What Can Harm the Reputation of a Trademark?

A Critical Re-Evaluation of
Dilution by Tarnishment
Michael Handler

Commentary: The Switch Thrown Wrong—How Railrunner Sent Intent-to-Use

Down the Wrong Tracks
Pamela S. Chestek

Commentary: An Idea Whose Time Has Come: Use of Takedown Notices for
Trademark Infringement
Susan D. Rector

Book Review: Domain Name Law and Practice: An International Handbook

(Second Edition). Edited by Torsten Bettinger & Allegra Waddell
Jane F. Collen

General Court of the European Union, Seventh Chamber: Statement in

Intervention in Case T-142/15 (Tribunal de l’Union Européenne septième
chambre: Memoire en Intervention dans l’affaire T-142/15)

May–June, 2016 Vol. 106 No. 3

The Trademark Reporter ®

WHAT CAN HARM THE REPUTATION OF A

TRADEMARK? A CRITICAL RE-EVALUATION OF
DILUTION BY TARNISHMENT

By Michael Handler ∗

I. INTRODUCTION
Dilution remains one of the most controversial concepts in
trademark law throughout the world. Dilution by blurring has
been the subject of sustained criticism for decades. Over the last
decade in particular, commentators have raised serious doubts
about the very nature of the harm of “blurring”—whether such
harm is conceived of as an increase in consumers’ “imagination
costs,” or as a type of “progressive harm” that should not be
allowed to go unchecked, even if no deleterious effects are evident
in any particular case, or as conduct that might result in
suboptimal investment in the development of new brands. This
scholarship has queried whether such harms rise to a level that
justifies legal intervention by way of an anti-dilution provision, or
whether such harms exist at all. 1 However, relatively little

∗ Associate Professor, Faculty of Law, University of New South Wales, Australia,

[Link]@[Link]. My warmest thanks go to Graeme Austin, Rob Batty, Robert
Burrell, Dev Gangjee, Megan Jones, Peter Karol, Doris Estelle Long, Sarah Lux-Lee, Lisa
Ramsey, Sandy Rierson, and Marlene Williams for reading drafts of this article and for
their invaluable suggestions for improvement. Particular thanks are due to Tobias Cohen
Jehoram, Martin Senftleben, and Jaap Spoor for their discussions about, and assistance in
locating and interpreting relevant sources of, Dutch and Benelux trademark law. Earlier
versions of this article were presented at the University of Auckland Business School
Seminar Series, September 2014; the INTA Annual Meeting Academic Day, San Diego, May
2015; and the Centre for Commercial Law/Intellectual Property Research Institute of
Australia Media & IP Law Conference, University of Melbourne, November 2015. I would
like to thank the organizers of these events, and also Irene Calboli, Earl Gray, Dan Hunter,
Signe Naeve, Peter Shand, David Tan, David Welkowitz, and Neil Wilkof for their feedback
on my presentations at these events. Final thanks go to Jessica Elliott, the editorial board of
the TMR, and the anonymous peer reviewers for their invaluable comments.
1. Critical commentary from the time of the Trademark Dilution Revision Act of 2006
includes Christine Haight Farley, Why We Are Confused About the Trademark Dilution Law,
16 Fordham Intell. Prop. Media & Ent. L.J. 1175 (2006) (arguing that anti-dilution law
provides a remedy without a supportable theorization of the harm); Clarisa Long, Dilution,
106 Colum. L. Rev. 1029, 1037 (2006) (arguing that the harm of dilution is elusive because it
is not clear whom the law is trying to protect); Mary LaFrance, No Reason to Live: Dilution
Laws as Unconstitutional Restrictions on Commercial Speech, 58 S.C. L. Rev. 709 (2007)
(concerned about the free speech implications of anti-dilution protection); Robert G. Bone, A
Skeptical View of the Trademark Dilution Revision Act, 11 Intell. Prop. L. Bull. 197 (2007)

Vol. 106 TMR 639

640 Vol. 106 TMR

scholarly attention has been paid to the other limb of dilution—

namely, tarnishment.
It is perhaps easy to see why. As Sandra Rierson recently
noted, the concept of tarnishment has a stronger intuitive appeal
than has blurring: 2 it seems easier to conceive of a right to prevent
the unauthorized association of a famous mark with insalubrious
content or poor-quality goods than it is to understand an action
designed to prevent the far more abstract “dispersal of identity” of
the mark. For this reason, examples of dilution by tarnishment are
often presented by judges and commentators as if the examples
speak for themselves. Famous illustrations might include Judge
Posner’s reference in Ty, Inc. v. Perryman to the hypothetical use
of “Tiffany’s” as a name for a striptease joint, 3 or, in the European
Union (“EU”), Advocate General Sharpston’s example in her
opinion in Intel Corp. Inc. v. CPM United Kingdom Ltd. of “Coca-
Cola” used for “low-grade engine oils or cheap paint strippers.” 4
Tarnishment also has seemingly clear parallels with other areas of
the law that seek to prevent harm to reputation, such as the torts
of defamation or injurious falsehood, or moral rights protections
for authors of copyright works, as well as with laws regulating
obscenity. Further, as will be seen below, there is little case law
around the world where dilution by tarnishment has been asserted
as the sole cause of action or where tarnishment has had
independent work to do beyond a separately pleaded claim of
trademark infringement. The very few cases in which dilution by
tarnishment has been established have tended to involve
“unsavory” uses of famous marks.

(arguing that dilution lacks a coherent policy foundation); Rebecca Tushnet, Gone in Sixty
Milliseconds: Trademark Law and Cognitive Science, 86 Tex. L. Rev. 507 (2008) (doubting the
imagination cost argument); Graeme W. Austin, Tolerating Confusion about Confusion:
Trademark Policies and Fair Use, 50 Ariz. L. Rev. 137 (2008) (pouring scorn on the
imagination cost argument); Robert Burrell & Michael Handler, Dilution and Trademark
Registration, 17 Transnat’l L. & Contemp. Probs. 713 (2008) (doubting the “progressive
harm” justification for anti-dilution protection); Dev Gangjee & Robert Burrell, Because
You’re Worth It: L’Oréal and the Prohibition on Free Riding, 73 Mod. L. Rev. 282, 290 (2010)
(criticizing attempts to justify trademark regimes by reference to the need to preserve
incentives for “brand entrepreneurs”); Sarah Lux, Evaluating Trade Mark Dilution from the
Perspective of the Consumer, 34 U.N.S.W. L.J. 1053 (2011) (arguing that consumer-based
justifications for anti-dilution protection lack internal consistency); Sandra L. Rierson, The
Myth and Reality of Dilution, 11 Duke L. & Tech. Rev. 212 (2012) (arguing that the costs of
anti-dilution laws to free speech and competition outweigh any private benefits to mark
owners); Barton Beebe, The Suppressed Misappropriation Origins of Trademark Antidilution
Law: The Landgericht Elberfeld’s Odol Opinion and Frank Schechter’s “The Rational Basis of
Trademark Protection,” in Intellectual Property at the Edge: The Contested Contours of IP,
59 (Rochelle Cooper Dreyfuss & Jane C. Ginsburg, eds., 2014) (suggesting that the theoretical
underpinnings of anti-dilution protection were deliberately obscured from the outset).
2. Rierson, supra note 1, at 246.
3. 306 F.3d 509, 511-12 (7th Cir. 2002).
4. Case C-252/07, [2008] E.C.R. I-8823, at paras. 9 and 81 (A-G Sharpston).
Vol. 106 TMR 641

To the extent that critical commentary specifically directed

toward dilution by tarnishment exists, much of it has tended to
accept the normative foundations of the cause of action, and
instead has focused on a more limited set of formal or doctrinal
issues. For example, there have been critical analyses of how the
tarnishment action should be classified: that is, whether it is
merely a subset of blurring, or whether it is, in fact, a type of
“dilution” at all. 5 There have also been detailed studies of the ways
in which U.S. courts have been interpreting the tarnishment
provisions introduced by the Trademark Dilution Revision Act of
2006 (“TDRA”), 6 and how courts in the European Union have
interpreted the analogous provisions of the original Trade Marks
Directive 7 and the Community Trade Mark Regulation. 8 A notable
example is Ilanah Simon Fhima’s wide-ranging analysis of anti-
dilution laws in the United States and the European Union—the
chapter devoted to tarnishment touches briefly on the nature of
the action and some of its potential problems, including its impact
on free commercial expression, but its primary focus is on
categorizing the types of conduct that courts have considered to be
“tarnishing” under decided case law, with relatively little critical
engagement with this body of law. 9 To the extent that trademark
scholars and practitioners might be uncomfortable with more
fundamental aspects of dilution by tarnishment (in the way that

