court dismissed a dilution claim based on defendant’s alleged sale of seats on American flights at higher fares than those charged by American itself, and the perception that American endorsed such activity. The court noted that defendant did not use its own mark, but rather was using American’s mark to refer to American (essentially a nominative use, although the court did not cite the nominative use exclusion). However, another court, in a default judgment situation, upheld a dilution-by-tarnishment claim by Hilton hotels against a telemarketer that falsely represented that it was partnering with Hilton to sell vacation packages.Microsoft received a default judgment in its favor where it claimed dilution and tarnishment as a result of defendant’s use of various Microsoft marks to engage in hacking and cybertheft.However, a post-TDRA case indicates how broadly tarnishment can apply in gray market goods cases. In Dan-Foam A/S v. Brand Named Beds, LLC, the court upheld a dilution claim where there were “material differences” in the goods such that the sale of the gray market goods could harm the reputation of the famous mark. Furthermore, the “likelihood of dilution” standard and the broad definition of tarnishment in the TDRA make claims against counterfeit marks a more defensible fit within the dilution statute, despite the fact that a confusion claim would seem to be the most logical one.Not surprisingly, courts have continued to find dilution by tarnishment in counterfeiting cases, though rarely do they seriously attempt to fit it within the statutory language (many of the claims are resolved by default judgments).In the area of use on poor quality goods, the key issue is the actual quality of the defendant’s goods. Except in counterfeiting cases, courts are reluctant to assume that the second user’s goods are not of good quality.c.Subjecting the Mark to Ridicule or CriticismThis is perhaps the most controversial form of tarnishment. In large part, the problem stems from the fact that this type of use is a directly communicative use and therefore implicates free speech values. The TDRA provides explicit avenues of defending against liability for many such uses, primarily in the exclusions of “noncommercial” uses, and “fair use” for parody and commentary. These exclusions are discussed in detail elsewhere in this book. Suffice it to say that in several cases courts have applied the noncommercial use exclusion to thwart tarnishment claims, and one would expect an increasing use of the fair use exclusion. Moreover, it was clearly the intent of Congress that the FTDA and TDRA not be used to prevent the dissemination of satire, parody, or commentary. As discussed in connection with state-law claims, the success of this type of claim may turn on the “commercial” nature of the use, and thus the level of protection it can claim under the First Amendment. Moreover, the Supreme Court’s decision in Matal v. Tam, striking down the Lanham Act’s ban on registering disparaging trademarks, could be applied to prevent dilution by tarnishment claims against disparaging uses of another’s mark. Similarly, the Court’s decision in Iancu v. Brunetti, striking down the “immoral or scandalous” bar to registration, could be applied to bar tarnishment claims against uses in unwholesome contexts. It is not surprising, therefore, that such claims have been relatively unsuccessful outside of a narrow set of circumstances.The most successful invocations of this form of tarnishment have been in Internet cases, in which a critical Web site uses the exact trademark of the well-known organization as its domain name. In Jews for Jesus v. Brodsky, for example, the court found both blurring and tarnishment by a Web site critical of the plaintiff organization. In Kraft Foods Holdings, Inc. v. Helm, the court rejected a parody claim in which the defendant used the nickname “King VelVeeda” on a Web site offering sexually oriented material (though it did not use the plaintiff’s Velveeta mark as part of the domain name). The defendant claimed that, in addition to commercial services, his site offered commentary and protectable art and therefore was entitled to protection. However, the court found that the commercial aspects of his site, and the use of the “King VelVeeda” name, could be separated from the noncommercial aspects and rejected his defense. However, even on the Internet, tarnishment claims in this situation are often unsuccessful. For example, a tarnishment claim was rejected in Bally Total Fitness Holding Corp. v. Faber. The defendant created a Web site critical of Bally, called “Bally sucks.” The court held that the use of the Bally mark was for commentary purposes and should not be considered tarnishment. Unlike other Internet cases, such as Jews for Jesus, the defendant did not use BALLY.COM or another confusingly similar name as its domain name address. Thus, tarnishment had to be based on the associations made by a person who made more than a minimal effort to view the “Bally sucks” site, since one would not ordinarily reach the site by accident.In an unusual case of a successful tarnishment claim against a parodic use, a district court upheld a jury finding of actual dilution via defendant’s sale of T-shirts containing allegedly offensive uses of a university mascot and other trademarks. Another successful tarnishment claim in a parody-type case was Pfizer, Inc. v. Sachs. The defendants put the plaintiff’s Viagra and Viva Viagra marks on a missile, which they exhibited publicly—including in front of Pfizer’s New York headquarters—as a means of publicizing their outdoor advertising business. But this was not the only place that the missile was displayed. As stated by the court: “Defendants exhibited the Viagra-branded missile at an adult entertainment exhibition, and informed Pfizer that the missile would be displayed again, with two models ‘riding’ the missile and distributing condoms.” This was found (with no discussion) to be a tarnishing use; however, the sexual context clearly influenced the court’s decision. This also was a somewhat unusual case in that the defendant’s commercial purpose was clearly the primary reason for the use.In another somewhat unusual case, a district court found dilution by tarnishment in a party’s use of “National Association for the Abortion of Colored People” in conjunction with the NAACP mark on a website criticizing what it believed was the position of the NAACP on abortion. However, that ruling, and its rejection of a claim of protected speech, was reversed on appeal to the Fourth Circuit. Although the appeals court agreed that there was a prima facie case of tarnishment, it found the use in question to fall within the statutory exclusions for fair use and noncommercial use.In VIP Products, LLC v. Jack Daniel’s Properties, Inc., the court found dilution by tarnishment where the offending use was a line of dog toys called “Bad Spaniels,” described by the court as “in the shape of a liquor bottle and features a wide-eyed spaniel over the words ‘Bad Spaniels’, ‘the Old No. 2, on your Tennessee Carpet.’ … At the bottom of the ‘Bad Spaniels’ toy, it reads: ‘43% POO BY VOL.’ and ‘100% SMELLY’.” The court credited the expert for Jack Daniel’s, who found that consumers would acquire negative associations with the Jack Daniel’s brand from seeing the “Bad Spaniels” toys. In other cases, tarnishment claims based on ridicule have been less successful. One prominent example is Lyons Partnership v. Giannoulas, in which the owner of the Barney dinosaur character sued the creator of the Famous Chicken (a character that entertains at sporting events). The chicken performed a skit with a “putative Barney” that portrayed Barney in an unflattering manner. The court rejected a tarnishment claim, finding no “shoddy quality” or “unwholesome or unsavory context” in the defendant’s parody.In Tommy Hilfiger Licensing, Inc. v. Nature Labs, LLC, the defendant created a line of “pet perfumes,” whose names parodied those of actual perfumes. The one in question was called Timmy Holedigger. Noting a lack of any evidence of harm to the Tommy Hilfiger mark, the court rejected a tarnishment claim by Tommy Hilfiger’s licensing arm. In Louis Vuitton Malletier, S.A. v. Haute Diggity Dog, LLC, the Fourth Circuit rejected a tarnishment claim under the TDRA brought against the maker of a Chewy Vuiton dog toy.In Ford Motor Co. v. 2600 Enterprises, the court rejected a tarnishment claim in which the defendant owned the Web site “fuckgeneralmotors.com” which automatically redirected users to the “ford.com” Web site of the Ford Motor Company. The court noted that the domain name did not use any of Ford’s marks, and the court did not believe the use “in any way hampers [Ford’s] commercial success in an unlawful manner.” The court further stated that the FTDA was not intended to be “a tool for eliminating Internet links that, in the trademark holder’s subjective view, somehow disparage its trademark.”In Mattel, Inc. v. MCA Records, Inc., the court rejected a dilution claim against the distributor of a song entitled “Barbie Girl” that denigrated the lifestyle symbolized by the Barbie doll. Although primarily based on the noncommercial use exclusion of the FTDA, the court’s opinion was generally cautious about applying the FTDA to a parody such as the song in question. The court indicated that it doubted even that tarnishment would occur, noting that the alleged “wholesome” image of Barbie already had been degraded by identification with less-than-perfect values. Moreover, the court stated that it “must be wary of applying anti-dilution statutes to permit ‘a trademark owner to enjoin the use of his mark in a noncommercial context’ simply because they [sic] find such parodies ‘negative or offensive.’”After the Supreme Court’s decision in Moseley v. V Secret Catalogue, Inc., and prior to the TDRA, several courts rejected tarnishment claims of this nature on the merits. In Caterpillar, Inc. v. Walt Disney Co., the court denied Caterpillar’s attempt to enjoin the movie George of the Jungle 2, which, in one scene, prominently portrayed a Caterpillar bulldozer being used by people to destroy the jungle. The court rejected the claim, noting that “even the most credulous viewer or child” would not attribute the bad acts of the users to Caterpillar. Similarly, in Wham-O, Inc. v. Paramount Pictures Corp., the court rejected a tarnishment claim based on a misuse of the plaintiff’s Slip ’N Slide toy in a movie. In neither of these cases was the ridicule directly aimed at the product, but it is submitted that the courts correctly understood that they were used as “cultural props” and therefore were important parts of the commentary incorporated into the movies. And in World Wrestling Federation Entertainment, Inc. v. Big Dog Holdings, Inc., the court rejected a dilution claim based on the defendant’s parody T-shirts, noting that the plaintiff’s own T-shirts contained far more offensive material than the T-shirts of the defendant. Although these cases were decided under an actual-dilution standard, it is probable that they would be decided the same way under the likelihood-of-dilution standard. In all of these cases, it appeared that the court did not believe that the second use harmed the reputation of a famous mark. That is a requirement for dilution by tarnishment. (These uses probably would be exempt under Section 43(c)(3)(A)(ii) or (c)(3)(C).)Louis Vuitton Malletier, S.A. v. Haute Diggity Dog, LLC  is one example of an unsuccessful post-TDRA tarnishment claim. The court rejected an allegation that an inexpensive Chewy Vuiton dog toy in the shape of a woman’s handbag tarnished the Louis Vuitton trademark. Another is Anheuser-Busch, Inc. v. VIP Products, LLC, which also involved a dog toy, this one containing a replica of a Budweiser label, but with the word “Buttwiper” substituted for Budweiser. Although there was some evidence of possible unwelcome associations, the court rejected the claim, citing a lack of any evidence of harm to the mark’s reputation.In Allied Interstate, LLC v. Kimmel & Silverman P.C., the court rejected a tarnishment claim based on defendant’s use of plaintiff’s mark in keyword advertising and on defendant’s website. The references were all to plaintiff, not to any products or services of defendant; the court held that such uses did not constitute dilution.The caution shown by most courts in this area is appropriate. Congress had indicated that it does not wish dilution claims to be used to suppress legitimate communication. Particularly where confusion is absent, courts must carefully balance the communicative rights of the second user against the trademark owner’s rights. Moreover, as more than one court has commented, a successful parody can reinforce the tie between the famous mark and the single source—indeed, it relies on the connection for its success.Still unknown is whether a humorous alteration of a trademark used in a comparative advertisement will constitute tarnishment under the TDRA. The Second Circuit, interpreting New York law, held that this constituted tarnishment. It may be argued that an alteration of the mark is not a “fair” use. But when the alteration does not associate the mark with some clearly unwholesome activity, courts should grant greater leeway in the tarnishment analysis for comparative advertising, in accordance with Congress’s mandate. In the final analysis, the test is whether the defendant’s use is likely to harm the reputation of the famous mark, not whether it is possible that the use will be harmful. A “broad” view of tarnishment is not consistent with the TDRA’s limited definition of tarnishment, nor is it consistent with the TDRA’s exclusions, which further indicate an intent to narrow the use of tarnishment.In various tarnishment cases decided under state and federal laws, courts have required some evidence of tarnishment, rather than relying on an assumption of harm. This is particularly true of cases not involving pornography or drug-related references. This requirement is not illogical. As the USTA noted when it proposed a federal dilution statute, tarnishment is a different sort of wrong than dilution. If one views tarnishment as similar to a claim for trade disparagement, then a requirement of evidence would be analogous to a requirement of special damages.The importance of actual evidence of tarnishment in parody cases was illustrated in Tommy Hilfiger Licensing, Inc. v. Nature Labs, LLC. The court found no tarnishment of the plaintiff’s Tommy Hilfiger mark from the defendant’s “pet perfume” parodies, noting a lack of any evidence of harm to the mark. Even after the TDRA, courts should not lightly assume that any seemingly undesirable association would be likely to harm the reputation of a famous mark. Indeed, the first post-TDRA appellate decision appeared to agree. In Louis Vuitton Malletier, S.A. v. Haute Diggity Dog, LLC, the court rejected a tarnishment claim brought against the maker of a Chewy Vuiton dog toy. The court found nothing in the record to suggest that the toys would harm the reputation of the Louis Vuitton mark. In Starbucks Corp. v. Wolfe’s Borough Coffee, Inc., the court rejected a tarnishment claim against Mr. Charbucks coffee, and rejected the plaintiff’s interpretation of its survey evidence. Even though there was some evidence that people may have a negative reaction to Charbucks coffee, the court found no indication that this attitude affected the survey respondents’ reaction to the Starbucks mark. Although actual evidence of harm may not be a prerequisite, the conclusion that the reputation of the famous mark is likely to be harmed should not be based simply on speculation.3.First Amendment Concerns With Dilution by TarnishmentThe Supreme Court’s decisions in Matal v. Tam, striking down the Lanham Act’s bar to registering disparaging marks, and Iancu v. Brunetti, striking down the bar to registering immoral or scandalous marks, raise the issue of whether dilution—particularly dilution by tarnishment—might run afoul of the First Amendment, notwithstanding the statutory exclusions. In Tam, the Court found that the ban on disparaging trademarks constituted improper viewpoint discrimination. Apart from that, the Court was unable to reach consensus on an overarching rationale. However, some of the language in the Justices’ opinions, if applied to dilution, could make many tarnishment claims susceptible to First Amendment challenge. For example, Justice Alito, writing for all eight Justices, said of the ban on disparaging marks, “Speech may not be banned on the ground that it expresses ideas that offend.” Thus, the fact that another’s use of a famous mark causes offense—either to the mark owner or to consumers—cannot be the basis for banning that use. In his concurring opinion (written on behalf of four Justices), Justice Kennedy went even further:The Government may not insulate a law from charges of viewpoint discrimination by tying