5. See, e.g., Layne T. Smith, Tarnishment and the FTDA: Lessening the Capacity to
Identify and Distinguish, 2004 BYU L. Rev. 825; Barton Beebe, The Semiotic Analysis of
Trademark Law, 51 UCLA L. Rev. 621, 697-98 (2004).
6. Sarah L. Burstein, Dilution by Tarnishment: The New Cause of Action, 98 TMR
1189 (2008).
7. First Council Directive 89/104/EEC of 21 December 1988 to approximate the laws of
the Member States relating to trade marks [1989] O.J. L40/1, codified in Directive
2008/95/EC of the European Parliament and of the Council of 22 October 2008 to
approximate the laws of the Member States relating to trade marks [2008] O.J. L299/25
(“Original Trade Marks Directive”). See also Directive (EU) 2015/2436 of the European
Parliament and of the Council of 16 December 2015 to approximate the laws of the Member
States relating to trade marks [2015] O.J. L336/1 (“2015 Trade Marks Directive”). The 2015
Trade Marks Directive entered into force on Jan. 12, 2016, and Member States have until
Jan. 14, 2019, to bring into force laws that are necessary to comply with the key articles of
the new Directive.
8. Council Regulation (EC) No. 40/94 of 20 December 1993 on the Community trade
mark [1994] O.J. L 11/1, codified in Council Regulation (EC) No. 207/2009 of 26 February
2009 on the Community trade mark [2009] O.J. l78/1. See also Regulation (EU) 2015/2424 of
the European Parliament and of the Council of 16 December 2015 amending Council
Regulation (EC) No. 207/2009 on the Community trade mark [2015] O.J. L341/21. The
amending Regulation has effect from Mar. 23, 2016, and the amended Council Regulation
(EC) No. 207/2009 is now known as the “EUTMR.”
9. Ilanah Simon Fhima, Trade Mark Dilution in Europe and the United States ch. 5
(2011).
642 Vol. 106 TMR

many such commentators clearly are with dilution by blurring),

this is rarely articulated in detail in the literature. 10
In this article, I seek to challenge the normative foundations of
tarnishment, by querying whether this form of dilution—to the
extent it encompasses conduct beyond the boundaries of the
traditional, confusion-based, trademark infringement action—is, in
fact, a “harm” of which the law should take cognizance. I do so by a
detailed analysis of three broad issues.
The first issue, which I explore in Part II, relates to the
historical origins of the tarnishment action in both the United
States and Europe. My focus is on mid-twentieth century U.S.
state dilution statutes and earlier case law, and on developments
in the Benelux countries in the 1960s and 1970s that helped shape
the dilution provisions of the original European Trade Marks
Directive. My research reveals that the type of harm against which
such early laws were designed to protect was arguably more
limited than “tarnishment” as it is currently conceptualized in U.S.
and EU law, which should cause us to devote serious thought to
the normative basis of the modern action. In Part III, I turn to
consider the question of “reputation” in the modern tarnishment
action. Specifically, looking at case law from the United States and
the European Union, I investigate the ways in which famous
marks’ reputations have been constructed in tarnishment actions,
demonstrating that there has been insufficient judicial
engagement with the complex, multivalent notion of brand
“reputation.” In Part IV, I seek to unpack the issue of “harm to
reputation.” Looking in part to literature on consumer psychology
and using two decided cases as studies—the 2010 decision of the
U.S. Court of Appeals for the Sixth Circuit in V Secret Catalogue,
Inc. v. Moseley 11 and the Benelux Court of Justice’s 1975 decision
in Claeryn/Klarein12—I analyze whether, in the absence of
consumer confusion, unauthorized uses of famous marks do, in
fact, create reputational harm, at least of a type and magnitude
that might justify the current levels of legal intervention under
U.S. and EU law. I also address the extent to which the internal
limits of the dilution actions around the world adequately
accommodate free speech concerns in cases where junior uses of
famous marks seek to comment on, critique, or parody those marks
in a manner that might cause some degree of reputational harm.
Taken together, my investigation of these issues shows that
tarnishment, rather than being the more easily accepted limb of

10. A recent exception is Christopher Buccafusco, Paul J. Heald & Wen Bu, Testing
Tarnishment in Trademark and Copyright Law: The Effect of Pornographic Versions of
Protected Marks and Works (Dec. 8, 2015), at [Link]
t_id=2700840, discussed in detail infra Part IV.B.
11. 605 F.3d 382 (6th Cir. 2010).
12. Reported in English at (1976) 7 I.I.C. 420; in Dutch at [1975] N.J. 472.
Vol. 106 TMR 643

dilution, raises as many complex and unresolved issues as blurring

with regard to the “harms” against which anti-dilution laws are
said to protect. This should give us cause to reconsider whether the
entire dilution action (that is, for both blurring and tarnishment) is
really about preventing cognizable harm at all, or is ultimately an
action concerned with the morality of trade behavior. I touch on
these points, and what the consequences of a re-conceptualization
of the action along these lines might be, by way of conclusion in
Part V.

II. HISTORICAL UNCERTAINTIES

A. Current Scope, and Orthodox Histories of,
the Tarnishment Action
Despite the fact that existing international conventions do not
call for the enactment of anti-dilution protection (at least to the
extent it extends beyond cases of consumer confusion), 13 the key
jurisdictions that have provided a cause of action for dilution by
tarnishment have done so in a broadly similar manner. Thus,
under federal U.S. law, as a result of the TDRA, we see “dilution
by tarnishment” defined as “association arising from the similarity
between a mark or trade name and a famous mark that harms the
reputation of the famous mark,” 14 and actions for dilution that
apply “regardless of the presence or absence of actual or likely
confusion, of competition, or of actual economic injury.” 15 Under
the original European Trade Marks Directive, Member States
could provide a cause of action allowing the owner of a registered
trade mark that has “a reputation in the Member State” to prevent
the use, “without due cause,” of an identical or similar sign that “is
detrimental to . . . the repute of the trade mark.” 16 Under the 2015
Trade Marks Directive, Member States are required to make such
an action available. 17 The European Court of Justice has explicitly
referred to this action as one that prevents the “tarnishment” or
“degradation” of the registered mark, 18 and which does not depend

13. See Michael Handler, Trade Mark Dilution in Australia?, 2007 Eur. Intell. Prop.
Rev. 307, 308-10; Novelty Pte. Ltd. v. Amanresorts Ltd. [2009] 3 S.L.R. 216 (Singapore
Court of Appeal).
14. 15 U.S.C. § 1125(c)(2)(C).
15. 15 U.S.C. § 1125(c)(1) (injunction), §§ 1063 and 1052(f) (opposition) and §§ 1064,
1092 and 1052(f) (invalidation).
16. Original Trade Marks Directive, art. 5(2). This was also a ground of opposition
to/invalidation of registration: id., arts. 4(3) and 4(4)(a).
17. 2015 Trade Marks Directive, art. 10(2)(c). This remains a ground of opposition
to/invalidation of registration: id., art. 5(3)(a). Analogous provisions apply under the EU-
wide scheme: EUTMR, art. 9(2)(c) (infringement), art. 8(5) (opposition) and art. 53(1)(c)
(invalidity).
18. Case C-487/07, L’Oréal SA v. Bellure NV [2009] E.C.R. I-5185, para. 40 (E.C.J.).
644 Vol. 106 TMR

on the existence of confusion. 19 A number of non-European

countries, whose laws have been modelled on the original Trade
Marks Directive or British law implementing the Directive, 20 or
which have entered into bilateral trade agreements requiring them
to adopt anti-dilution laws, 21 have similar “tarnishment”
provisions in place.
The conventional histories of statutory protection against
dilution by tarnishment under both U.S. and EU law have been
reasonably well traversed. 22 To summarize these histories in
outline, in the United States, legal protection against tarnishment
(as well as blurring) is generally traced back to 1947, to a
Massachusetts statute 23 that added a new Section 7A to Chapter
100 of the General Laws of Massachusetts. This new section
provided:
Likelihood of injury to business reputation or of dilution of the
distinctive quality of a trade name or trade-mark shall be a
ground for injunctive relief in cases of trade-mark
infringement or unfair competition notwithstanding the
absence of competition between the parties or of confusion as
to the source of goods or services.
This level of protection extended beyond that which was available
to trademark owners in Massachusetts at that time. Similarly
worded amendments to state statutes, also designed to protect
“injury to business reputation” and the “dilution of the distinctive
quality” of a trademark in the absence of competition or source
confusion, were adopted in Illinois, Georgia, and New York in the
1950s. These quickly became known as “anti-dilution” statutes (the
term “anti-dilution” here being used as shorthand for both types of
injury contemplated by the statutes). 24 The United States
Trademark Association (“USTA”) revised its Model State

19. Case C-408/01, Adidas-Salomon AG v. Fitnessworld Trading Ltd. [2003] E.C.R. I-

12537, para. 27 (E.C.J.).
20. See, e.g., Trade Marks Act 1999 (India), section 29(4) (infringement) and section 11
(ground of opposition/invalidity); Trade Marks Act 2002 (N.Z.), sections 89(1)(d) and 89(2)
(infringement). Even though New Zealand law was modelled on post-Directive U.K. law, the
action for infringement by tarnishment differs from U.K. law, in that there is a requirement
in New Zealand that the use be “as a trade mark” but no requirement that the defendant’s
use be “without due cause.” Further, dilution by tarnishment cannot be raised in New
Zealand as a ground of opposition or invalidity. For consideration, see Paul Sumpter, Anti-
Dilution Law Down the Road—But Downsized? A Perspective from New Zealand, 80 Intell.
Prop. Forum 11 (2010).
21. Singapore-U.S. Free Trade Agreement 2003, art. 16.1.2(b)(i). See Trade Marks Act
1998 (Singapore), section 2 (definition of “dilution”), section 55(3)-(4) (action for
infringement) and section 8(4) (ground of opposition/invalidity).
22. My summary of U.S. law draws on Burstein, supra note 6, at 1192-96.
23. 1947 Mass. Acts, ch. 307 (p. 300).
24. For an early example, see Beverly W. Pattishall, The Case for Anti-Dilution Trade-
Mark Statutes, 43 TMR 887 (1953).
Vol. 106 TMR 645

Trademark Bill in 1964 to include “anti-dilution” provisions

similar to those contained in the Massachusetts statute quoted
above. A substantial number of other states, including California
and Texas, had adopted anti-dilution statutes by the time
Congress passed the Federal Trademark Dilution Act of 1995
(“FTDA”). Despite some uncertainty as to whether tarnishment
was, in fact, actionable as a type of dilution under the FTDA, 25 the
situation is now clear under the TDRA.
In Europe, the provisions of the original Trade Marks
Directive, referred to above, owe their origins to Article 13A(2) of
the Benelux Merkenwet (“BMW”). The Netherlands, Belgium and
Luxembourg had started negotiations toward a common
trademark law in the late 1940s, which continued throughout the
1950s. These resulted in the Benelux Convention Concerning
Trademarks, and a uniform trademark law known as the BMW.
While the text of both was settled by the early 1960s, 26 the
Convention entered into force only in 1969, 27 and the BMW took
effect only at the start of 1971. 28 The BMW provided for a unitary

25. See Moseley v. V Secret Catalogue, Inc., 537 U.S. 418, 432 (2003). This uncertainty
followed from the gap between the text of the FTDA (where “dilution” was defined in § 4 as
“the lessening of the capacity of a famous mark to identify and distinguish goods or
services,” language very much based around the harm of blurring) and what Congress
asserted the FTDA was designed to protect (namely, “famous trademarks from subsequent
uses that blur the distinctiveness of the mark or tarnish or disparage it”: see H.R. Rep. No.
104-374, p. 2, at [Link] This
conceptual confusion was compounded by the fact that, less than a decade earlier, Congress
had arguably seen “tarnishment” as something different from “dilution.” Specifically, in
1987, the Trademark Review Commission of the USTA made various recommendations and
drafted amendments to the Lanham Act, which became reflected in a bill that would become
the Trademark Law Revision Act of 1988. (In 1993, the USTA changed its name to the
International Trademark Association.) The Commission recommended the adoption of a
federal anti-dilution provision, with “dilution” understood as blurring and with the term
defined in both its recommendations and the resulting bill in essentially the same language
as the provision eventually adopted in the FTDA: see United States Trademark Association
Trademark Review Commission and Recommendations to USTA President and Board of
Directors, 77 TMR 375, 454-59 (1987). In a different part of its report, the Commission
separately recommended an addition to the infringement provision in § 43(a) of the Lanham
Act for conduct that was “likely to disparage or tarnish the mark of another,” a
recommendation also reflected in the bill: id. at 433-35. Neither the “dilution” nor the
“tarnishment” provisions of the bill were enacted, but at no stage did Congress suggest that
one of the problems with those provisions of the bill was the distinction drawn between
“tarnishment” and “dilution.” For analysis, see Smith, supra note 5, at 840-44; David S.
Welkowitz, Trademark Dilution: Federal, State, and International Law 120, 222-28 (2d ed.
2012).
26. See Antoine Braun, Précis des Marques de Produits: Loi Uniforme Benelux, Loi
Belge, Droit International 16-18 (1971) (in French). The text of what would become art. 13A
appears to have been settled as early as 1958: Projet de Convention et de Loi Uniforme
Benelux en Matiere de Marques de Produits, Document 15-1 de Conseil Interparliamentaire
Consultatif de Benelux 38 (Dec. 2, 1958) (in French) (copy on file with author).
27. The text is available in French at [Link]
265148.
28. Charles Gielen & Benoit Strowel, The Benelux Trademark Act: A Guide to
Trademark Law in Europe, 86 TMR 543, 544 (1996).
646 Vol. 106 TMR

right across the three countries, supplanting the three former

national trademark systems. Article 13A of the BMW provided:
Without prejudice to the application of the ordinary law
governing civil liability, the proprietor may, by virtue of his
exclusive right oppose:
(1) any use made of the mark or of a similar sign for the goods
for which the mark is registered or similar goods;
(2) any other use without due cause in the course of trade of
the mark or a similar sign made under such conditions that
would cause damage to the owner of the mark.
Article 13A(2) came to be interpreted expansively by the newly
created Benelux Court of Justice in the mid-1970s, so as to afford
mark owners protection against the dilution of their marks in the
absence of source confusion. Even more significantly, following the
determined efforts of Dutch negotiators, the Benelux provision
provided a “partial template” for Article 5(2) of the original
European Trade Marks Directive, 29 an article that the Court of
Justice subsequently held protects against the harms of “blurring”
and “tarnishment.” 30 Despite being an optional provision, Article
5(2) was implemented in the domestic laws of most EU Member
States. Under the 2015 Trade Marks Directive, the analogous anti-
dilution provision, Article 10(2)(c), is now mandatory.
What is immediately striking about both the 1947
Massachusetts statute and Article 13A(2) of the BMW is the
breadth of the language used. Under the Massachusetts statute,
liability was to be found for “injury to business reputation” even in
the “absence of competition between the parties or of confusion as
to . . . source.” The BMW placed no limits on the type of damage
caused to the owner of the mark or the circumstances in which
such damage might be caused (other than that the defendant’s use
take place in the course of trade and not be without due cause). It
would seem to be clear from a literal reading of these provisions
that they were intended to extend trademark protection well
beyond the traditional tests for infringement that existed at the
time—in particular, to situations where the parties were providing
vastly different goods or services. It is perhaps for this reason that
the intent of the statutes has been taken largely at face value—
that is, it seems to have been accepted that these were intended to
be “proto-dilution” statutes—unambiguous acknowledgments by
legislatures that dilution by what we would now describe as
blurring and tarnishment, in the absence of source confusion, were

29. Ilanah Simon Fhima, Exploring the Roots of European Dilution, 2012 Intell. Prop.
Q. 25, 36. See also Gielen & Strowel, supra note 28, at 572.
30. Case C-487/07, L’Oréal SA v. Bellure NV [2009] E.C.R. I-5185, para. 40 (E.C.J.).
Vol. 106 TMR 647

harms against which, from that point on, the law would provide
protection. 31
While the statutes did come to be interpreted that way, there
has been relatively little critical inquiry into what the statutes
were originally designed to achieve. 32 This requires a closer look at
the state of the law in Massachusetts, in other U.S. states and at a
federal level, and the state of the law in the Netherlands, around
the time of the adoption of the new statutes. Such a historical
analysis, far from confirming the modern understanding of the
statutes, in fact casts doubt over their intended scope and, in
particular, the extent to which they were designed to create free-
standing causes of action against dilution by both blurring and
tarnishment.