censorship to the reaction of the speaker’s audience. … The danger of viewpoint discrimination is that the government is attempting to remove certain ideas or perspectives from a broader debate. That danger is all the greater if the ideas or perspectives are ones a particular audience might think offensive, at least at first hearing. An initial reaction may prompt further reflection, leading to a more reasoned, more tolerant position.Indeed, a speech burden based on audience reactions is simply government hostility and intervention in a different guise. The speech is targeted, after all, based on the government’s disapproval of the speaker’s choice of message. ……A law that can be directed against speech found offensive to some portion of the public can be turned against minority and dissenting views to the detriment of all. The First Amendment does not entrust that power to the government’s benevolence. Instead, our reliance must be on the substantial safeguards of free and open discussion in a democratic society.In Brunetti, a majority found that the ban on registering “immoral or scandalous” marks was viewpoint discrimination. That the ban precluded “offensive” marks meant that it regulated differently depending on the point of view expressed by the mark. Dilution by tarnishment may well cross this line, at least in some circumstances. Thus, the kind of use at issue in University of Kansas v. Sinks, where the university was offended by nasty parodies of its mascot on T-shirts, could be categorized as viewpoint discrimination (it is unlikely that a tarnishment claim would be brought against someone making T-shirts praising the university, although a blurring claim might be brought). In cases where one creates a mark that ridicules another mark (the “Lardashe” mark might be an example; another could be a mark like “Coke sucks”), one could also defend a tarnishment claim by asserting that banning the use constitutes viewpoint discrimination. As noted earlier, a presumption that associating a famous mark with sex or drugs constitutes tarnishment could also be deemed viewpoint discrimination. However, under Brunetti, the fact that such unwholesome associations might simply be offensive is similar to the viewpoint-based regulation problem found by the Court. In many cases, the fair use or noncommercial use exclusion would prevent liability. In others, a court could find that the parody itself precludes a dilution claim. However, all courts may not agree, and Tam’s and Brunetti’s implications of the First Amendment are powerful reminders that courts must tread lightly when considering tarnishment claims.Dilution by blurring is more complicated. Justice Kennedy’s concurring opinion carefully avoided opining on the effect of Tam on other trademark issues. Depending on how the Court views dilution by blurring, it may see such claims as more analogous to infringement claims that “protect consumers and trademark owners.” Moreover, to strike down blurring, the Court would have to distinguish its opinion in San Francisco Arts & Athletics, Inc. v. United States Olympic Committee, where the Court upheld the grant of dilution protection of the word “Olympic” for the United States Olympic Committee. And the most recent pronouncement of the Court on that case suggests that such a distinction may not be forthcoming, particularly if the Court determines that blurring is viewpoint-neutral.F.Dilution by Confusion1.Discussion and CommentaryAlmost all courts, state and federal, have assumed that the existence (or likelihood) of confusion means that dilution also exists. However, if a court is inclined to give the TDRA a close reading, it could come to a different conclusion. The issue arises from the definition of dilution. Under the TDRA, dilution by blurring is an “association [between the famous mark and the second user’s mark] that impairs the distinctiveness of the famous mark.” That normally means that consumers view the trademark as signifying another source in addition to the owner of the famous mark. But consider what confusion implies. If a consumer is confused, he or she believes that the use of the famous mark by a second user means that the second user has some connection with the famous mark’s owner. In other words, the consumer does not believe that the famous mark signifies more than one source. Rather, the consumer believes that the second use of the famous mark denotes the famous mark owner. Because the consumer continues to believe that the famous mark signifies only the famous mark owner, the distinctiveness of the famous mark (in the sense that it signifies a unique source) is not weakened. If anything, it reveals how strong this distinctiveness continues to be in the face of a second use. One may conclude, then, that dilution by blurring is not a subset of (or a result of) confusion.On the other hand, if one considers dilution to mean a loss of “distinctiveness” in a broader sense, or a loss of selling power, then confusion may be said to result in dilution. The unauthorized use is beyond the control of the mark owner. Particularly when the mark is used on non-competing products, it may alter the consumer’s perception of the “message” conveyed by the mark. In that way, it changes the “distinctive quality” of the mark by altering the distinctive message it previously conveyed. This may cause a loss of selling power as well. Alternatively, if one interprets the TDRA definition of blurring as referring to a famous mark’s capacity to signify particular products or services, then confusion (in which consumers erroneously view the mark as signifying products not connected with the famous mark owner) would reduce the “particularity” of the mark.Focusing on the message of the mark rather than on its distinctiveness might be more likely to fit the definition of dilution by tarnishment, which focuses on harm to the reputation of the famous mark.Originally, Congress probably intended that whatever had been recognized as diluting under various state laws (with the exception of uses covered by the exclusions in Section 43(c)(4)—now 43(c)(3)) would be considered diluting under the FTDA. However, the TDRA comes after court opinions reading the statute strictly, so one must assume its terms were drafted carefully. Thus, an interpretation that is both narrow and consistent with the plain language of the statute should not be surprising.2.Confusion Usually Viewed as Tantamount to DilutionDespite the theoretical problems raised by confusion and dilution, courts tend to regard a finding of confusion as tantamount to dilution. In Nabisco, Inc. v. PF Brands, Inc., for example, the Second Circuit, although recognizing that confusion was not necessary for dilution, clearly regarded it as sufficient. “Consumer confusion would undoubtedly dilute the distinctive selling power of the mark. … A junior use that confuses consumers as to which mark is which surely dilutes the distinctiveness of the senior mark.” That is not necessarily correct, however; if the consumer misidentifies the defendant’s mark as the plaintiff’s, that may not translate into a reduction in the plaintiff’s mark’s selling power, or its uniqueness. The mistaken consumer still believes that the mark signifies a single source.In a district court case following Nabisco, the court determined that the existence of a form of confusion was highly relevant to a finding of dilution. One can find similar sentiments expressed in other cases. Indeed, in an unpublished opinion, the Fifth Circuit specifically rejected an argument that dilution and confusion were mutually exclusive claims. However, the Sixth Circuit found confusion to be “irrelevant” to dilution.In Apple, Inc. v. Samsung Electronics Co., the court excluded a confusion survey conducted by defendant Samsung on the grounds that it would be irrelevant to a dilution claim (Apple had dismissed its infringement-by-confusion claims). The exclusion was somewhat unusual in that the survey was proffered by the defendant, not the plaintiff. This demonstrates a one-way effect of confusion; confusion would be deemed relevant to dilution, but the lack of confusion is irrelevant.Dilution does not follow from every confusing use; the requirement of a famous mark limits dilution claims where infringement could still exist. Moreover, it is worth repeating that dilution should not be used as a fallback for a weak claim of infringement.G.Dilution by Genericide“Genericide” refers to the destruction of a mark’s source-identifying power, resulting in a generic term, rather than a trademark. Even if it results from some action on the part of another, such as generic use in a commercial, or generic use in a dictionary, it does not fit the classic mold of dilution by use on a non-competing product or service. Normally, uses that cause genericide would be either an advertisement by a competitor or a nontrademark use of some sort. The legislative history of the original FTDA included “diminishment” among the categories of dilution Congress believed it was including in the definition. Although there is no discussion of what was intended by “diminishment,” it is not unreasonable to assume that genericide was included. After all, most other forms of dilution would be included in the other listed categories of blurring and tarnishment. The TDRA has now narrowed dilution to two categories: dilution by blurring and dilution by tarnishment. If genericide cannot be made to fit the current definitions of one of those categories, then it would not be actionable. It may be possible to fit genericide into dilution by blurring. After all, it does attack the distinctiveness of the mark by making it completely nondistinctive. However, it is not clear that genericide operates by creating an “association” in the mind of the consumer between the second mark and the famous mark.Thus far, cases decided under the FTDA and TDRA have largely avoided this category of dilution. A few cases decided under state laws appear to have recognized this category. Some cases arose in connection with comparative advertisements or advertisements that involved compatible products. Such suits face two obvious obstacles. First, there is the familiar principle that collateral, nondeceptive uses of another’s mark are permissible in comparative advertising under a fair use or nominative use rationale. Second, the TDRA excludes “[a]ny fair use … of a famous mark … including use in connection with [] advertising or promotion that permits consumers to compare goods or services” from liability. Although one could argue that a deliberate use of a mark as a generic term in a comparative advertisement would not be a “fair” use, this may depend on how close the famous mark is to being the only way to describe the goods or services. If it is a descriptive use, then the fair use exclusion would apply as well.A third problem with genericide is that prohibiting uses in dictionaries, books, or other similar works would raise serious free speech issues. Such uses also dovetail with the noncommercial use exclusion of the TDRA, which was intended to preserve free speech values by excluding problematic uses (those not deemed “commercial”) from liability. Thus, the lack of cases discussing this category is not surprising.A panel of the Seventh Circuit expressed skepticism about whether genericide was even covered by the (pre-TDRA) FTDA. Writing for the court in Ty, Inc. v. Perryman, Judge Posner surmised that the plaintiff’s dilution claim against the defendant’s “bargainbeanies.com” Web site was intended to prevent the Beanie Babies trademark from becoming generic. He noted that there are both costs (the mark owner must search for a new mark) and benefits (“an addition to ordinary language”) that result from a mark’s becoming generic. Ultimately, the court found the extension of dilution law in this context could threaten “the public interest” and rejected it. A later Ninth Circuit case also found genericide not actionable, although it was in a case that did not raise the issue as a dilution claim.H.Dilution and Internet Domain Names1.“Cybersquatting” and the Challenge for DilutionThe use of famous trademarks as components of Internet domain names has caused considerable difficulty for courts. A typical situation involves someone who is not the trademark owner but who registers a domain name in the form of “trademark.com” or “trademark.net” with a domain name registrar. In some instances, people have registered domain names with numerous variations on the trademark. The purpose is often to attempt to sell the domain name to the trademark owner at a high price. The legislative history of the FTDA indicates some concern about this type of activity, often termed “cybersquatting.” Most notable is this observation by Senator Patrick Leahy:[I]t is my hope that this antidilution statute can help stem the use of deceptive Internet addresses taken by those who are choosing marks that are associated with the products and reputations of others.When the FTDA became effective in 1996, many trademark owners attempted to use it to remedy cybersquatting. However, courts have not always found it easy to fit domain name registration problems into the traditional dilution categories of blurring and tarnishment. Where the domain name registrant uses the famous mark to lure Internet users to a pornographic site, courts were fairly receptive to a tarnishment claim. Cases not involving pornography, drugs, or other unwholesome activity are more difficult to categorize. The act of registering a “blocking” or even confusing domain name does not, by itself, tarnish the famous mark. Unless the registrant uses the domain to purvey an unsavory setting, tarnishment does not fit the situation. Blurring also is not an appropriate description of the problem. The domain name itself seldom is used in a trademark sense, particularly when the issue is cybersquatting (registering a domain name in order to profit by selling it to the famous mark owner). In such situations, consumers will not be likely to associate the famous mark with two sources or two products, since the registrant is not trying to use the mark as a source identifier. It has also been held that registering the domain names, or providing a forum for buying and selling them, does not constitute a “use” of the marks as required by the FTDA or the TDRA.In 1999, Congress, recognizing some of the unique difficulties that this problem causes for conventional trademark and dilution analyses, enacted the Anticybersquatting Consumer Protection Act (ACPA). The ACPA provides a cause of action to trademark owners against those who register trademarks as domain names with a “bad faith intent to profit” from this activity.The ACPA supplies a more direct method of remedying the problem than dilution or infringement, and it incorporates certain dilution standards in its analysis. However, the ACPA does not exclude other remedies; one still may bring a dilution claim in addition to a claim under the ACPA. Moreover, the analysis of dilution in these cases may be relevant to certain issues under the ACPA.Indeed, in Porsche Cars North America, Inc. v. Porsche.net, the Fourth Circuit held that one cannot maintain an in rem action for violation of the FTDA by cybersquatting. The court noted that Congress had provided a specific remedy in the ACPA for cybersquatting that included an in rem procedure, and that a contrary result would permit the plaintiffs to circumvent the specific requirements of the ACPA.The lack of a trademark use may be decisive after the TDRA. As noted elsewhere in this chapter, it appears that the statute (for both dilution by blurring and dilution by tarnishment) requires that the diluting use be “use of a mark.” If uses on the Internet are not considered uses as trademarks, then there may be no actionable dilution.2.Development of a “Per Se” Dilution Test for Internet CasesPrior to the TDRA, several courts, when faced with an Internet domain name dilution claim, fashioned an analysis that used neither blurring nor tarnishment analyses. Some of these courts stated that blurring and tarnishment are not the only possible dilution categories. With that premise, a number of courts concluded that the inability of the famous mark owner to use the domain name and/or the owner’s loss of control over its use constituted dilution. This analysis created a virtual per se rule that using another’s famous mark as the primary component of one’s domain name constitutes dilution.The leading cases in the use of the (pre-TDRA) FTDA against Internet domain name registrants are Panavision International, L.P. v. Toeppen, and Intermatic, Inc. v. Toeppen. Both involved the same defendant, Dennis Toeppen, who registered a number of domain names incorporating various trademarks. In Panavision, the Ninth Circuit stated that dilution need not fit into the blurring or tarnishment categories to be actionable. The court affirmed the district court’s determination that