B. Rethinking the History of Tarnishment

under U.S. State Law
Understanding the scope of the 1947 Massachusetts statute
first requires an appreciation of the state of trademark and unfair
competition law at that time. By 1947, no uniform standard had
emerged across state and federal trademark and unfair
competition laws with respect to those instances where a
defendant would be held liable for using another party’s mark
when the parties were not in direct competition with one another.
Starting with Aunt Jemima Mills Co. v. Rigney & Co. in 1917, 33
federal judges had gradually expanded the confusion-based test for
liability, finding parties to be liable when they used others’ marks
or similar variants in relation to related, but non-competing,
goods. 34 Courts stretched the notion of the “relatedness” of goods in
some of these cases, 35 and started to articulate the harms caused
to the plaintiff more in terms of damage to reputation than

31. See, e.g., Burstein, supra note 6, at 1192; Robert G. Bone, Schechter’s Ideas in
Historical Context and Dilution’s Rocky Road, 24 Santa Clara Computer & High Tech. L.J.
469, 497 (2008).
32. An important exception, in the case of the Massachusetts statute, is Bone,
Schechter’s Ideas, supra note 31.
33. 247 F. 407, 409-10 (2d Cir. 1917) (owner of AUNT JEMIMA for flour able to prevent
another’s use of “Aunt Jemima” for syrup).
34. For other key cases, see Vogue Co. v. Thompson-Hudson Co., 300 F. 509 (6th Cir.
1924) (owner of VOGUE and V device on hats able to restrain the use of similarly presented
VOGUE and V on magazines); Wall v. Rolls-Royce of Am., Inc., 4 F.2d 333 (3d Cir. 1925)
(owner of ROLLS-ROYCE in relation to cars and aeroplanes able to prevent the use of
“Rolls-Royce” on radio tubes); Yale Elec. Corp. v. Robertson, 26 F.2d 972 (2d Cir. 1928)
(enjoining the use of a “Yale” device in relation to electric flashlights and batteries in the
face of another party’s YALE trademark for hardware and locks); L.E. Waterman Co. v.
Gordon, 72 F.2d 272 (2d Cir. 1934) (owner of WATERMAN for fountain pens able to prevent
the sale of “Waterman” razor blades).
35. See, e.g., Wall, 4 F.2d at 334.
648 Vol. 106 TMR

diversion of custom. 36 In some states, notably New York, it was

possible to point to far-reaching decisions in the 1930s where
courts stretched the confusion-based test even further, finding
liability in cases involving use of identical or similar marks on non-
competing goods, without finding that the parties’ goods and
services were related. 37 However, even as late as 1947, other states
were continuing to interpret their unfair competition laws in a far
more conservative manner.
Massachusetts was one such state. Before the passage of the
1947 statute, liability would only be found “where the defendant
was soliciting the same trade from the same customers in the same
territory” 38—that is, the plaintiff and defendant needed to be in
direct competition. Although the historical record relating to the
adoption of the Massachusetts statute has rightly been described
as meagre, 39 it seems clear that at least one aim of the 1947
statute was to bring Massachusetts law into line with federal law,
and the law of other states, by ensuring that liability for unfair
competition or infringement could be found in the absence of
competition, where the injury to the plaintiff extended beyond the
diversion of trade. 40
A more pertinent, and much more difficult, question is
whether the 1947 statute intended to go further than this—in

36. See, e.g., Yale, 26 F.2d at 974. Judge Learned Hand’s later decision in S.C. Johnson
& Son, Inc. v. Johnson, 116 F.2d 427, 429 (2d Cir. 1940), has been argued to be a
“turnabout” from the broad approach taken in Yale: see Gerard N. Magliocca, One and
Inseparable: Dilution and Infringement in Trademark Law, 85 Minn. L. Rev. 949, 997
(2001); Bone, Schechter’s Ideas, supra note 31, at 496. Although Judge Hand was more
explicit in S.C. Johnson about the potential monopoly concerns involved in cases involving
non-competing goods, the case involved a defendant using his own surname, which helps
explain the limited injunction that permitted him to continue to trade under his name with
an appropriate disclaimer. Crucially, Judge Hand maintained that a key factor in cases
involving non-competing goods was harm to reputation. See infra, notes 52-57 and
accompanying text.
37. Tiffany & Co. v. Tiffany Prods., Inc., 147 Misc. 679 (N.Y. Sup. Ct. 1932), aff’d, 237
A.D. 801 (N.Y. App. Div. 1932), aff’d, 262 N.Y. 482 (1933) (owner of TIFFANY for jewellery
able to prevent “Tiffany” for film distribution); Phila. Storage Battery Co. v. Mindlin, 163
Misc. 52 (N.Y. Sup. Ct. 1937) (manufacturer of radio sets and storage batteries under
PHILCO mark successful in its action against the maker of “Philco” razor blades). See id. at
55 (collecting further examples from state and federal law, and making reference to two
passing-off cases from the United Kingdom (Eastman Photographic Materials Co. Ltd. v.
John Griffiths Cycle Corp. Ltd. (1898) 15 R.P.C. 105 (Ch.) and Walter v. Ashton [1902] 2 Ch.
282 (Ch.)), in which misrepresentations had been found in cases involving the use of marks
on dissimilar goods).
38. Patrick H. Harrington, Jr., Note: Governmental Regulation of Business—Unfair
Competition—Trade Name Legislation, 27 B.U. L. Rev. 489, 489 (1947).
39. Walter J. Derenberg, The Problem of Trademark Dilution and the Antidilution
Statutes, 44 Cal. L. Rev. 439, 452 (1956).
40. Harrington, supra note 38, at 491 (“[i]t is unfortunate that the reluctance of the
Massachusetts Court to expand its concept of unfair competition has necessitated this
statute”); see also Robert N. Klieger, Trademark Dilution: The Whittling Away of the
Rational Basis for Trademark Protection, 58 U. Pitt. L. Rev. 789, 812, n.126 (1997).
Vol. 106 TMR 649

particular, whether it meant to establish an entirely new form of

liability, untethered to consumer confusion, such that it can be
classified as a “proto-dilution statute.” 41 Some commentators
writing at the time believed this to have been the case, 42 and the
use of the term “dilution” and the fact that the statute was to
apply “notwithstanding the absence of . . . confusion as to the
source of goods or services” certainly suggests as much. However,
it is vital to note that the statute was to apply “in cases of trade-
mark infringement or unfair competition.” No new cause of action
was created. Rather, the statute made clear that the harms of
“injury to business reputation or of dilution of the distinctive
quality of a trade name or trade-mark” could give rise to injunctive
relief. The fact that the claim still had to be for trademark
infringement or unfair competition raises a question about the
precise roles that “dilution” and “injury to business reputation”
were intended to play under the new Massachusetts law, and
what, exactly, was intended by the reference to “the absence of . . .
confusion as to the source of goods or services.” Addressing these
issues requires an understanding of how the language of “dilution”
and “injury to business reputation” had been deployed by courts
and commentators up to 1947.
Courts started using the term “dilution” in trademark cases in
the 1930s. But, in a point that is easily lost in modern accounts of
the history of the dilution action, courts used the term differently
from how the concept of dilution had been originally conceived
(and from the way we now understand it). Frank Schechter, in his
1927 Harvard Law Review article, had proposed what might be
loosely described as a right against dilution by blurring as a
conceptually distinct category of legal protection from traditional,
confusion-based trademark infringement. 43 This proposal received
very little direct support. 44 However, in the 1930s, the courts

41. Similar questions can be asked about the near-identical statutes adopted in Illinois
in 1953 (1953 Ill. L. 455, § 1) and New York in 1955 (1955 N.Y. Laws 466, § 1).
42. See, e.g., Ralph S. Brown, Advertising and the Public Interest: Legal Protection of
Trade Symbols, 57 Yale L.J. 1165, 1194 n.121 (1948).
43. Frank I. Schechter, The Rational Basis of Trademark Protection, 40 Harv. L. Rev.
813 (1927).
44. See Bone, Schechter’s Ideas, supra note 31, at 489-92 (on the mild reception of
Rational Basis up to 1937), id. at 492-94 (providing two convincing reasons as to why this
was so: the lack of general interest among practitioners in the theoretical foundations of
trademark law, and the ascendency of the “goodwill-as-property” theory in supporting the
expansion of trademark law in the 1920s and 1930s. Bone’s third reason, that Schechter did
not push aggressively for a dilution cause of action, is, however, harder to square with
Schechter’s drafting of the Perkins Bill and his advocacy before Congress in 1932 as to the
desirability of adopting stand-alone anti-dilution protection, discussed in detail in
Derenberg, supra note 39, at 449-50. A more compelling reason might simply be the
complexity of Rational Basis, and the lack of clarity over whether Schechter’s proposal was
meant to set up a new general theory of protection, or merely to apply to the limited subset
of fanciful, arbitrary, or coined marks).
650 Vol. 106 TMR

adopted and applied the concept of “dilution” in infringement and

unfair competition actions as a means to describe the harm
potentially suffered by a plaintiff where confusion as to source had
arisen out of concurrent use of a similar trademark by a defendant
on non-competing goods. 45 Most notably, in Tiffany & Co. v.
Tiffany Prods., Inc., 46 the New York Supreme Court restrained the
defendant’s use of “Tiffany” for film production and distribution
services (noting the defendant’s use of the mark in conjunction
with a “diamond” image in its marketing) on the basis that this
constituted unfair competition in light of the substantial
reputation of TIFFANY for jewelry. In finding that there was
source confusion, Judge Dore quoted Schechter, explaining that
the injury to the plaintiff was best described 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.” 47 One of the reasons that courts might have been attracted
to using “dilution” in this way is that it provided an explanation
for the harm experienced by plaintiffs in the absence of
competition at a time when the courts were expanding the
confusion-based tests for liability. More specifically, the concept of
“dilution” provided an avenue for describing the “missing injury
element” 48 in cases where junior users were providing goods of a
different quality from those of the plaintiff, which, as a matter of
strict logic, might mean that consumers would be less likely to be
confused as to the origin of the goods than if the two sets of goods
were of the same quality. 49
What is of vital importance for present purposes is that
“dilution,” by 1947, did not have a settled meaning. While the idea
of dilution as a stand-alone cause of action had been revived by
Rudolf Callmann in his 1945 treatise, 50 in case law “dilution” had
been used only to buttress the expansion of the confusion-based
test for infringement—a development of which even Schechter had
come to express his approval by the mid-1930s. 51 The presence of