diminishing Panavision’s ability to use the Internet to sell its goods and services (because only one person can register a particular domain name) constituted dilution under the FTDA. In the court’s terms, the inability of Panavision to use the panavision.com domain name “‘lessens the capacity of the Panavision marks to identify and distinguish goods and services on the Internet.’” The court noted that allowing others to register domain names with Panavision’s trademark makes it more difficult for consumers to find Panavision’s own site and may deter consumers.A similar analysis was used in Intermatic. Toeppen had registered intermatic.com as a domain name. As in the later Panavision case, the court held that Intermatic’s loss of the ability to use the Internet site most conducive to its Intermatic mark constituted dilution by hindering Intermatic’s ability to use the Internet to promote its products. The court also cited Intermatic’s potential loss of control over its mark and reputation as supporting the finding of dilution.Following these two cases, a number of courts found that virtually any unauthorized use of a domain name containing a famous mark constitutes dilution. A typical example is Virtual Works, Inc. v. Network Solutions, Inc., a suit involving the plaintiff’s registration of the domain name vw.net. The plaintiff sued for tortious interference with its use of the domain name, and Volkswagen counterclaimed for infringement and dilution of its VW marks as well as for violations of the ACPA. Discussing Volkswagen’s dilution claim, the district court said simply, “[r]ecent case law holds that internet cyberpiracy constitutes per se trademark dilution.” Similarly, in Lozano Enterprises v. La Opinion Publishing Co., the court stated flatly that “[a] defendant who uses Plaintiff’s name as defendant’s Internet domain name violates the Federal Anti-Dilution Act.” The Panavision analysis was also followed in Toys “R” Us, Inc. v. Abir.Not all courts agreed with the per se approach. In some cases, courts have rejected it outright. In Hasbro, Inc. v. Clue Computing, Inc., the court rejected the use of a per se rule for Internet cases. However, it cautioned that its ruling was in the context of a mark owner whose mark was a common term (Clue) and where another party may have a legitimate reason to use the term as a domain name. Thus, Hasbro was not a classic cybersquatting case. Had the registration been made to extort money from the plaintiff or block the use of the domain name by the plaintiff, a different result (though perhaps not by a per se rule) may have ensued.HQM, Ltd. v. Hatfield  arguably went a step further. The defendant registered his name, Hatfield, as a domain name, hatfield.com. The only alleged use of the registration was for e-mail—the defendant apparently had no Web site. Citing the Fourth Circuit’s expressed fears that dilution not become a property right in gross, the court rejected the plaintiff’s attempt to use a per se rule. Instead, the court stated that the fact that the registration may complicate the search for the plaintiff’s goods on the Internet was not sufficient to be considered diluting of the mark. Although the defendant in HQM apparently had a legitimate basis for registering the domain name, the court clearly rejected the major premise of the per se rule, that making it more difficult for consumers to find the famous mark owner’s goods on the Internet constitutes dilution.The rationale of HQM was followed in Strick Corp. v. Strickland. The defendant, whose last name is Strickland, registered the domain name strick.com. When the plaintiff later tried to register the same name, it discovered the defendant’s registration. Eventually, the plaintiff sued to force the defendant to transfer the registration. The court rejected the plaintiff’s argument that the defendant’s registration, by preventing the plaintiff’s customers from using strick.com to locate Strick’s Web site, was per se dilution. Instead, it noted that the inconvenience to customers posed by the defendant’s ownership of strick.com was “trivial” and that Internet users could find the plaintiff’s site by using a search engine. Finally, the court stated that trademark law did not grant famous marks a monopoly on all domain names arguably similar to their marks. As in HQM, it should be noted that the defendant had a legitimate reason to register the domain name in question, since it was the closest available one to his last name.In Avery Dennison Corp. v. Sumpton, the Ninth Circuit denied the plaintiff’s motion for summary judgment against the defendant’s use of avery.net and dennison.net as e-mail domains. The court noted that cases like Panavision were based on the purported inhibition of commerce resulting from the defendant’s registration of a “blocking” domain name. But when the offending domain name is a “.net,” not a “.com,” the court was unwilling to presume that Internet searchers would be deterred in their search for a commercial entity, which would be expected to use a “.com” domain.The per se rule also was rejected in Ford Motor Co. v. GreatDomains.com, Inc. That court first stated that the intent of the registrant should not matter in the dilution analysis—i.e., the existence of dilution should not depend on a showing of bad faith. Thus, under Panavision’s per se theory, any registration of a domain name containing a famous trademark should be considered diluting. However, the court rejected the per se theory as beyond the proper scope of the dilution statute. Quoting the Strick case, the court agreed that the per se rule effectively granted “‘property rights-in-gross as a matter of dilution law.’”One court, while purporting to reject any per se rule, found dilution based largely on the rationale that the defendant’s registration of domain names using the plaintiff’s mark kept the plaintiff from using the Internet effectively to sell its goods, which is little different from a per se rule.Another court, although agreeing that Internet domain disputes may be actionable even if they fall outside the traditional categories of blurring or tarnishment, added the requirement that the defendant must make a trademark use of the famous mark (i.e., use as a domain name was not itself sufficient) before a dilution claim could exist. In other Internet cases, courts have applied the dilution tests from their respective circuits and found dilution to exist under those tests. In most cases, use of a domain name that is identical, or virtually identical, to the plaintiff’s famous trademark will be enjoined. However, where the mark is a common word, particularly where the second user’s domain name adds other terms, courts have not been as ready to enjoin the second user’s domain name.And, although several courts have cited the loss of control over one’s reputation as dispositive, not all courts were quite as solicitous of the mark owner’s desire for control. In Toys “R” Us, Inc. v. Feinberg, for example, the court rejected a blurring claim by Toys “R” Us against the owner of a Web site using the domain name gunsareus.com.3.The TDRA Undermines the Rationale for a Per Se Test The definition of dilution under the TDRA undermines much of the rationale for a per se rule. First of all, to the extent that the claim relies on blurring, the current definition does not fit the per se rationale used by courts. Courts have justified using dilution on the theory that the cybersquatter has “blocked” the mark owner from using its trademark as a domain name. However, “blocking” does not necessarily harm the distinctiveness of the famous mark, which is a key component of the definition of blurring. Moreover, the multiple statutory factors attached to the definition of blurring indicate that Congress favored a balancing approach rather than a rigid per se approach.Second, to the extent that courts using a per se test assumed that dilution encompassed categories other than blurring and tarnishment, that rationale is undercut by the TDRA’s limitation of dilution to those two categories.It is noteworthy that the first court to comment on this issue found no per se dilution test under the TDRA.There is one possible per se analysis available in Internet domain cases. As noted earlier in this chapter, the Sixth Circuit considers unauthorized sexually oriented uses of a famous mark to be presumptively tarnishing. That could justify a limited, quasi per se test—not a full per se test because the presumption (or inference) can be rebutted.4-A. Cybersquatting as Dilution After the TDRAEven though the TDRA undermines any per se test for dilution by cybersquatting, this does not mean that the TDRA cannot be applied to such activity. In one case, for example, a court found likely dilution by tarnishment where defendant did not provide any goods or services in connection with the offending domain names.However, it should be noted that the Anticybersquatting Consumer Protection Act (ACPA) provides a specific remedy, including the possibility of statutory damages, for cybersquatting, and it is to be expected that the ACPA will be the primary source of claims for cybersquatting.4.Domain Names and Critical CommentarySeveral Internet domain name cases have involved Web sites whose purpose is to provide criticism of an organization or a company. These cases divide generally into two categories. The first category includes cases where the domain name essentially consists of the trademark of the company or organization—i.e., in the form trademark.com, trademark.net, or trademark.org. The second category includes cases where the domain name contains the trademark, but also contains other indications that the site is not authorized by the trademark owner.One concern with Web pages containing critical commentary is whether such uses constitute a “use in commerce” of the famous mark, particularly when the purpose of the site is to comment on the business of the well-known mark owner. The operative language of the TDRA requires a “use in commerce” of the famous mark by the second user, and the phrase “use in commerce” is specifically defined in the Lanham Act. Moreover, if the commentary qualifies as a “noncommercial use,” then a dilution claim would be barred by Section 43(c)(3)(C). In cases prior to the TDRA, courts found sufficient commercial connections when the Web site contained links to money-raising ventures (and/or contained advertising) or where the site solicited funds from users. However, several cases applying the original FTDA’s “commercial use in commerce” requirement found no commercial use, even where the domain name was essentially trademark.com. Thus, in Northland Insurance Cos. v. Blaylock, the court gave greater leeway to the critical Web site. The defendant used northlandinsurance.com as the domain name for his Web site criticizing the company. The court found a lack of evidence of any commercial use and refused a preliminary injunction. Significantly, the court rejected the plaintiff’s argument that by interfering with the plaintiff’s use of the Internet to attract customers, the defendant made commercial use of the mark. As noted earlier, much the same argument about interfering with the plaintiff’s ability to use the Internet to communicate with customers has been used by courts (after finding commercial use) to support the existence of dilution.Similarly, in Taubman Co. v. Webfeats, the plaintiff, owner of a mall named The Shops at Willow Bend, sued the registrant of the domain name shopsatwillowbend.com. The Sixth Circuit, reversing the trial court, found that the defendant’s Web site (which originally contained information about the mall but later contained information about the lawsuit) did not constitute a commercial use. Significantly, although the site originally contained a link to a commercial business, once that link was removed, with a promise not to recreate it, the court found that any commercial connection had been severed. Moreover, the court rejected the plaintiff’s assertion that the defendant’s acceptance of an offer to buy the domain name would be a “commercial” use.And in Bosley Medical Institute, Inc. v. Kremer, the Ninth Circuit held that the defendant’s “gripe” site, www.bosleymedical.com, was not a commercial use and was not actionable as either infringement or dilution. In doing so, the court defined commercial use in commerce as “use in connection with the sale of goods.” It found no commercial links on the defendant’s Web site; a link to a public interest group representing the defendant was deemed not commercial. Finally, and perhaps most important, the court rejected the argument that commercial use was satisfied by blocking the plaintiff’s ability to use the domain name.Thus, prior to the TDRA, courts were not consistent about the treatment of trademark.com sites involving commentary, particularly regarding the commercial nature of the use.In the second category, where the domain name is not trademark.com, the “Bally sucks” case is probably the best known. There, the court rejected infringement and dilution claims based on the defendant’s creation of a Web site entitled Bally sucks. The defendant did not use bally.com or even ballysucks.com as its domain name, although it did use an address containing Bally in a subdomain: www.compupix.com/ballysucks. However, the court’s broad statement that commentary was not a “commercial” use and, even if it were, it would be protected speech indicates that it would have viewed ballysucks.com as being similarly protected.The two categories are distinguishable. If a person uses trademarksucks.com as a domain name for nondefamatory critical commentary, that does not prevent the mark owner from making efficient use of the Internet domain system to sell its services or promote its cause. As long as the domain name itself is not defamatory, its use should not be considered diluting, particularly not in the per se sense used in several cases. This principle was recognized by the Sixth Circuit which, in an infringement-by-confusion case, reversed an injunction issued against “sucks” Web sites.When the essence of the domain name is the trademark itself, the issue is more difficult. On the one hand, the purveyor of critical commentary does not need to use the unembellished famous mark as the domain name. It could use trademarkcriticism.com or even trademarksucks.com as the domain name. A search engine seeking Web sites probably would pick up the critical site in such circumstances. However, that would be far less effective as a means of reaching people.Assuming that there is no confusion, the critical commentary on the Web site should alert the Web surfer that this is not the site of another organization of the same name. Thus, an Internet user who accesses a site with the address plannedparenthood.com and finds an antiabortion site probably will not believe that there is now a second organization called Planned Parenthood that is dedicated to a cause antithetical to the better-known organization of the same name. The user would think that some other (perhaps unnamed) organization is using the Planned Parenthood name to attract attention to itself. In that case, the use of the domain name has not harmed the distinctiveness of the mark, which is what the TDRA’s definition requires.However, the actual Planned Parenthood case asserted that the use of the company’s or organization’s trademark is not essential to the critical message being transmitted on the Web site. This analysis separates the message on the Web site from the use of a famous mark to induce traffic to visit the site. Presumably, a domain name like www.plannedparenthoodkillsbabies.com would be deemed to send a message and would require a different analysis.5.Descriptive Uses of Famous Marks in Domain NamesSome uses of famous marks in domain names do not follow the classic “cyberpiracy” model of Trademark.com. These parties use the trademark as part of the domain name, but descriptively, as in brandnameusedcars.com. Unlike the cases where the defendant uses Trademark.com, these more descriptive uses do not block the use of the trademark in a domain name by the owner of a famous mark. One can argue that, at least as far as dilution is concerned, they do not interfere with the distinctiveness of the mark, since it is used to identify the actual mark owner. They are also different from the critical commentary sites in that there is no aura of tarnishment. One could portray these as nominative fair uses of the mark, meaning a use of a trademark necessary to indicate something about the products portrayed on the Web site. A fair nominative use of a famous mark would be excluded from liability by Section 43(c)(3)(A).The Seventh Circuit apparently accepted this rationale in Ty, Inc. v. Perryman. Defendant Ruth Perryman sold secondhand beanbag toys, mostly Beanie Babies made by Ty, which she advertised on her Web site, www.bargainbeanies.com. The court likened her use of Ty’s Beanies mark to a used