45. See George E. Middleton, Some Reflections on Dilution, 42 TMR 175, 184-85 (1952).
A further example is Triangle Publ’ns, Inc. v. Rohrlich, 73 F. Supp. 74 (S.D.N.Y. 1947),
where the publisher of Seventeen was successful in its unfair competition action against
“Miss Seventeen Foundations Co” for cosmetics, with the district court noting the potential
harm of dilution that might flow from the likely consumer confusion: id. at 76.
46. 147 Misc. 679 (N.Y. Sup. Ct. 1932), aff’d, 237 A.D. 801 (N.Y. App. Div. 1932), aff’d,
262 N.Y. 482 (1933).
47. Id. at 681, quoting Schechter, Rational Basis, supra note 43, at 825. See also Phila.
Storage Battery Co. v. Mindlin, 163 Misc. 52, 54-55 (N.Y. Sup. Ct. 1937).
48. Bone, Schechter’s Ideas, supra note 31, at 494 n.125.
49. See Magliocca, supra note 36, at 1000-01.
50. Rudolf Callmann, The Law of Unfair Competition and Trade-Marks (1945).
51. Frank I. Schechter, Fog and Fiction in Trade-mark Protection, 36 Colum. L. Rev.
60, 84 (1936).
Vol. 106 TMR 651

the word in the Massachusetts statute, and in similar state

statutes, needs to be interpreted in this context.
In a similar vein, “injury to business reputation” needs to be
understood as having a more specific meaning than the generality
of the phrase might otherwise suggest. To the extent that the
language had been used in the federal unfair competition
context, 52 it was being used at a time when courts were attempting
to articulate the harm suffered by a plaintiff in a case where the
defendant had been using the trademark on non-competing goods,
thus presenting no possibility of diversion of trade. By the mid-
1920s, the idea had started to take hold that the harm suffered by
a plaintiff in such a case was “harm to reputation,” in the sense
that the quality of the plaintiff’s reputation no longer laid within
its control. 53 Thus, by 1932, the U.S. Court of Appeals for the
Tenth Circuit in Standard Oil Co. of New Mexico v. Standard Oil
Co. of California 54 could observe that:
If one fraudulently sells his goods . . . as those of another,
injury may result to the latter although he is not engaged in
the manufacture or sale of like goods. Where one passes off his
goods, his services, or his business as the goods, services, or
business of another, equity will intervene to protect the good-
will and business reputation of the latter from any injury
liable to be caused thereby. 55
Closer to the time of the Massachusetts statute, in the decision of
the U.S. Court of Appeals for the Second Circuit in Standard
Brands, Inc. v. Smidler, 56 Judge Chase wrote:
The protection which the law gives the owner of a trade-mark
is not confined to the goods upon which it is, or has been, used
by the owner of it but extends to products which would be
reasonably thought by the buying public to come from the
same source if sold under the same mark . . . His mark is the
brand by which his goods can be identified, and when it is
used by another the reputation of his mark, and consequently
his own business reputation, are placed to that extent beyond
his control. Unless the use by that other is upon goods so
unlike his own or in territory so far from that which he has
exploited that it will not create confusion, it will be enjoined. 57

52. The language was also used in cases of libel or slander of title: see, e.g., Pullman
Standard Car Mfg. Co. v. Local Union No. 2928 of United Steelworkers of Am., 152 F.2d
493, 497 (7th Cir. 1946).
53. See, e.g., Yale Elec. Corp. v. Robertson, 26 F.2d 972, 974 (2d Cir. 1928).
54. 56 F.2d 973 (10th Cir. 1932).
55. Id. at 978-79.
56. 151 F.2d 34 (2d Cir. 1945).
57. Id. at 37. See also Phila. Storage Battery Co. v. Mindlin, 163 Misc. 52, 54 (N.Y. Sup.
Ct. 1937) (referring to a possible harm flowing from source confusion as being that the mark
652 Vol. 106 TMR

What is crucial is that this harm to “business reputation” was

recognized as arising in the context of unfair competition, which
still turned on a showing of consumer confusion. There was no
sense that the adoption of the language of “injury to business
reputation,” in isolation, was intended to introduce an independent
“proto-tarnishment” cause of action to determine liability for
conduct beyond that proscribed by unfair competition law.
More difficult to explain, then, is the reference in the
Massachusetts statute to the availability of an injunction in “the
absence of . . . confusion as to the source of goods or services.” This
might suggest that the legislative intent in creating the statute
was, in fact, to establish a new, “full” anti-dilution cause of action.
However, by its very terms, the statute was to apply only “in cases
of trade-mark infringement or unfair competition.” This part of the
statute is rarely commented on, but it is highly significant,
because it shows that the legislature did not set up a distinct cause
of action for parties that enabled them to seek an injunction to
restrain “dilution” or “injury to business reputation” at large—the
action still needed to be for trademark infringement or unfair
competition. 58 Moreover, if read literally, the statute would have
eliminated confusion as a necessary element in an action for
trademark infringement or unfair competition in all cases, not only
those involving non-competing goods. It seems highly unlikely that
a change so sweeping would have been the object of such a terse
amendment by the legislature.
What, then, did the Massachusetts legislature intend to
accomplish by referring in the statute to “the absence of . . .
confusion”? Two potential readings are possible. One is that these
words were intended to be broadly synonymous with “the absence
of competition between the parties.” Keeping in mind the highly
restrictive nature of Massachusetts unfair competition law at the
time, it is arguable that the intention of the statute was to make
doubly clear that courts were no longer required to find that the
defendant’s use of a similar mark would generate confusion that
would result in a diversion of trade from the plaintiff. That is,
“confusion” might have been used in the statute as shorthand for
both a mistaken belief as to source and a resulting, limited type of
harm (the diversion of trade), with the real concern of the statute

owner’s “reputation may be tarnished by the use of his mark upon an inferior product.” In
support of this proposition, Judge Shientag cited the U.K. House of Lords’ decision in Eno v.
Dunn ((1890) 15 App. Cas. 252, 258), in which a mark was denied registration for baking
powder on the basis that its use would deceive the public due to the existence of an earlier,
similar mark being used in relation to unrelated goods, namely, salts used to produce a fizzy
drink).
58. This is to be contrasted with the approach taken in the FTDA and TDRA, where
trademark infringement and anti-dilution actions are treated as ontologically separate (that
is, the latter is not a subset of the former).
Vol. 106 TMR 653

being to ensure that that limited type of harm was no longer an

essential element of an action for trademark infringement or
unfair competition. Alternatively, the words could have been
intended to capture the principle of trademark infringement that
existed at the time that a defendant could not avoid liability for
use of the same mark as the plaintiff for competing goods by
claiming that the mode of use of its mark would render confusion
unlikely. Such mode of use might include the use by defendant of
trade dress in addition to the mark, or use of disclaimers with the
mark. 59
This is not to suggest that the orthodox view of the
Massachusetts statute (that is, that it set up a “full” anti-dilution
measure) is implausible. However, beyond a literal reading of the
statutory text, the evidence that the legislature intended to enact a
change of such significance is thin. There is little to suggest that
the legislators were influenced by Callmann’s theories of
dilution, 60 or that those supporting the bill were, in fact, seeking
full anti-dilution protection, as distinct from expanded unfair
competition protection. 61 A more cautious reading of the
contemporaneous evidence suggests that the 1947 Massachusetts
statute was intended only to codify existing federal and state
judicial decisions that established that actionable “confusion” could
be found even in cases of non-competing goods. 62 Support for this
reading is contained in the 1948 decision of the U.S. District Court
in Food Fair Stores, Inc. v. Food Fair, Inc., 63 where Judge
Wyzanski interpreted the “injury to business reputation” element
of the statute as follows:
If one were to read that phrase without any acquaintance with
the prior case law and the mischief sought to be remedied, it
would be possible to construe it as affording protection only in

59. For detailed analysis, see Milton Handler & Charles Pickett, Trade Marks and
Trade Names: An Analysis and Synthesis: II, 30 Colum. L. Rev. 759, 777-82 (1930).
60. Cf. Bone, Schechter’s Ideas, supra note 31, at 498-501 (making a circumstantial
case for Callmann’s influence).
61. Cf. id. at 501-03 (suggesting that the bill was “sold to” the legislature as one that
merely expanded unfair competition law). Perhaps the strongest argument that the
legislature intended to introduce “full” anti-dilution protection is that “the original bill, as
first introduced into the House, was limited to ‘any coined or peculiar word,’ just as
Schechter had intended”: id. at 503 n.166. However, even this can potentially be explained
on the basis that to the extent that courts in New York had recognized the harm of dilution
in unfair competition cases involving non-competing goods, they had also recognized that
“[t]here may be situations in which protection should not be projected beyond the field of the
plaintiff’s exploitation of his mark,” for example, where “the mark may be so nondescript as
to render further dilution of its selling power improbable”: Phila. Storage Battery, 163 Misc.
at 56 (emphasis added).
62. Similarly, the adoption of the statute in New York might have been said to have
entrenched the dilution-influenced approach taken in Tiffany and Phila. Storage Battery,
rather than ushering in “full” anti-dilution protection by stealth.
63. 83 F. Supp. 445 (D. Mass. 1948).
654 Vol. 106 TMR

a case where it was probable that defendant would render

inferior services or sell inferior goods which would make
customers think less of plaintiff. That is, plaintiff would be
required to show—what has not been shown at bar—that
defendant’s services or goods were below plaintiff’s standard
. . . But such a literal reading is not warranted because the
statute was obviously designed to make the Massachusetts
law go at least as far as the unfair competition rules laid down
in federal courts . . . 64
This more cautious reading also chimes with the way the
Massachusetts statute, and similar state anti-dilution statutes,
were subsequently interpreted. For many years, Massachusetts
and other state courts read a requirement of confusion into the
state statutes in cases involving non-competing goods (whether the
harm was claimed to be injury to business reputation or
dilution). 65 A notable example from the 1950s was Esquire, Inc. v.
Esquire Slipper Manufacturing Co., where the owner of the
stylized mark ESQUIRE for magazines was unsuccessful under
Massachusetts law in preventing the use of “Esquire” in relation to
slippers. 66 Gerard Magliocca describes this result as “rather
astounding,” 67 and both this decision, and others of state courts
interpreting their anti-dilution statutes in the 1950s and 1960s,
received sharp criticism from commentators in favor of strong anti-
dilution protection. 68 These decisions seem much less surprising,
however, if one takes the view that the courts understood their role
as applying an “expanded” notion of trademark and unfair
competition law, one that continued to require a showing of
likelihood of confusion, but that now provided protection against
injuries to business reputation and also dilution of the
distinctiveness of the mark. Indeed, there is evidence for this in
Esquire itself, where Judge Woodbury held that “the statute was

64. Id. at 451 (referring to Harrington, supra note 38, at 492). See also Middleton,
supra note 45, at 182 (considering that, under the statute, “injury to the plaintiff’s business
reputation . . . could arise only out of confusion as to the defendant’s identity”).
65. See Magliocca, supra note 36, at 1006; Burstein, supra note 6, at 1195. This can
arguably be seen in the earliest case to have considered the 1947 Massachusetts statute,
265 Tremont St. v. Hamilburg, 321 Mass. 353 (D. Mass. 1947). Although the action was
brought before the coming into force of the 1947 statute, the Supreme Judicial Court
appeared to look forward to the impact of the new statute in holding that there was nothing
about the defendant’s use of “Shubert’s Jewelry Store” that would suggest “it was owned,
operated, or sponsored by” the owner of the “Shubert Theatre,” and that the absence of
confusion was fatal, “even though the absence of competition be treated as no objection”: id.
at 357.
66. 243 F.2d 540 (1st Cir. 1957).
67. Magliocca, supra note 36, at 1006.
68. See, e.g., Derenberg, supra note 39, at 458. See also Beverly W. Pattishall, The
Dilution Rationale for Trademark—Trade Identity Protection, Its Progress and Prospects, 71
Nw. U. L. Rev. 618, 624 (1976).
Vol. 106 TMR 655

meant to accomplish no more than to permit injunctive relief in

Massachusetts in suits for trade-mark infringement or unfair
competition grounded on dilution of a plaintiff’s mark as well as in
such suits grounded on direct infringement,” 69 where “a dilution
case” was understood as simply being one where the parties were
“not directly competing with one another in the sale of similar
goods.” 70
“Anti-dilution” decisions up to the time of the passage of the
FTDA produced “a confusing body of case law,” 71 as courts
grappled with the meaning and scope of the statutes. This case law
was complicated by two related factors. One was that actions were
frequently brought on grounds of both federal trademark
infringement and violation of the applicable state anti-dilution
statute, with courts tending to focus on the former and tending not
to engage in sustained, independent analysis of the latter. The
second factor was that, in considering claims brought under state
statutes, courts started to read the concept of “dilution of . . .
distinctive character” broadly, in a manner that encompassed
reputational injury (helping to explain how tarnishment became
known as a type of dilution). 72 What is most striking about
decisions under the anti-dilution statutes, however, at least up to
the late 1970s, is that the imposition of liability still required a
finding of confusion. Cases that are often cited as early examples
of dilution by tarnishment, such as the (in)famous “Enjoy
Cocaine” 73 and Dallas Cowboy Cheerleader 74 cases, involved claims
for both federal trademark infringement and breach of a state anti-
dilution statute, and were, in fact, determined to have involved
infringing conduct, on the basis of the existence of a likelihood of
confusion.

69. Esquire, 243 F.2d at 544.

70. Id. at 542. See also George E. Middleton, Some Aspects of Trademark Dilution, 47
TMR 1023, 1031-33 (1957) (accounting for the early lack of judicial enthusiasm for ‘full’
dilution on the basis that the harms sought to be prevented were adequately addressed by
the confusion-based infringement action).
71. David S. Welkowitz, Reexamining Trademark Dilution, 44 Vand. L. Rev. 531, 533
(1991).
72. Magliocca, supra note 36, at 1010-12 has traced the origins of the classification of
tarnishment as a type of dilution to Note, Dilution: Trademark Infringement or Will-O’-The-
Wisp?, 77 Harv. L. Rev. 520 (1964), where the author claimed that “dilution threatens two
separable but related components of advertising value. Junior uses may blur a mark’s
product identification, or they may tarnish the affirmative associations a mark has come to
convey”: id. at 531. This definition was quoted by Callmann in the subsequent edition of his
treatise: 3 The Law of Unfair Competition, Trademarks, and Monopolies § 84.2 (3d ed.
1967), which was in turn adopted by courts (calling it Callmann’s definition) by the early
1980s. See, e.g., Sally Gee, Inc. v. Myra Hogan, Inc., 699 F.2d 621, 625 (2d Cir. 1983).
73. Coca-Cola Co. v. Gemini Rising, Inc., 346 F. Supp. 1183 (E.D.N.Y. 1972).
74. Dallas Cowboys Cheerleaders, Inc. v. Pussycat Cinema, Ltd., 604 F.2d 200 (2d Cir.
1979).
656 Vol. 106 TMR

It was only in the early 1980s that courts started to grant

relief under state statutes based on tarnishment, absent any
showing of confusion. Strikingly, in only five cases prior to the
passage of the FTDA in 1995 was tarnishment found in the
absence of confusion, with the conduct in question involving well-
known trademarks being associated with an unholy trinity of sex, 75
drugs, 76 or lawnmowers. 77 Since the passage of the TRDA, the only
successful stand-alone tarnishment actions have also involved
either sexual 78 or drug-related 79 associations. While the courts
have asserted that, at the level of principle, “tarnishment is not
limited to seamy conduct,” 80 the case law on stand-alone
tarnishment actions suggests otherwise.
It is thus the case that the injury to business reputation or
dilution by tarnishment action has played only an extremely
limited independent role, aside from the confusion-based
trademark infringement action. This raises questions about the