automobile dealer advertising the brands of used vehicles it sold. And it stated clearly that such uses were not forbidden by dilution laws. Indeed, the court speculated that Ty’s real fear was that others’ use of Beanies in this manner would render its mark generic—and the court apparently did not believe that a finding of dilution was the proper remedy for that problem.However, pre-TDRA courts did not always follow that path. In Paccar, Inc. v. Telescan Technologies, L.L.C., the plaintiff was the owner of the marks Peterbilt and Kenworth, used on trucks and truck parts. The defendant operated a Web site, www.truckscars.com, used for locating new and used trucks. However, the defendant also registered domains that contained the trademarks of specific manufacturers, including Peterbilt and Kenworth. These domains included peterbiltusedtrucks.com, peterbiltdealers.com, kenworthusedtrucks.com, and kenworthtruckdealers.com. The court found both infringement and dilution, and rejected a “fair use” argument. Regarding dilution, the court asserted that the use of the marks put the plaintiff “at the mercy” of the defendant. The court thus viewed dilution as protecting control of the mark by the mark owner, not simply as protection against misdirection of potential customers. Post-TDRA, this case would be criticized on at least two grounds. First, the TDRA clearly limits dilution to uses that satisfy the definition of either blurring or tarnishment. This sort of use may not cause harm to the distinctiveness or reputation of the famous mark and, therefore, would not fit the statutory definition of dilution. Second, the expanded “fair use” exclusion of the TDRA expressly excludes descriptive and nominative fair uses from dilution liability. Cases like this would be likely candidates for that exclusion.6.Use of Marks in MetatagsAnother way in which trademarks may be used on the Internet is in metatags. Metatags are codes in a Web page that are invisible to a user, but that can be seen by search programs (search engines) to find Web pages relevant to an Internet user’s search terms. Normally, metatags contain information relating to the content of a Web page. A store that sells products produced by others might want to indicate that fact by inserting a metatag into its Web page containing the brand name of the products so that Internet users looking for sellers of those products could find the store. Thus, a toy store might want to include “Mattel” in a metatag to indicate that it sells Mattel toys. Obviously, metatags also can be used simply to attract attention, even where the Web page is completely unrelated to the metatag. When the second user has a legitimate purpose for using the metatag, courts have found that to be a fair use and not actionable. If the use were completely unrelated, or used for an unscrupulous purpose, such as directing people to a pornographic Web site, then this misuse of a metatag could be held actionable as an infringement under a theory of “initial interest” confusion. Conceivably, such misuse could be actionable as tarnishment if the metatag directs people who input the mark into a search engine to an unsavory Web site. A blurring claim for misuse of a metatag would be dubious, however. A misdirected user is not likely to think that a seemingly unconnected Web page is somehow a second source for goods or services under the famous mark. Thus, neither the uniqueness nor the selling power of the mark is undermined.7.Hyperlinks and TarnishmentIn some cases, trademark owners have complained about the use of hyperlinks to other Web sites contained on Web sites using the famous mark. The famous mark owners have claimed—though not necessarily successfully—that hyperlinks to undesirable Web sites cause tarnishment of the mark.In Voice-Tel Enterprises, Inc. v. JOBA, Inc., Voice-Tel complained that a Voice-Tel franchisee’s Web site contained a hyperlink to a pornographic Web site. (The franchisee’s Web site was using the Voice-Tel mark legitimately.) However, the link originally led to another, perfectly legitimate, business which went bankrupt and which, without the defendant’s knowledge, sold its domain name to a pornography site. The court rejected a dilution claim, holding that to allow the claim “would give the trademark owner far more protection than was intended by the statute.”I.Contributory DilutionVery few courts have discussed the idea of contributory dilution. Contributory dilution refers to a dilution claim against a defendant who is not the one making the diluting use of a distinctive mark, but rather is a person who encourages and/or facilitates another’s dilution of a distinctive mark. In trademark infringement doctrine, there is an established body of precedent permitting claims for contributory infringement. Generally speaking, liability for contributory infringement requires either directly inciting infringement, such as suggesting that another commit an infringing act, or facilitating the infringement by either supplying the means or not taking action to halt the infringement by another when one knows or should know that the other is infringing.However, the notion of contributory liability does not translate easily where the claim is for dilution. The TDRA limits its application to a “use of a mark or trade name in commerce that is likely to cause dilution by blurring or dilution by tarnishment of the famous mark.” Moreover, absent actual encouragement to commit dilution, the ability of an alleged facilitator to know whether someone to whom it supplies goods or services is likely to be causing dilution under federal law is questionable. It requires both a knowledge that the plaintiff’s mark is “famous,” which would often be difficult to determine, and that consumers, while not confused, are likely to make an association that will harm the famous mark’s distinctiveness. Imposing this form of secondary liability could seriously impair a variety of useful commercial activity.On the other hand, an argument can be made that an injunction should be available against those who encourage or facilitate diluting uses as a necessary adjunct to stopping the use itself. No significant precedent under the FTDA or TDRA has yet emerged. However, one court did deny leave to amend a complaint to add such a claim on the grounds that it would be futile. In Tiffany (NJ), Inc. v. eBay, Inc., the court noted the lack of precedent for such a claim, and expressed skepticism about whether such a claim exists. It also held that, even if such a claim existed, the plaintiff would not have been able to recover on the facts of the case. Upholding the district court, the Second Circuit supported this conclusion, although in dicta. The few cases mentioning the issue under state law have not been consistent and do not contain cogent analyses.Although there is no apparent appellate precedent, a few district courts have recognized a contributory dilution claim. A contributory dilution claim was permitted in Kegan v. Apple Computer, Inc., where the plaintiff, a publisher of computer guides, including MacGuide, sued Apple, which used the terms “Apple Guide” and “Macintosh Guide” to denote software used to assist computer users. The plaintiff claimed both direct dilution and contributory dilution; the latter claim was based on allegations that Apple encouraged others to develop Guide software. The court denied Apple’s summary judgment motion as to the dilution claims, both direct and contributory, asserting that issues of fact existed as to the distinctiveness of the plaintiff’s mark and a likelihood of blurring. However, it apparently assumed, there being no real discussion of the issue, that a claim for contributory dilution existed under Illinois law. Given the absence of any real discussion of the issue, Kegan is at best weak support for the existence of such a claim. Another federal court followed Kegan, calling a contributory claim “novel” yet “entirely plausible,” though without a serious discussion of the claim. And another district court denied a motion to dismiss a contributory dilution claim, relying in part on the fact that no appellate court had expressly rejected such a claim, plus the Kegan opinion.One court allowed a contributory dilution claim to go forward, citing several cases including Kegan, and stating that “willful blindness” is sufficient to permit such a claim. Another court allowed such a claim against a flea market operator. However, the same court dismissed a contributory claim against the owners of the property (who leased to the market operator), holding that merely owning property on which dilution occurred was not sufficient to support contributory liability.By contrast, a claim for contributory dilution under California law was dismissed, with almost no discussion, in Academy of Motion Picture Arts & Sciences v. Network Solutions, Inc. However, the lack of discussion of the issue in this case and Kegan makes them questionable precedent for any broad statement that such a claim should or should not exist.In another case, a federal district court granted summary judgment for the defendant on claims of contributory infringement and dilution, but without discussing whether a contributory dilution claim is theoretically valid.If such a claim exists, one would expect that the requirements would be no less stringent than for contributory infringement. Thus, some sort of actual or imputed knowledge of the other’s dilution should be required.III. The Required Level of Similarity toDemonstrate DilutionA.The Requisite Level of SimilarityThe concepts of dilution and similarity of marks are very much intertwined. Dilution is the end result of a second mark that calls to mind another mark, leading to some kind of harm to the other mark—either a loss of distinctiveness or harm to its reputation. Indeed, the definitions of blurring and tarnishment in the TDRA state that blurring and tarnishment occur when the famous mark’s distinctiveness or reputation is harmed by an association that results from similarity. Clearly, unless the second mark is similar to the other mark, it is unlikely that the second mark will trigger a recall of the other mark. But how similar must the second mark be to cause dilution (or be likely to cause dilution)? And is there a minimum threshold of similarity below which a dilution claim will not be allowed to proceed?Prior to the TDRA, courts had a fairly consistent answer to those questions. A dilution claim could not succeed without a requisite level of similarity between the marks. And the required level was, at least in theory, fairly high. Most courts required that the two marks be “substantially” similar. The Ninth Circuit seemed to go further, requiring the marks to be “identical or nearly identical.” However, in practice, the Ninth Circuit’s standard was not as exacting as its rubric suggested.Even after the TDRA, courts continued to use the substantial similarity, or nearly identical, test—that is, until late 2009. That was when the Second Circuit, in Starbucks Corp. v. Wolfe’s Borough Coffee, Inc., held that, in a blurring case, substantial similarity was no longer a threshold requirement under the TDRA. The court based its holding on the listing of factors for determining blurring, which includes “degree of similarity,” but without specifying “substantial” similarity. The court held that this indicated Congress’s desire to make similarity just one factor in the analysis and not a threshold requirement. The Ninth Circuit followed suit in 2011, essentially adopting the Second Circuit’s reasoning. These two cases have left the issue of similarity in an uncertain state, especially regarding blurring cases.Even if other courts follow suit, the reasoning of these cases was specific to the definition of blurring under the TDRA. That leaves open the question of the proper level of similarity to demonstrate likelihood of tarnishment. Unlike the definition of blurring, the definition of tarnishment contains no guiding factors. Thus, although the definition of tarnishment does not contain any specific threshold of similarity, there is no basis to believe that Congress intended to make similarity merely a factor, in the manner that the blurring factors arguably do for blurring. The original FTDA contained no specific reference to any level of similarity—the substantial similarity test was a judicial interpretation of Congress’s intent. Indeed, using the reasoning of Starbucks and Levi Strauss, one may conclude that the absence of any specific factors incorporating similarity was a deliberate decision on Congress’s part. In the absence of any indication in the statute or legislative history to the contrary, there is little basis to infer that Congress intended to change the applicable standard merely by including the word “similarity” in the definition of tarnishment.Even if a court were inclined to revisit the issue for tarnishment, there is good reason to continue to use the substantial similarity standard. Tarnishment by definition involves a use that harms the reputation of another’s mark. That sort of harm indicates that viewers of the second mark make a fairly strong connection between the marks—more than a mere calling to mind. There must be a strong association for a second mark to harm the reputation of the first. This requires a higher level of similarity than would a simple loss of uniqueness. (It is not a search cost problem but a true associational problem.) As discussed in Section II.E. above, outside of sexually oriented cases, many courts have been reluctant to assume that tarnishment will occur in the absence of any evidentiary showing that harm is likely to occur. This, too, counsels against a simplistic decision to change the standard. Finally, since many tarnishment cases implicate the communicative value of trademarks as other than mere selling vehicles, courts should not lightly reduce the burden on the plaintiffs to demonstrate a real likelihood of harm, as opposed to a possibility. Obviously, the main argument in favor of changing the tarnishment standard is consistency across the dilution statute. But, given that tarnishment expresses a harm to reputation, not distinctiveness, a different standard is not necessarily illogical.Nevertheless, one district court has held that the requisite level of similarity for tarnishment is the same as that for blurring. The court reasoned that Congress intended the word “similarity” in the definitions of blurring and tarnishment to have the same meaning.B.Illustrating Similarity: Examples from the Case LawWhether the standard is “substantial” similarity or whether it is simply “degree of similarity,” courts will have to determine what level of similarity exists between the two marks at issue. Unfortunately, even a standard like “substantial” similarity or “identical or nearly identical” does not clearly identify which situations meet the standard. Thus, one must look at case law to find appropriate analogies. Most of the recent case law predates the recent questioning of the substantial similarity test, but it gives clues to the proper methodology in determining similarity.In two cases, courts examining the similarity of marks in the genre of well-known drugs analyzed this issue in roughly the same manner. In one case, the court found the mark Tempanol to be substantially similar to Tylenol. The court cited similarities in spelling, sound (both begin with T and end with “nol”), and number of syllables. The court found this sufficient to show “a mental association” between the marks. In the other case, the defendant’s Herbrozac mark was found to be substantially similar to Prozac. The court noted the identity of the “rozak” ending and the similarity of the “B” and “P” sounds before the ending. It characterized the defendant’s mark as “effectively stat[ing] ‘herbal Prozac.’ ”Two cases compared marks of the form “Greatest ____ on Earth” to the mark Greatest Show on Earth, and neither found substantial similarity. In the first case, the court found that The Greatest Bar on Earth was not substantially similar to The Greatest Show on Earth. The court found that although the “Greatest Show . . .” phrase is famous, it is made up of common terms. The key word “Show” in Ringling’s mark had been replaced with “Bar,” and the latter term overshadowed the rest of the phrase. In the second case, the court found that The Greatest Snow on Earth was not substantially similar to Ringling’s mark. As with the “Greatest Bar . . .” mark, the court cited the fact that the key words—snow and show—while similar in sound (and rhyme), were common words. Moreover, the court cited the dissimilarity of the contexts, stating that both phrases had strong descriptive effects and thus the context would dispel any tendency to conjure up a single mark.The Sixth Circuit examined the issue of similarity in AutoZone, Inc. v. Tandy Corp., a dilution case involving the marks Autozone and Powerzone that were used for different goods. The court first emphasized the difference between similarity in an infringement context and in a dilution context, concluding that “we require a plaintiff to demonstrate a higher degree of similarity than is necessary in infringement claims in order to prove that actual dilution has occurred.” The court also noted that several other courts had demanded a higher standard of similarity when considering dilution claims. Applying this standard to the instant case, the court noted:The two marks look different; they are written in different fonts and colors. They utilize different designs (slanted versus nonslanted tradename) and are meant to convey different messages. They also sound different because there is little acoustic similarity between AUTO and POWER. As a comparison, AUTOZONE and a hypothetical AUDIOZONE would have a higher level of acoustic similarity. Additionally, the common word between AUTOZONE and POWERZONE is pronounced second and is consequently deemphasized.The Fourth Circuit cited the AutoZone language when it dismissed a claim by the owner of the CareFirst mark against the owner of First Care. The court indicated that the two marks were not particularly similar even standing alone, and noted further that the CareFirst mark was always accompanied by a house mark (Blue Cross Blue Shield).In Louis Vuitton Malletier v. Dooney & Bourke, Inc., (a pre-Starbucks district court case in the Second Circuit) the court found that a Louis Vuitton handbag design, claimed as a trademark (the Multicolore Monogram design) was not substantially similar to the defendant’s competing design. The court referred to its confusion discussion, which also found the marks to be dissimilar.In Starbucks Corp. v. Wolfe’s Borough Coffee, Inc., although the Second Circuit remanded to the district court to reconsider its analysis of blurring, it did uphold the district court’s finding that the Starbucks mark and the Mr. Charbucks mark were only minimally similar. On remand, the district court reiterated its conclusion, noting that “contextual features,” such as the addition of “Mister” before Charbucks, justified its conclusion.In Hershey Co. v. Promotion in Motion, Inc., the court found that although there was “some similarity” between the mark SWISSKISS used on chocolate and the Hershey’s Kisses mark, it was not enough to favor a finding of dilution. The court declined to decide whether to use a “substantial similarity” test or a lesser test, such as the one used in Starbucks, but found insufficient similarity under either standard.Post-TDRA, the TTAB has had several opportunities to discuss similarity. In one case, the Board found that Citibank and Capital City Bank were not substantially similar, and the dilution claim failed. In a subsequent case, the TTAB found that The Other White Meat and The Other Red Meat were “highly similar”—without ever mentioning “substantial similarity”—and the dilution claim succeeded. In two other cases, one comparing Just Jesu It to the Nike mark Just Do It, and another comparing “Crackberry” to “BlackBerry,” the TTAB also found that the marks were sufficiently similar to conclude that the second user would conjure up the famous mark. In doing so, the TTAB considered “appearance, sound, connotation and commercial impression.” On the other hand, the Board found that Rolex and Roll-X were not particularly similar, despite their similar pronunciation.C.Trade Dress and SimilarityTrade dress cases provide particularly challenging situations for analyzing similarity. Word marks can be compared for sound, cadence, and context. However, packaging and product design rely on visual cues, some of which can be ambiguous. Indeed, as one court admitted, the test is often a “subjective ‘eyeball’” one. As with word marks, where the product design is copied, the issue is simple. In other situations, a more nuanced discussion is required.In Nabisco, Inc. v. PF Brands, Inc., the marks at issue were crackers, and the court rejected several arguments made against similarity. The defendant (the senior mark owner) made Goldfish crackers, in the shape of a goldfish. The plaintiff produced crackers based on characters in the CatDog cartoon shown on the Nickelodeon Television Network: a two-headed “catdog,” a bone, and a fish-shaped cracker were all packaged together. First, the court rejected the argument that the products’ packaging was so dissimilar that it dispelled any tendency of consumers to associate the two crackers. It cited the possibility of post-sale dilution, where snackers are presented with Nabisco’s offending crackers outside of the packaging. Second, Nabisco argued that only a quarter of its crackers were fish-shaped and, in context with the other crackers, would conjure the CatDog Show, not Goldfish crackers. However, the court believed that many consumers, perhaps unfamiliar with the cartoon, would not make that connection, and would associate the fish-shaped crackers in the Nabisco product with the famous Goldfish crackers.In Binney & Smith v. Rose Art Industries, the plaintiff claimed that the trade dress of its Crayola boxes, “a green chevron laid over a yellow background,” was diluted by the defendant’s trade dress for its competing products. The court found the defendant’s packaging, consisting of a green overlay on an angle on a yellow background, was similar to the plaintiff’s trade dress.In adidas America, Inc. v. Skechers USA, Inc., the Ninth Circuit affirmed the district court’s finding of dilution by blurring. Skechers used a three-stripe design on its tennis shoes and, although there were some dissimilarities (in its confusion analysis the court noted “a difference in the thickness of the stripes, the inclusion of a strip between the three stripes on the Cross Court, and the fact that the stripes do not continue to the sole of the shoe”), the court held that these were not sufficient to overcome the strength of the three-stripes mark. (In the confusion analysis, which was largely the basis for the dilution analysis, the court also cited defendant’s somewhat questionable intent in adopting this design.) In Deere & Co. v. FIMCO, Inc., the court found that the green and yellow colors used by defendant on its farm vehicle accessories were very similar to those used by Deere & Co. on its tractors.In Gucci America, Inc. v. Guess?, Inc., the court held a bench trial on claims of infringement and dilution involving the trade dress of various products made by Gucci. The court separately discussed the similarity of the printed designs on Gucci and Guess products. As to one of the designs, the court concluded that the two were similar based on the general visual appearance plus the marked color similarities. However, as to another design, the differences in color led to the conclusion that the designs were dissimilar. The court further found that a “Block G” design used by Guess was not similar to a “Stylized G” design used by Gucci. The court’s conclusions as to likelihood of dilution mirrored its conclusions concerning similarity.One trade dress case where similarity was lacking was American Cyanamid Co. v. Nutraceutical Corp. Both parties made vitamins and herbal supplements. At issue were the labels used by each. The plaintiff’s label contained adjacent, horizontal-colored rectangles with colors that gradually changed from blue to red, as in a rainbow. The defendant’s labels also had colored bands, some of which ran through the spectrum of colors from blue to red and back to blue. The court found the overall impression of the labels not similar in its analysis of possible confusion. Apart from the use of a spectrum of colors in both labels, the court found the marks not to be similar.Hershey Foods Corp. v. Mars, Inc. presented a situation analogous to American Cyanamid. The plaintiff claimed that its trade dress for peanut butter cups (Reese’s) was diluted by Mars’ packaging for its M&M peanut candies. The key elements claimed by the plaintiff included an orange background with yellow lettering surrounded in brown. Mars’ trade dress also used an orange background, but contained solid brown lettering, with its logo in a different part of the package than Reese’s logo. Although they used similar colors, the court viewed “[t]he overall look” as “different.”Hershey also considered the problem of how much of the trade dress should be viewed as relevant. The defendant argued that the court should view the package as a whole, including the logo and the trade name. The court acknowledged the relevance of the whole package in a confusion analysis, but was skeptical of the relevance of the logos and trade names to a dilution analysis, where the issue is one of mental association, not whether the trade dress as a whole dispels confusion. Because of its determination that there was no similarity, even without the logos, the court deemed it unnecessary to decide the issue.The issue Hershey raised but did not decide is an important one. The court’s assumption that the logo itself would not be relevant, since different names might dispel confusion, but would not avoid a mental association, seems logical. However, it is an untested assumption about consumer behavior. It may be that, faced with vastly different logos on otherwise similar packaging, consumers will focus on the logo, not the packaging, particularly if the logo is well known. In that case, consumers may not make the requisite mental association, since they do not view the packaging in the abstract. Marketing surveys testing this assumption would be useful to courts faced with packaging issues, especially where a logo or trade name is a prominent feature of the package.An interesting twist on the trade dress problem is provided by Landscape Forms, Inc. v. Columbia Cascade Co. This case involved two sellers of outdoor furniture, where the trademarks in question were the pieces of furniture. Although the court acknowledged that the parties’ furniture lines were “substantially similar,” it nevertheless found the marks not to be sufficiently similar for dilution. The reason is that the furniture was normally purchased through catalogs and other publications that prominently displayed each party’s trade name. The court determined that these contexts would prevent potential customers, who were sophisticated landscape architects, “from making the mental association between the parties’ respective trade dress that is required to show blurring.”The importance of viewing the mark in context is also illustrated by Elvis Presley Enterprises, Inc. v. Capece. The district court ruled that the defendant’s use of The Velvet Elvis as the name of its bar was not similar to the plaintiff’s Elvis Presley marks. The court viewed the defendant’s bar as being a symbolic parody of 1960s and 1970s bad taste, and saw The Velvet Elvis as tied to that overall parody, rather than to Elvis Presley. On appeal, the Fifth Circuit reversed but did not examine the court’s conclusions as to dilution. However, it did criticize the district court’s conclusions about similarity in the confusion context. Thus, the district court’s conclusions about lack of similarity for dilution must be viewed with skepticism. Nevertheless, the court’s approach seems correct—to view the use of the mark in context when determining similarity—even if its application to the facts of the case was questionable.In most cases, the issue of similarity probably will not be a difficult one. The Internet cases, for example, tend to be about marks that are virtually identical. And many other cases involve marks so similar that the issue is not a serious problem. Thus, the cases discussed here can only provide a general framework for the discussion. However, they state important principles. The marks as a whole should be examined, and not overly parsed for either similarities or differences. Context should not be ignored, because it is the initial mental reaction of consumers to the marks in the marketplace that is being measured. And, with respect to trade dress, care must be taken to determine what is relevant to consumers about the marks in question.IV. The Need for a Trademark UseThe TDRA grants a claim to the owner of a famous mark against one who “commences use of a mark or trade name in commerce that is likely to cause dilution.” One would expect, then, that the only uses that are actionable are uses as trademarks, not just any commercial use of the famous mark. Furthermore, the definition of dilution by blurring states that it is an “association arising from the similarity between a mark or trade name and a famous mark,” and the definition of dilution by tarnishment is an “association arising from the similarity between a mark or trade name and a famous mark that harms the reputation of the famous mark.” Finally, Section 43(c)(5), which, under limited circumstances, permits the owner of the famous mark to obtain monetary and other non-injunctive relief, has as its first requirement that “the mark or trade name that is likely to cause dilution by blurring or dilution by tarnishment was first used in commerce by the person against whom the injunction is sought after October 6, 2006.” These provisions of the TDRA seem to reinforce the idea that only a use as a trademark by the offending party is actionable as dilution.One part of the TDRA creates some ambiguity. The “fair use” exclusion in Section 43(c)(3)(A) applies only to uses that are not trademark uses (uses “other than as a designation of source for the [second user’s] own goods or services”). A trademark use requirement would appear to make this exclusion superfluous. The legislative history of the TDRA is unclear on this point. As originally introduced in February 2005, H.R. 683 gave a cause of action against one who “commences use of a mark or trade name in commerce as a designation of source of the person’s goods or services.” Dilution by blurring was defined as an “association arising from the similarity between a designation of source and a famous mark that impairs the distinctiveness of the famous mark.” The use of the “designation of source” language presumably indicated the need for the diluting use to be a trademark use. However, the American Intellectual Property Law Association (AIPLA) objected to the phrase “designation of source” and the House Committee eliminated the phrase “designation of source of the person’s goods or services.” The AIPLA believed that the “designation of source” language would limit the use of the TDRA and that the concerns about imposing liability for nontrademark uses should be handled by the TDRA’s exclusions from liability in Section 43(c)(3). However, when the House altered H.R. 683 to eliminate the “designation of source” language, it did not indicate whether it was adopting the AIPLA’s view of the change. It is uncertain how courts will use this legislative history. If the Supreme Court’s Moseley decision is any indication, then the history will not be that significant; in Moseley, the Court looked at the language of the statute and gave short shrift to the legislative history.Furthermore, the three provisions cited earlier, namely the definitions of blurring and tarnishment, and the section on non-monetary relief all point strongly in the direction of requiring a trademark use. The definitions call for a comparison between a mark or trade name and the famous mark, indicating that the two are not the same—i.e., the offending use must be a “mark or trade name.” Section 43(c)(5)(A) is even more specific, referring to “the mark or trade name” that is the likely cause of dilution, nor merely to a “use” that is likely to cause dilution.The original FTDA also had language indicating a need for a trademark use. Section 43(c)(1) originally permitted an injunction “against another person’s commercial use in commerce of a mark or trade name [that causes dilution].” Moreover, such a requirement is consistent with the background against which the FTDA was enacted. Perhaps the most notable aspect of this background is the Restatement (Third) of Unfair Competition. Section 25 of the Restatement is the dilution section. Two portions of Section 25 are pertinent to the type of use required. Subsection (1) imposes liability “if the actor uses such a designation in a manner that is likely to associate the other’s [distinctive] mark with the goods, services, or business of the actor.” In other words, the designation must be used as a source identifier, or trademark. Subsection (2) limits liability where the use is “not in a manner that is likely to associate the other’s mark with the goods, services or business of the actor.” This also refers to a trademark use. The comments to the Restatement reinforce the requirement of a trademark use. If the Restatement’s view is the proper interpretation of the TDRA, then many uses often associated with dilution, particularly parodies, but also including many Internet domain names that do not consist solely of the famous mark, will not be subject to liability.However, it