75. Pillsbury Co. v. Milky Way Prods., Inc., 215 U.S.P.Q. 124 (N.D. Ga. 1981)
(plaintiff’s corporate mascot shown engaging in sex acts in a pornographic magazine
considered to injure the plaintiff’s business reputation and dilute the mark’s distinctive
character); American Express Co. v. Vibra Approved Labs. Corp., 10 U.S.P.Q.2d 2006
(S.D.N.Y. 1989) (American Express’s slogan “Never Leave Home Without It” used on
condoms considered to be dilution); Eastman Kodak Co. v. Rakow, 739 F. Supp. 116
(W.D.N.Y. 1989) (use of “Kodak” by a comedian in a comedy act relating to bodily functions
and sex held to constitute dilution by way of both blurring and tarnishment (with the court
citing “Callmann’s definition,” supra note 72)).
76. Coca-Cola Co. v. Alma-Leo USA, Inc., 719 F. Supp. 725 (N.D. Ill. 1989) (involving a
bubble gum product in the form of a white powder resembling cocaine sold in a plastic
container shaped like a Coca-Cola bottle. This was considered to dilute the distinctiveness of
Coca-Cola’s mark and injure Coca-Cola’s reputation).
77. Deere & Co. v. MTD Prods., Inc., 41 F.3d 39, 43 (2d Cir. 1994). This decision,
involving a comparative advertisement that made mocking use of an animated deer taken
from John Deere’s logo, only came to be classified as a tarnishment case in Hormel Foods
Corp. v. Jim Henson Prods., Inc., 73 F.3d 497 (2d Cir. 1996). In neither decision was it made
clear why the defendant’s alteration of the John Deere mark risked the possibility that
consumers would come to attribute unfavourable characteristics to the mark and associate
it with inferior goods and services.
78. V Secret Catalogue, Inc. v. Moseley, 605 F.3d 382 (6th Cir. 2010) (VICTORIA’S
SECRET held to have been tarnished by use of “Victor’s Little Secret” and “Victor’s Secret”
in relation to a store selling goods including lingerie and sex toys); Univ. of Kansas v. Sinks,
644 F. Supp. 2d 1287 (D. Kan. 2008) (University of Kansas marks used on T-shirts in
context of sexual activity and obscenity). See also Anheuser-Busch, Inc. v. Andy’s
Sportswear, Inc., 40 U.S.P.Q.2d 1542 (N.D. Cal. 1996) (owner of BUDWEISER successful in
securing a temporary restraining order in relation to “Buttwiser” T-shirts); Rolls-Royce
Motor Cars Ltd. v. Davis, Civ. No. 15-0417 (D.N.J., Mar. 11, 2016) (owner of ROLLS-
ROYCE obtained default judgment against a hip-hop artist called “Rolls Royce Rizzy,” with
the judge noting the use of the artist’s name in an advertisement featuring a scantily clad
woman and advertising a “Booty Shaking Contest,” and on the cover of the artist’s album,
titled “Pimp’n”).
79. Lorillard Tobacco Co. v. Cal. Imports, LLC, 886 F. Supp. 2d 529 (E.D. Va. 2012)
(use of “Newprot” in relation to a product associated with synthetic marijuana said to have
tarnished plaintiff’s NEWPORT mark for cigarettes).
80. Hormel, 73 F.3d at 507.
Vol. 106 TMR 657

real intent of the original anti-dilution statutes, and the extent to

which they were ever intended to capture conduct without the
requirement of source confusion. Further, it needs to be asked
whether the independent role that the tarnishment action has
come to play is in fact justifiable, which involves returning in Parts
III and IV to some of the cases mentioned above, in considering
how the reputations of the marks were said to have been harmed
in the absence of source confusion.

C. Rethinking the History of Tarnishment

under Benelux Law
There are some notable parallels between the U.S. history
outlined in Part II.B and the history of the equivalent Benelux law
(the BMW), particularly when this is read in light of the pre-
Benelux Dutch law. As noted in Part II.A, understanding the
development of anti-dilution laws in the Benelux countries in the
1960s and 1970s is particularly important because of the way
these laws helped shape the anti-dilution provisions of the original
European Trade Marks Directive.
Until the early 1970s in the Netherlands, protection for
trademarks was afforded primarily through the Merkenwet of
1893. At the time the uniform Benelux law was being negotiated,
Article 10 of the Merkenwet afforded trademark owners the right
to prevent the use of similar marks on the same kind of goods.
Following a 1956 amendment, trademark owners also had the
right to prevent the use of similar marks in a manner that would
cause confusion as to the origin of the goods. Article 4bis(3) of the
Merkenwet applied in a similar manner to prevent the registration
of such problematic marks. 81 In applying the “confusion” ground,
courts examined not only how closely related the two sets of goods
were and the nature of the potential consumers of such goods, but
also the fame and distinctiveness of the earlier mark. 82 Owners of
well-known marks, separate from any rights they had under the
Merkenwet, could also bring actions for unfair competition under
Article 1401 of the Dutch Civil Code of 1838. 83 By the mid-1950s
such actions for unfair competition had extended to cases of

81. See generally L. Wichers Hoeth, Letter from The Netherlands, 58 TMR 31, 42
(1968).
82. Id. at 43-44; see also T.J. Dorhout Mees, Kort Begrip van het Nederlands
Handelsrecht 131-32, 134-36 (4th ed. 1964) (in Dutch).
83. See L. Wichers Hoeth, Note to Claeryn/Klarein [1975] N.J. 472, para. 1 (in Dutch).
Although art. 1401 required a party that had committed “an unlawful act” to compensate
the injured party, from 1919, following the decision of the Dutch Supreme Court in
Lindenbaum/Cohen [1919] N.J. 161, actions were able to be brought in relation to conduct
that was “contrary to good morals.” See Antoine Hol, Fault in Legal Doctrine in the
Netherlands, in The Development and Making of Legal Doctrine, 164, 168 (Nils Jansen, ed.,
2014).
658 Vol. 106 TMR

“slavish copying” of product appearance, provided such acts were

likely to deceive the public. 84 Thus, by around that time, it was
possible to point to cases where owners of well-known marks had
been able to prevent the registration or use of identical or similar
marks on dissimilar goods, 85 although it could not be said with
certainty how far such protection might extend.
By way of contrast, it appears that the trademark laws of both
Belgium and Luxembourg were not nearly as advanced as those in
the Netherlands. In particular, it appears that under Belgian law
a trademark owner could prevent the use of similar marks only in
relation to the same kind of goods. 86
Article 13A of the BMW, set out in Part II.A, established a
differently worded test for infringement compared with the
infringement provisions of the former laws of the Netherlands,
Belgium, and Luxembourg. Each article of the new BMW
contained its own explanatory notes, and the notes to Article 13
began by stating that the new provision was intended to provide
broader rights than those available under the national laws that
were being superseded. Exactly how much broader, keeping in
mind the differences between the three countries’ former laws,
was, however, not specified. The explanatory notes indicated the
limits placed on the new article: it was emphasised that the right
to prevent “any other use without due cause in the course of trade
. . . that would cause damage to the owner of the mark,” would only
catch uses taking place in commerce (as distinct from uses in
dictionaries and scientific publications). But, significantly, both the
text of Article 13 and the explanatory notes were silent on a
number of key questions. It was unclear whether Article 13A was
intended to apply to all marks, or only those with particular
reputations. There was no guidance on what, exactly, was meant
by the “due cause” exception. And, above all, nothing in either the
text or the notes indicated what sort of “damage” was
contemplated, and what sort of evidence would be needed to show
such damage. As one commentator asked, if protection against
damage was to extend to the impairment of distinctiveness of the

84. See Hyster Karry Krane [1954] N.J. 90 (Dutch Supreme Court). See generally
Rogier de Vrey, Towards a European Unfair Competition Law: A Clash between Legal
Families 114-23 (2006).
85. For example, the registration of “Steiner” for toothpaste was annulled on the basis
of confusion with an earlier identical mark used in relation to paper towels and associated
containers ([1962] B.I.E. 88, Court of The Hague), while the following applications for
registration were refused: “Chevrolet” and “Buick” for watches ([1952] B.I.E. 122, Court of
The Hague); “Lucky Smile” for chewing gum, in light of “Lucky Strike” for cigarettes ([1954]
B.I.E. 101, Court of s’Hertogenbosch); and “Pontiac” for refrigerators ([1965] B.I.E. 240,
Court of The Hague).
86. Wichers Hoeth, Note to Claeryn/Klarein, supra note 83, at para. 3 (“deze stap was
intussen voor Belgie en Luxembourg groter dan voor Nederland, omdat in die landen de
‘règle de la spécialité’ nog volop werd toegepast”).
Vol. 106 TMR 659

mark, or injury to reputation, to what extent was this protection to

be conditioned on a finding of a likelihood of confusion as to the
source of the dissimilar goods? 87
Early in the negotiating history of the uniform law,
statements were made by a group assembled under the auspices of
the International Association for the Protection of Intellectual
Property (“AIPPI”) suggesting that what was intended was a new,
broad right against “usurpation” of a mark. 88 The concern
expressed by these members of the AIPPI was that “the weakening
of the attractive power of a famous brand by its use on dissimilar
products” was an unfair action that would cause harm to the
trademark owner, against which the owner ought to be protected
under the new law. 89 This might suggest that “full” dilution
protection was intended to be part of the BMW from the very start.
It is, however, worth noting that the AIPPI’s statements were
made prior to the expansion of Dutch trademark law in 1956 by
the addition of “confusion” to the infringement test, which provided
stronger protection for owners of well-known marks. An even more
telling point is that the drafters of what became Article 13A of the
BMW shied away from using the specific language of “dilution” or
“usurpation,” instead adopting the more neutral language of
“damage.” This was so notwithstanding that the concept of
“usurpation” was in use in international intellectual property law
as a recognised type of harm for which protection was required
even in the absence of confusion. 90 This is not to suggest that
dilution or usurpation were not contemplated by the drafters, but
rather to emphasize that a decision appears to have been made to
leave the full scope of the new article to be worked out by courts
over time.
This was also reflected in the lack of consensus in academic
and practitioner publications in the Benelux countries in the early
1970s concerning the potential scope of the new Article 13A.
Antoine Braun, writing from a Belgian perspective, anticipated
that the new “damage” condition would be fulfilled “each time the
power of attraction of the famous mark is weakened as a result of