has been held that whether the defendant is making a trademark use should be viewed objectively, and should not depend on the defendant’s subjective intent.The trademark use requirement was accepted by the Fifth Circuit in National Business Forms & Printing, Inc. v. Ford Motor Co. The court held that a printer of decals that contained Ford’s trademarks did not “use” the marks within the meaning of the TDRA. Specifically, the court held that the printer was not liable for dilution because it did not use Ford’s marks “in identifying its own goods or services.”Similarly, a district court in New York dismissed a dilution-by-tarnishment claim, stating, among other things, that because defendant’s use of plaintiff’s mark was to refer to plaintiff, there could be no dilution. In Pellegrino v. Epic Games, Inc., the court dismissed a dilution claim because there was no trademark use. And in Lions Gate Entertainment, Inc. v. TD Ameritrade Holding Corp., the court, while denying a motion to dismiss, appeared to accept the need for a trademark use, but found that the complaint adequately pled such a use.However, in Rosetta Stone Ltd. v. Google, Inc., the Fourth Circuit reversed a district court ruling that had dismissed a dilution claim for lack of a trademark use. The court reasoned that the fair use exclusion’s requirement of a non-trademark use would be rendered superfluous by a trademark use requirement for dilution. The court did not address the other statutory provisions seemingly requiring trademark uses.A few district court opinions predating the TDRA denied dilution by blurring claims on the grounds that the defendant’s use was not a trademark use. In one, a parody of Barney the Dinosaur by The Famous Chicken was said not to constitute blurring because it was not a trademark use. In a third case, a search engine was programmed to display certain advertisements when the terms “playboy” or “playmate” were researched. This, too, was held not to be a trademark use. In the third case, the court refused a preliminary injunction against a defendant who converted Ford cars into limousines and left certain original car badges on the cars. The court held that this was not actionable because it was not a trademark use of Ford’s trademarks. This issue was also raised in Avery Dennison Corp. v. Sumpton, where the court found that a “commercial use” under the FTDA “requires the defendant to be using the trademark as a trademark, capitalizing on its trademark status.”A similar result was reached in Interactive Products Corp. v. a2z Mobile Office Solutions, Inc. The court held that the defendant’s use of the plaintiff’s Lap Traveler mark in the “post-domain path” of the defendant’s Internet URL was not a trademark use of the mark and thus did not violate the FTDA.In Ford Motor Co. v. Greatdomains.com, Inc., one defendant provided an online forum for the purchase and sale of domain names and other defendants registered various domain names that incorporated the plaintiff’s trademarks. For the court, the key issue was whether the defendants had made a “use” of the marks within the meaning of the FTDA. The court held that a trademark use was required under the FTDA, meaning a use “in connection with goods and services.” Applying that requirement, the court found that the sale of domain names incorporating trademarks does not constitute use in connection with goods and services, particularly where the registrant defendants had not operated Web pages using the domain names in question. In DaimlerChrysler AG v. Bloom, the Eighth Circuit held that licensing a vanity telephone number that spelled 1-800-MERCEDES to Mercedes dealers, without any advertising of the name by the licensor, did not constitute a “use” for purposes of dilution or infringement.The Second Circuit raised the issue in Nabisco, Inc. v. PF Brands, Inc., calling it “a complicated question.” However, the court did not need to decide the issue because it found that Nabisco’s crackers constituted a trademark use of the goldfish shape.Clearly, despite the Restatement’s admonishment to the contrary, many courts have imposed liability in tarnishment cases for nontrademark uses. However, in blurring cases a requirement of trademark use is easily supportable. Blurring assumes that the second use will cause consumers to associate more than one source with a particular mark. For that to occur, it is logical that the offending use should be as a trademark. That would eliminate certain possible diluting uses, notably those constituting genericide. However, requiring a trademark use causes interpretive problems regarding the statutory exclusions from liability. The “fair use” exclusion is expressly limited to non-trademark uses, and the noncommercial use exclusion has been held not to apply to trademark uses. V. Proving Dilution: Case Law andSuggested ApproachesA.IntroductionDilution liability can be established by showing either actual dilution or likelihood of dilution. One aspect of trademark dilution that has received relatively little attention from courts is the proper methodology for proving that dilution has actually occurred or that the occurrence of dilution is likely (as opposed to simply possible). In part, this undoubtedly stems from the difficult task it presents. Unlike confusion, dilution is often described as a gradual process, where the damage is incremental and hard to measure. Indeed, the most influential multifactor tests for dilution used by courts prior to the TDRA did not even list actual dilution as a factor. Although the TDRA contains factors for use by courts in determining blurring, they do not necessarily lead to easy conclusions about the real issue—whether the defendant’s use harms or is likely to harm the distinctiveness (or reputation, if tarnishment is claimed) of the famous mark. They are, at best, indirect measures. However, the TDRA makes “actual association between the mark or trade name and the famous mark” a factor in the determination of dilution by blurring. Therefore, practitioners can no longer ignore the need to provide such evidence.B.Case Law Discussing Evidence of DilutionA relatively small number of cases has discussed evidence of dilution. Thus, cases can offer only partial guidance to practitioners. In infringement cases, the most common evidence of likelihood of confusion is a survey. In dilution cases, litigants also have introduced surveys to assist their cases. In many situations, the surveys are used to demonstrate mark strength, in an effort to support a claim that the mark is famous. However, some cases have used surveys, especially associational surveys, as a way of indicating dilution. The Supreme Court’s Moseley decision in 2003 forced litigators to attempt to show actual dilution, making more direct evidence of dilution critical, at least until the TDRA was enacted. However, few cases before or after Moseley have attempted to demonstrate dilution directly, as opposed to using association as an indirect means of indicating the likelihood of dilution. When the TDRA changed the operative standard from actual to likelihood of dilution, it made evidence of actual association a factor in the analysis of blurring, thus encouraging the use of survey evidence. In those cases in which surveys have been introduced, the results have been mixed.In an early FTDA case, Wawa, Inc. v. Haaf, the court credited a survey indicating that 29 percent of the respondents associated the defendant’s Haha market with Wawa. The Wawa survey appears to have been an “association” survey, where respondents were shown a picture of a Haha market and asked whether they associated it with anything else. Apparently, the court believed that such association was sufficient to show dilution. In another case, a survey that found a 14 percent association was held to be insufficient to demonstrate dilution. That same court stated that dilution claims should require a greater showing of association than confusion claims because dilution does not require a showing of confusion. In a post-TDRA case, a court cited a survey showing a 73 percent rate of association in support of a finding of dilution.Kellogg Co. v. Exxon Mobil Corp. presented an unusual situation in which a survey was offered. Kellogg, owner of the Tony the Tiger mark, sued Exxon, claiming that Exxon’s use of a cartoon tiger to promote its products diluted the Tony the Tiger mark. In the course of the litigation, Kellogg abandoned its claim against Exxon’s use of the tiger in connection with petroleum products, and chose to focus on Exxon’s use of the cartoon tiger in connection with sales of food and beverage items at convenience stores associated with service stations. In opposition to Exxon’s summary judgment motion on its dilution claims, Kellogg submitted a survey in which customers were shown two beverage containers with Exxon’s cartoon tiger and an Exxon gasoline advertisement. The customers were asked what company or brand came to mind, and then what other companies or brands came to mind. Almost 70 percent of the respondents mentioned Kellogg as a company that they associated with the containers and/or the advertisement. Exxon, however, pointed out that essentially the same percentage of people thought of Kellogg when shown Exxon’s gasoline advertisements as thought of Kellogg when shown the beverage containers, and that it was unclear whether any dilution was caused by the use on beverages as opposed to the gasoline advertisements. Ultimately, the court found that Kellogg raised an issue of fact by its assertion that the Exxon cartoon tiger was used “‘primarily, if not exclusively … to promote foods, beverages, and convenience stores—not gasoline.’” However, the court noted that this survey went only to the issue of blurring, not tarnishment, which was also part of Kellogg’s suit.In a few other cases, the courts rejected surveys designed to demonstrate dilution. The first was Ringling Bros.-Barnum & Bailey Combined Shows, Inc. v. Utah Division of Travel Development, where the plaintiff complained of the defendant’s use of the phrase Greatest Snow on Earth in tourism advertisements. The plaintiff introduced a nationwide survey in which respondents were asked to fill in the following: “The Greatest ____ on Earth.” The results showed that a total of 46 percent of Utah respondents completed the blank with “Show”: 25 percent filled it in with the word “Show,” and another 21 percent filled it in with “Show” and with “Snow.” An additional 24 percent of the respondents filled in the blank only with the word “Snow.” By comparison, 41 percent of non-Utah residents surveyed filled in the blank with “Show,” and virtually none filled it in with “Snow.”The court found these results inadequate to show dilution by blurring. Primarily, the court’s rejection focused on two factors. First, the survey question did not test respondents’ reaction to the actual mark used by the plaintiff, but to a portion of the mark. Second, the court pointed out that the respondents who filled in the blank with “Snow” almost invariably associated it with Utah, and those who filled in with “Show” almost invariably associated it with Ringling. Those filling in both responses correctly separated the two sources, associating “Snow” with Utah and “Show” with Ringling. Thus, the court found no evidence that Utah’s phrase had done any harm to the source-identifying capacity of Ringling’s mark. In contrast to the Wawa court, the Ringling court was skeptical that a demonstration that the “incomplete” phrase called to mind a second source was evidence of dilution. Indeed, the Supreme Court later cited this survey in support of its assertion that association alone does not necessarily demonstrate dilution.In a post-Moseley, pre-TDRA decision, the Fourth Circuit rejected a survey purporting to demonstrate confusion on the grounds that confusion does not demonstrate dilution as defined by the FTDA. A district court in New York, post-TDRA, also rejected a survey that showed confusion on the grounds that it did not measure blurring. Another district court also found evidence of possible confusion insufficient to show actual dilution, finding no diminution of the mark’s source-identifying power.Another dilution survey was rejected in Hershey Foods Corp. v. Mars, Inc. The survey in this case attempted to show that the plaintiff’s trade dress (packaging) was diluted by the defendant’s trade dress (packaging). The plaintiff’s expert surveyed only people who purchased certain chocolate and/or peanut butter candy. It was a controlled survey, using representations of various candies as controls. Although the court held that the plaintiff’s trade dress was not famous, it proceeded to analyze the elements of dilution. The survey tested responses to representations of the trade dress of various peanut-related candies. It did not use the actual packaging of the candies. Instead, it used recreations of the major design elements of the packages, with brand names and logos eliminated. The survey asked respondents if they knew who made the product represented by the packaging. It showed that over half of the respondents mistakenly identified the M&Ms representation as coming from Reese’s, with many of the respondents citing the color used in the representation as the reason for the mistake. The court rejected two challenges to this survey, but accepted a third challenge. It rejected the contention that the survey should have used actual packaging. The court found that trade dress can consist of elements that do not include an entire packaging scheme. A survey that properly tests this could be accepted. The court also rejected an argument that the plaintiff’s survey was designed to show confusion, not dilution. Although it appears that the survey design was directed at confusion, the court believed that it could also measure dilution, because respondents were permitted to answer that they did not know whose packaging was being represented. However, the court accepted the defendant’s third objection, that the survey did not show all of the elements of M&M’s trade dress to respondents. The court pointed out that the M&M logo in the actual packaging used a different type style and was located in a different place in the representation than on the actual package. Moreover, some of the control pictures retained various design elements whose equivalents were eliminated from the M&M package representation. Thus, the court concluded that respondents may have been misled and discounted the survey.And, in National Football League Properties, Inc. v. Prostyle, Inc., the court also found survey evidence wanting. This case involved the sale of unauthorized Green Bay Packers shirts. After the court rejected the plaintiff’s expert’s survey as flawed, he did a second survey to show dilution. The survey consisted of one question, asking respondents what they thought of when shown a T-shirt with green and gold colors and the words “Green Bay Football” or “Green Bay P.” The court rejected the survey, citing the complete absence of controls.Although hardly a representative group of cases, these cases indicate some of the difficulties facing litigators using surveys. First, one must be aware that the defendant will have experts who will examine the methodology carefully. Second, the difficulty courts have in defining dilution creates problems for survey design. It is hard to design a survey when the goal is not well stated. Third, proper controls are essential to an acceptable survey.Even if an association survey is admissible, it may not be sufficient to demonstrate the connection between association and likely harm. In Starbucks Corp. v. Wolfe’s Borough Coffee, Inc., the court found a survey showing a 30.5 percent association between Charbucks and Starbucks to favor the plaintiff, but only “minimally.” The major problem with the survey, which was conducted by telephone, was that it did not appropriately account for the context in which defendant used the term “Charbucks”—the survey did not include (and could not include) the visual context or the word “Mister” before Charbucks or “blend” after Charbucks.Similarly, in Hershey Co. v. Promotion in Motion, Inc., the court was skeptical of a survey that showed respondents the defendant’s mark on a blank card, instead of using a picture of an actual product, or at least the mark in the colors actually used on the product. The court criticized this as not effectively replicating market conditions, and found it not reliable. In addition, the court noted that the association rate of about one-third was lower than other cases.However, in Radiance Foundation, Inc. v. NAACP, the court credited a survey in support of its finding of dilution by tarnishment. The survey found a 42 percent rate of association of “National Association for the Abortion of Colored People” with the “National Association for the Advancement of Colored People” and a 37 percent rate of tarnishment. However, the survey arrived at its rate of tarnishment by assuming that those respondents who associated the two marks but did not, unaided, characterize the unauthorized use as parody or criticism, saw it as simply tarnishing. Although this is evidence of association, it is less clear that it should have been credited as evidence of tarnishment, as it did not measure whether the respondents’ views of the NAACP mark changed after seeing the Radiance version. In any event, the district court’s decision was reversed on the grounds that the use in question fell within the statutory exclusions.In Deere & Co. v. FIMCO, Inc., the court accepted surveys of users of farm equipment that showed photographs of defendant’s equipment and asked respondents what, if any, companies came to mind. The surveys found 38 percent and 43 percent association with Deere (the surveys were designed to measure association with the green and yellow colors of Deere equipment). These surveys were used to demonstrate actual association.In VIP Products, LLC v. Jack Daniel’s Properties, Inc., the court accepted one survey that indicated that consumers would have negative associations with the Jack Daniel’s brand as a result of exposure to the offending “Bad Spaniels” dog toys. The same court rejected a rebuttal survey because the respondents were told in advance that the toys were intended to be a parody.Surveys are not the only possible evidence of dilution. Courts have permitted a marketing expert to testify about the likely effect of one mark on consumer perceptions of a famous mark in a dilution case. One example is the remand of VIP Products LLC v Jack Daniel’s Properties, Inc. After a bench trial, the court credited Jack Daniel’s expert, who relied entirely on consumer marketing research to conclude that the type of use made in that case was likely to cause harm to the reputation of the Jack Daniel’s brand—the expert did not present any direct evidence, such as a survey aimed at the particular use in question.Another court permitted defendant’s expert to rebut plaintiff’s evidence without requiring the expert to conduct independent marketing research, which the court held was not required.However, it should be noted that, in line with precedent in confusion cases, courts have stated that neither evidence of actual dilution, nor evidence of actual association, is essential to a dilution claim. On the other hand, the effect of the absence of such evidence will vary, depending on the circumstances of the case. In a post-TDRA case denying a dilution claim, a district court pointedly noted that the plaintiff had failed to produce a survey when it would have been logical to have one. In another post-TDRA case, a district court upheld a jury finding of actual dilution (to support a damages claim) based on a combination of the testimony of an employee of the plaintiff and blog entries indicating unfavorable reactions to the defendant’s use of the plaintiff’s marks (belonging to the University of Kansas).C.Dilution Surveys: Characteristics and Pitfalls1.Designing a Survey: Determining What to MeasureClearly, the most common evidence of actual confusion has been a survey. The case law and literature discussing confusion surveys is rather extensive. The advent of a federal dilution statute has stimulated discussion of methods of measuring dilution in published commentary and empirical studies. This literature contains several useful suggestions for designing dilution surveys.However, before one can design an appropriate survey, one must first ask what precisely the survey is supposed to measure. One commentator suggested that a brand association survey would be appropriate evidence of dilution—i.e., one that shows that consumers make a mental association between the junior user and the famous mark. However, the value of such surveys should not be overstated. The Supreme Court has stated that “mental association” is not sufficient to show dilution, and further indicated that, in its view, dilution is shown when, after viewing the junior use, the consumer’s view of the famous mark is altered or diminished in some way. This is consistent with the TDRA’s definition of dilution, which states that dilution is the result of an association that harms the distinctiveness or the reputation of a famous mark. Using this construct of dilution, a survey that attempts to demonstrate dilution, not just association, would have to test a consumer’s association with the mark before and after seeing the offending use. On the other hand, under the TDRA, evidence of mental association is clearly relevant, even if not sufficient. The definition of dilution by blurring states that it is the loss of distinctiveness of the famous mark that is the cornerstone of the issue. Showing a mental association is one step in the process of demonstrating a loss of distinctiveness, but it is not definitive.If a court deems confusion to be relevant to the dilution case, then a survey showing confusion may be relevant. However, the absence of confusion does not demonstrate a lack of dilution; thus, a survey negating confusion may not be relevant.Even before the TDRA, there was no real consensus about what to measure. The Fourth Circuit had required evidence of actual harm to the mark’s selling power, a position seemingly rejected by the Moseley Court. Other courts have emphasized the potential loss of the famous mark’s distinctiveness as the appropriate measure of dilution. This is not the same as a loss of selling power. Loss of distinctiveness implies a loss of uniqueness. That is, when consumers no longer view a mark as pointing uniquely at the famous mark owner, it has lost some of its distinctive quality. Alternatively, one might view dilution as a diminishing of the mark’s unique message to the consumer; where once it signified great wealth, perhaps now it is associated also with goods of middle-class quality. This has been called “brand equity.” The first does not necessarily cause a loss of the mark’s selling power, or even a blurring of its message (as implied by the second message). Thus, it may not measure loss of selling power. The brand equity approach may measure loss of selling power, but only if careful questioning is performed. One would have to establish that consumers now view products that they believe come from the famous mark owner in a different light (one not desired by the famous mark owner).One other possibility is to attempt to measure the search-cost effect related to an allegedly diluting use. Although studies have attempted to demonstrate this effect, both the theory and the empirical data supporting it have been subject to criticism.Once one settles on an appropriate measure of dilution (and this may vary depending on which jurisdiction will receive the evidence), one must design a proper survey. More than one commentator has suggested a comparison survey. A comparison survey involves a control group, which has not been exposed to the second user, and a comparison group, which is familiar with the famous mark and has been exposed to the second use. Ideally, a famous mark owner would survey the control group prior to the advent of the second use. This requires that a mark owner be vigilant enough to realize the value of maintaining a control survey in the event that it is needed for litigation. The analysis then examines the difference between the control responses and those of the “contaminated” group. If the difference is large enough, then the effect could be attributed to dilution.Although this may seem obvious, the survey must take proper account of the context in which plaintiff’s and defendant’s uses occur. In the long-running dispute between Starbucks and a coffee roaster using “Charbucks” in various ways, the Second Circuit affirmed a district court finding that a telephone survey that did not give respondents the proper context for defendant’s uses—“Mister Charbucks” and “Charbucks blend ”—was, at best, of minimal value as a measure of actual association. In another case, a court discounted a survey that showed respondents a blank card with defendant’s mark, instead of a picture of an actual product, thus lacking marketplace context.Useful research may not result exclusively from litigation. In an example of a comparison study performed outside of the litigation context, researchers tested several hypotheses concerning dilution. One striking aspect of this study is the number of different hypotheses it tested. In part, this reflects the different possible interpretations of “dilution.” The study looked at several different versions of dilution: whether the second user weakens consumer associations with the first user’s products, whether dilution is more pronounced if the products are competitive, and dilution as measured by increased recall of the second user. The study indicates the many variables with which courts and litigants must grapple as they evaluate dilution claims. Ironically, one of its indications is that very strong marks may be impervious to dilution and weaker marks are more susceptible to dilution. As further research becomes available, practitioners will have to be aware of such conclusions, lest their opponents bring them out in court.2.Designing and Introducing a Survey: Some Litigation PitfallsAs discussed in the previous section, there are a number of important considerations for designing a proper dilution survey. Courts often note that the poor design of a survey does not necessarily make it inadmissible; the flaws may affect its evidentiary weight, rather than admissibility. Nevertheless, they should take seriously the role of “gatekeeper” assigned to the trial judge by Daubert v. Merrill Dow Pharmaceuticals, Inc. As the court noted in National Football League Properties, Inc. v. Prostyle, Inc., “‘helpfulness to the trier of fact remains the ultimate touchstone of admissibility.’” Daubert sets forth general guidelines for analyzing the admissibility of expert evidence, which includes a survey. It must be remembered that a survey will be introduced through an expert; courts should examine the expert’s testimony in a manner congruent with the examination of other scientific experts and refuse to admit surveys that are so flawed as to be unhelpful or misleading to the trier of fact.When designing a survey one first must insure that the survey measures something deemed relevant to the court. Under the TDRA, a brand-association survey would be relevant, though it may not be sufficient. In states using an actual-dilution standard, such a survey also may be deemed insufficient. But equally important, perhaps even more important, is to insure that the survey conforms to recognized survey principles. One court has noted that a controlled survey is essential. Other considerations include a relevant sample, both in size and composition; a sufficient number of questions to make responses meaningful; non-leading questions; proper supervision of questioners; and proper techniques of coding responses. Naturally, appropriate statistical methods must be employed. The results should be case-specific, and not simply general statements of research about dilution.The introduction of the survey will require one to follow the Federal Rules of Evidence, as well as the Federal Rules of Civil Procedure regarding the pretrial disclosure of expert testimony. And finally, the NFL Properties case provides a useful practical reminder that one should examine the expert’s testimonial history carefully; the court’s opinion cited several previous instances in which courts had rejected surveys performed by the plaintiff’s expert.VI. Dilution Protection for Product Configurations:A Difficult Dilution ProblemA.The IssuesOne of the more difficult issues under the TDRA is the protection of product configurations. Unlike the situation with regard to word marks, picture marks, or even packaging, protection for product configurations aims at the product itself. Thus, one can protect Buick against the use of its mark on a candy bar without preventing the production of the candy bar itself. However, if a candy bar maker could show that the shape of its candy bar is “famous,” it might be able to prevent anyone else from making a candy bar (or, conceivably, any number of other products) in that same shape. In the U.S. system of intellectual property, such monopoly uses of the product are normally the province of patent law, or sometimes copyright. But patents and copyrights are limited in duration, while trademark protection is perpetual. Thus, the “bargain” implicit in a patent of a limited-time monopoly, in exchange for disclosure of the invention so that others can make it, is nonexistent in trademark law.Traditionally, courts have pointed to two aspects of trademark law to distinguish it from patent law when product configurations are at issue: (1) the doctrine of functionality, and (2) the requirement of likelihood of confusion. However, dilution protects marks without regard to likelihood of confusion, and functionality is a weak basis of distinction between patent and trademark. Thus, dilution potentially offers greater protection than the patent system for product configurations.There is no indication that, prior to the TDRA, Congress ever considered the possibility that dilution protection would be extended to product configurations, either in 1995, when it enacted the FTDA, or in 1988, when a dilution statute almost became law. However, the definitions in the original statute did not exclude product configurations from coverage. Section 43(c) has always granted protection to the “owner of a famous mark.” Section 45 defines a mark as including “any trademark, service mark, collective mark, or certification mark.” Focusing simply on trademarks, the Lanham Act defines them as “includ[ing] any word, name, symbol, or device, or any combination thereof [used] … to identify and distinguish [the] goods [of one user] from those manufactured or sold by others and to indicate the source of the goods … .” Clearly, the terms “symbol or device” are broad enough to include product configurations. Courts interpreting the scope of Section 43(a) have long permitted its use to protect them.The TDRA contains new provisions that make it clearer that product configurations are protected. Section 43(c)(4) creates a new burden of proof for the element of fame for owners of unregistered trade dress who seek dilution protection. In addition, Section 43(c)(7), denominated a “Savings Clause,” disclaims any intent to override the operation of the patent laws. That can only mean that Congress considered the problem of product configurations and wanted to make sure that courts did not think that the TDRA allowed dilution to override the restrictions of patent. But it also means that Congress probably assumed that product configurations would be subject to dilution protection. In the final analysis, the major limitations on the use of dilution for product configurations would be the limits derived from the test for dilution, the doctrine of functionality, and, possibly, the collision between dilution and the patent laws or the Patent Clause of the Constitution.B.Tests for Dilution Do Not Limit Protection for ProductConfigurationsNone of the tests used by courts under the original FTDA—the actual harm test of the Fourth Circuit (a variation of which was adopted by the Supreme Court), the multifactor test of the Second Circuit, or the simplified test of fame plus similarity apparently used in the Seventh Circuit, or even the so-called Sweet test used in a number of other cases—had done much to harmonize perpetual dilution protection with limited-time patent protection. However, the TDRA made it clear that Congress did not intend that dilution protection override the operation of the patent laws. It is less clear how courts will seek to harmonize dilution protection and patent protection.Because there have been very few cases on this issue after the TDRA, most of the judicial discussions are from cases operating under a variety of pre-TDRA tests. It is unclear whether they provide appropriate guidance under the TDRA. Moreover, the limited case law under the TDRA is itself pregnant with questions.1.Product Configurations and the Test for Dilution by BlurringThe use of the TDRA’s six-factor test in a product configuration case will probably result in a finding in favor of the plaintiff. Since the configuration must be famous to be eligible for dilution protection, it is likely that the second factor, degree of recognition of the mark, will favor the plaintiff, as would the degree of distinctiveness. Assuming that the second user has made a copy, or something close to a copy, of the original, then similarity would also favor the plaintiff. The factor of predatory intent in a product configuration case is problematic. It may be that the second user deliberately copied the configuration in order to compete in the market for that product. Thus, the second user also presumably meant to capitalize on the fact that the famous configuration had created a strong market for this product design. That may give a court reason to decide that this factor favors the plaintiff. But intentionally copying a product is not the same as intentionally copying a word or logo mark. In the latter case, it is more reasonable to assume bad faith. Moreover, even if the object of this factor is not simply bad faith, the policy in favor of competition suggests that predatory “intent to create an association” should be analyzed more carefully in product configuration cases.The other two factors, mental association and the degree to which the famous mark owner is the exclusive user of the mark, may well favor the plaintiff. However, they are also connected to the possibility of functionality (which would make the configuration unprotectable as a mark). If the core of a product configuration consists of design elements that are relatively common, then that configuration may be deemed functional. Otherwise, competition would be unduly stifled. And if the design elements are somewhat common, then either (1) it is unlikely that the configuration will evoke a mental association between the two marks, or (2) any mental association will be due to the familiar (and functional) elements, and not necessarily the unique (and possibly nonfunctional) aspects of the famous product design.Clearly, courts should proceed cautiously in this area. Through the “Savings Clause” in the TDRA, Congress demonstrated its concern that dilution protection for product configurations could unwittingly become a substitute for patent protection. This would be an unwarranted anticompetitive result. Therefore, more than a mechanical application of the factors is in order. Courts must consider the issue in the broader context of the harmonization of trademark and patent and the prevention of anticompetitive effects.2.Tarnishment of Product ConfigurationsOwners of product configurations may complain that cheap, “knockoff” versions of their products will create an unwelcome association between the cheaper version and the originator’s product. An example of such a case under the FTDA is Libbey Glass, Inc. v. Oneida, Ltd. The plaintiff claimed that the defendant copied its line of glassware and that the lower quality of the defendant’s goods would cause tarnishment of the plaintiff’s designs. The court found an issue of fact as to the existence of tarnishment, but did not dispute the viability of this theory of liability.Similarly, in Abbott Laboratories v. Revitalyte LLC, the court denied a motion for judgment on the pleadings regarding dilution by tarnishment of the Pedialyte trade dress, noting allegations that defendant’s “advertisements feature references to heavy drinking by young adults, encourage excessive consumption of alcohol, and often feature other conduct that could be described as vulgar and crass, including the use of expletives and suggestive placement of the product.”Tarnishment claims in product configuration cases must be analyzed with care. If the senior manufacturer is able to claim too easily that the second user’s product is inferior and diluting, then consumers will lose the value of competition and lower-priced goods. At a minimum, courts should require a very strong showing that the second user is (or is likely to be) causing harm to the reputation of the trademark represented by the product configuration.C.Functionality, Fame, and Distinctiveness as Gatekeepers Between Dilution and PatentThe distinctions that exist between federal dilution protection and patent protection must come from something other than the test for dilution itself. Other possible distinctions are functionality, and fame and distinctiveness. All of these can be viewed as prerequisites to dilution protection.1.Functionality as Barrier to “Dilution Patent”In order for a product configuration to qualify for dilution protection, it must be protectable as a trademark. One of the general prerequisites for trade dress protection is nonfunctionality. Section 43(c)(4) expressly requires the owner of an unregistered trade dress to prove non-functionality. Presumably, Congress assumed that the owner of a federally registered trade dress had already demonstrated non-functionality to the Patent and Trademark Office. Even before the TDRA added Section 43(c)(4), however, it was assumed that the FTDA protected product configurations, provided that they were non-functional. As defined in trade dress cases, a functional design is one that is “dictated by” the function of the article, is the best or cheapest method of making this article, or, more generally, is one that is needed for effective competition in this market. Just as the refusal to protect generic word marks preserves a competitor’s interest in using the term, functionality recognizes the competitor’s need to use a particular product design.To the extent that functionality prevents monopolization of a useful design, it distinguishes trademark protection (including dilution) from utility patent protection. But design patents also protect nonfunctional designs. Unless functionality is interpreted in a way that precludes monopolization of certain nonfunctional designs, no significant distinction may exist between dilution and design patent protection. The seldom-used trademark doctrine of “aesthetic functionality” does preclude monopolization of some ornamental designs. Aesthetic functionality is based on the principle that certain designs are necessary for effective competition and should not be given trademark protection. If a design is essential to competition, then it will be deemed functional and not capable of protection. However, aesthetic functionality has not been well received by courts and commentators. It raises many doctrinal questions, notably which market should be considered when measuring the needs of competition. Unless the doctrine gains more acceptance, it will not be a strong distinguishing factor between patent and trademark dilution.In Apple, Inc. v. Samsung Electronics Co., the Federal Circuit discussed functionality in a case involving utility patents, design patents, and trade dress (both registered and unregistered). A jury found Samsung liable for both patent infringement and trademark dilution. Although the Federal Circuit affirmed the jury verdict as to patent infringement, it reversed the dilution verdict on the grounds that the trade dresses in question—both registered and unregistered—were functional. The Federal Circuit divided its discussion into unregistered and registered trade dress, because the latter is accorded a presumption of validity. It also used the Ninth Circuit’s functionality doctrine, because the case was filed in that Circuit. In particular, the Federal Circuit noted that product configurations are scrutinized more carefully for functionality, because of the potential overlap with patent and copyright. The unregistered trade dress consisted of several elements: a rectangular product with four evenly rounded corners;a flat, clear surface covering the front of the product;a display screen under the clear surface;substantial black borders above and below the display screen and narrower black borders on either side of the screen; andwhen the device is on, a row of small dots on the display screen, a matrix of colorful square icons with evenly rounded corners within the display screen, and an unchanging bottom dock of colorful square icons with evenly rounded corners set off from the display’s other icons.Applying the Ninth Circuit’s test, requiring that the unregistered trade dress “serve[] no purpose other than identification,” the Federal Circuit found that each of these elements served a utilitarian purpose:rounded corners improve “pocketability” and “durability” and rectangular shape maximizes the display that can be accommodated. J.A. 40869–70; J.A. 42612–13. A flat clear surface on the front of the phone facilitates touch operation by fingers over a large display. J.A. 42616–17. The bezel protects the glass from impact when the phone is dropped. J.A. 40495. The borders around the display are sized to accommodate other components while minimizing the overall product dimensions. J.A. 40872. The row of dots in the user interface indicates multiple pages of application screens that are available. J.A. 41452–53. The icons allow users to differentiate the applications available to the users and the bottom dock of unchanging icons allows for quick access to the most commonly used applications.In addition, the court found that alternative designs (not required under the Supreme Court’s decision in TrafFix) did not provide identical functionality and that Apple’s advertising had touted some of these functional advantages. In sum, the court found the unregistered trade dress to be unquestionably functional.Turning to the registered trade dress, namely, “the design details in each of the sixteen icons on the iPhone’s home screen framed by the iPhone’s rounded-rectangular shape with silver edges and a black background,” the court found this, too, enhanced the usability of the iPhone. Indeed, Apple’s own expert testimony indicated that the design of icons, for example, enhances the ability of users to understand how best to operate the iPhone. Thus, Samsung was able to overcome the presumption of validity and Apple did not demonstrate the non-functionality of its design.Thus, in this case functionality proved to be a barrier to dilution protection for a product configuration. However, this case appears to have been decided on utilitarian functionality grounds; aesthetic functionality was not analyzed (and, given the result, was unnecessary).In Evig LLC v. Fantasy Inc., the court found that the issue of the functionality of plaintiff’s trade dress was precluded by a prior court ruling that it was functional.2.Fame and Distinctiveness as Gatekeepers of Dilution ProtectionThe major limitation on product configuration protection in the TDRA is the requirement of fame. The TDRA gives a right of action to the owner of a “famous mark that is distinctive.” And distinctiveness is a prerequisite for any kind of trademark protection. Moreover, in the area of product configurations, the Supreme Court has now somewhat clarified the conditions for distinctiveness. In Wal-Mart Stores, Inc. v. Samara Bros., Inc., the Court held that a product configuration must be shown to have secondary meaning before it can be protected. The Court explained that although product configurations may be viewed as unique and attractive, they are not presumptively viewed as trademarks, that is, as source identifiers. Therefore, protection depends on an affirmative showing that the product configuration operates as an identification of source, not just as a pleasing design.As discussed in Chapter 8, the Second Circuit’s interpretation of the original FTDA—that marks must be “inherently distinctive” in order to be eligible for protection—made product configurations ineligible for dilution protection. However, the TDRA reworded the statute, which now protects marks that are distinctive, “inherently or through acquired distinctiveness.” Thus, product configurations should be eligible, as long as they can demonstrate fame and acquired distinctiveness.a.Demonstrating Secondary Meaning in a Product ConfigurationSecondary meaning can be shown by both direct and indirect evidence. Direct evidence usually is in the form of a consumer survey or unsolicited third-party references indicating the source-identifying properties of the design. Indirect evidence ordinarily is in the form of advertising expenditures, length of time that the mark has been in use, exclusivity of use, and evidence of copying. Where a product configuration is concerned, indirect evidence is an especially weak indication of secondary meaning. Long and exclusive use and advertising dollars may successfully create a demand for the product, but it is unclear that they create an association of source with the product’s shape. Copying, which with word marks or packaging would reasonably lead to the conclusion that the copyist believes the mark (or packaging) has associational value, is less instructive for product designs. One may copy a product because the public finds it attractive and because of a desire to compete in that product line.In one post-Wal-Mart case, a court required more evidence of secondary meaning than in cases of word marks or packaging. “Specifically, [plaintiff] must create an evidentiary basis for inferring that the source-identifying function has subordinated its other functions.” That court scrutinized the proffered evidence, which included several surveys, skeptically, deciding that the plaintiff had failed to sustain its burden. Perhaps most interesting was the court’s determination that the surveys must show not only that consumers view the product configuration as a source identifier, but that they also view it as emanating from a single (though perhaps anonymous) source. By making an exacting inquiry into secondary meaning, the court also placed a barrier in the way of state dilution claims, which may be less restrictive than the TDRA, but which also require distinctiveness. A heightened secondary meaning requirement, based on the assumption that the public is less apt to view product configurations as trademarks, would help distinguish product configuration dilution from patent infringement.b.A “Famous” Product Configuration MarkThe requirement that a mark be “famous” to earn protection under the TDRA can be a serious gatekeeper to dilution protection. The First Circuit, ruling in a product configuration dilution case, made clear that a court “should be discriminating and selective in categorizing a mark as famous.” The standard for fame was required to be even more rigorous than the standard for distinctiveness, even though many of the factors in both tests look the same. As discussed elsewhere, the Supreme Court’s decision in Wal-Mart v. Samara complicates the test for a famous mark under the TDRA. The statute protects a “famous mark that is distinctive, inherently or through acquired distinctiveness.” If “inherent distinctiveness” can no longer indicate that the product configuration serves as a trademark, it is difficult to make appropriate use of it in determining whether the configuration is a famous mark. Hopefully, courts will not engage in “gotcha” jurisprudence and assume that Congress has silently overruled Wal-Mart. The third factor to be considered in determining fame under the TDRA is “the extent of actual recognition of the mark.” In both situations, the goal is to show that the configuration serves as a source identifier. Rigorous application of this concept requires some sort of survey or surveys from the plaintiff. The surveys will need to demonstrate more than just source identification. They will need to show a high level of such identification.In addition, if the product design is not a registered trademark, then the mark owner also must comply with Section 43(c)(4), which requires unregistered trade dress to be shown to be famous without any consideration of other marks (like a famous word mark) that may appear on the product.In one case, a district court found that the plaintiff had established at least an issue of fact as to the fame of its product configurations—a “three stripe” adornment on athletic shoes, and a trade dress consisting of stripes, a particular flat sole, a particular shoe toe, and a colored “heel patch.” The court cited advertising, sales, and other evidence of awareness of the product design.In another case, the design of an iPhone 3GS was considered famous, while other iPhone designs and the design of an iPad were found not famous or functional.Prior to the TDRA, the use of “niche market” fame analysis loosened the barrier of a famous mark to dilution protection. (Niche market fame refers to fame that exists among consumers only in a specific product market.) However, the TDRA has eliminated niche market fame, requiring a mark to be famous among the “general consuming public of the United States.”D.The Possible Clash Between the TDRA and the Patent ClauseUse of dilution to protect product configurations may also conflict with the “limited times” provision of the Patent Clause of the Constitution. Because dilution provides protection without the need for confusion, it arguably protects against the simple act of copying the product. This may provide unlimited protection in an area where the Constitution intends protection to be limited in time. The TDRA contains a “Savings Clause” that appears designed with this issue in mind.The litigation between Apple and Samsung Electronics illustrates the problem. Apple’s claims included patent infringement (both utility and design patents) and trade dress dilution. The jury found for Apple on several claims under both patent infringement and dilution theories. However, the court did not discuss the possible constitutional issue.