87. Rudolf Kraβer, The Extent of Trademark Protection in the Benelux Trademark Act,
7 Int’l Rev. Indus. Prop. & Copyright L. 358, 360 (1976).
88. See the 1954 explanatory statement prepared by a Benelux group of the AIPPI (in
French), reported in [1954] B.I.E. 17, and cited in Braun, supra note 26, at 189.
89. Cited in Braun, supra note 26, at 189 (“l’affaiblissement du pouvoir attractif d’une
marque célèbre par son usage pour des produits non similaires . . . autant de manoeuvres
déloyales de nature à causer préjudice au titulaire de la marque”).
90. See Lisbon Agreement for the Protection of Appellations of Origin and their
International Registration of 1958, art. 3 (in the context of a provision requiring parties to
afford protection to registered appellations of origin). See generally Dev Gangjee, Relocating
the Law of Geographical Indications ch. 4 (2012).
660 Vol. 106 TMR

‘dilution.’” 91 Some commentators, such as the Dutch practitioner

Wichers Hoeth, criticized the fact that it would be left up to
individual judges to determine whether actionable damage would
be incurred in any given case, 92 and expressed concern that the
“exceptions” carved out for non-commercial use and for use with
“due cause” were overly generous. 93 But others, such as Koman
and Verkade, while noting that the wording of the new Article 13A
was broader than the infringement provisions under former Dutch
law, anticipated that courts would apply much the same criteria in
interpreting the new article. 94 The only thing that can be said with
certainty is that the precise scope of Article 13A—in particular, the
circumstances in which use on dissimilar goods would “cause
damage” to the mark owner—was to be left to the courts.
It might have been expected that this issue would have been
addressed in a slow, incremental manner by the judiciary. Instead,
the Benelux Court of Justice, in its very first decision in
Claeryn/Klarein, gave close to the widest possible reading of
Article 13A(2). Colgate-Palmolive had sought to enter the Dutch
market with a detergent product called “Klarein.” Lucas Bols Co.,
the Dutch owner of the phonetically identical mark CLAERYN for
genever (a high-quality gin), sought to restrain Colgate-Palmolive
from advertising and selling its product, which, given the
dissimilarity of the goods, required Lucas Bols to make a case
under Article 13A(2). The Court of Appeals in Amsterdam found in
favor of Lucas Bols. On appeal, the Dutch Supreme Court referred
a number of questions about the scope of Article 13A to the
Benelux Court of Justice. Noting that the intent of the BMW was
to expand the protection given under former law, the Court of
Justice asserted, without explanation, that this meant that the
new law was designed to go beyond situations involving confusion,
and instead covered conduct that impaired the “attraction” of the
mark and the “capacity of the mark to stimulate the desire to buy”
the goods for which it was registered. 95 The Court held that the
protection afforded by Article 13A(2) was not limited to famous
marks, although the fame of the mark might be deemed relevant
in determining whether the defendant had impaired the “capacity

91. Braun, supra note 26, at 222 (“on peut estimer que chaque fois que le pouvoir
d’attraction de la marque célèbre sera affaibli, suite à la ‘dilution’ de la marque dans le
grand public, cette condition sera remplie”).
92. Wichers Hoeth, Note to Claeryn/Klarein, supra note 83, at para. 3 (“de vraag of een
bepaald merkgebruik (bijv. voor ongelijksoortige waren) schadelijk is in de zin van art. 13
kan door onderscheidene rechters zeer wel verschillend worden beantwoord”).
93. L. Wichers Hoeth, Kort Commentaar op de Beneluxmerkenwet 90 (1970) (in
Dutch).
94. A. Koman & D.W.F. Verkade, Het Nieuwe Merkenrecht 21-22 (1970) (in Dutch).
95. Claeryn/Klarein (1976) 7 I.I.C. 420, 423.
Vol. 106 TMR 661

of the mark to stimulate the desire to buy.” 96 The Court further

held that the “without due cause” exception would apply only in a
case where the defendant was under a compulsion to use the mark,
or could show a prior right to the mark (but not including that the
mark had been used in another country). 97
The breadth of this decision surprised Dutch commentators, 98
and it was sharply criticized by Rudolf Kraβer in a 1976 article.
Kraβer argued that the Court of Justice’s reasoning, based on the
explanatory notes, as to the scope of Article 13A(2) was
unconvincing, since that reasoning relied on an assumption that
the former national laws of all three countries already provided for
full protection against the risk of confusion (that is, including in
cases involving non-competing goods). 99 More tellingly, Kraβer was
particularly concerned that the Court had failed to appreciate how
wide a scope it had been given to put flesh on the bones of the new
Article 13A(2). More specifically, Kraβer argued that it had been
left to the Court to explain the relevance of the trademark’s
reputation, to clarify how the lower courts were to assess whether
conduct would cause detrimental associations at a level
warranting legal intervention, and to impose appropriate
boundaries upon the right in order to ensure that it would not
“develop into an instrument for unlimited expansion of trademark
protection.” 100 Kraβer’s criticism is a strong reminder that Article
13A was capable of multiple interpretations, and that the Court of
Justice’s holding that “damage” could be divorced entirely from
consumer confusion was by no means compelled.
Unsurprisingly, however, the Claeryn/Klarein decision
imposed upon lower courts in the Benelux countries an
exceptionally wide reading of the “damage” requirement in Article
13A(2), subject to few limits. Similarly broad approaches were
taken by Dutch courts in cases in the 1970s and 1980s involving
successful actions by the owner of SKOL for beer, enjoining “Sjolk”
for chocolate milk; 101 by the Coca-Cola Company, restraining the
distribution of a movie featuring a Coca-Cola bottle used as part of
a sex act; 102 and by the owner of BLUE BAND, a well-known mark
in the Netherlands for margarine, preventing a blues band from

96. Id. at 424.

97. Id. at 425.
98. See D.W. Feer Verkade, De Zaak Claeryn/Klarein (Benelux-Gerechtshof 1 Maart
1975, Zaak A/74/1), in Liber Amicorum Ludovic De Gryse, 331 (Brigitte Dauwe et al., eds.,
2010), at [Link] (in Dutch).
99. Kraβer, supra note 87, at 361.
100. Id. at 362-64.
101. [1979] B.I.E. 11 (District Court of Zutphen).
102. [1977] N.J. 59 (Court of Amsterdam).
662 Vol. 106 TMR

continuing to use the name “The Blue Band.” 103 In a particularly

striking case, the owner of the mark POUR VOUS MADAME for
stockings obtained an injunction to prevent the use of the mark on
a magazine devoted to perfume. 104 The Dutch Supreme Court
avoided the issue of “damage” under Article 13A(2) altogether,
holding that it was sufficient to find that damage could be caused,
and unnecessary to determine whether and how it might occur. 105
As noted in Part II.A, Article 13A of the BMW came to provide
a partial template for the anti-dilution provision of the original
European Trade Marks Directive. Just as importantly, the
Benelux Court of Justice’s interpretation of Article 13A in
Claeryn/Klarein remains relevant under the harmonized
European trademarks regime. Since the original Directive came
into force, English and Dutch commentators, 106 English courts, 107
and the Advocate General for the European Court of Justice 108
have all cited Claeryn/Klarein as an example of what might
constitute “detriment to repute” as set out in the Directive.
Because of the ongoing significance of this case, it is vital to
appreciate the historical context in which it was made. The
Benelux Court of Justice’s reading of Article 13A in
Claeryn/Klarein was only one of several possible interpretations
available, and one that did not closely address the ways in which
reputational “damage” could be caused, something that the text of
the article had deliberately left open. Indeed, as will be seen in
Part IV, a critical re-reading of Claeryn/Klarein shows that the
decision, rather than being an uncontroversial illustration of
“detriment to repute,” in fact should make us call into question
whether harm to reputation in the absence of source confusion is of
sufficient gravity to warrant a legal remedy at all.

103. [1981] B.I.E. 266 (District Court of Arnhem). For further examples, see Feer
Verkade, De Zaak Claeryn/Klarein, supra note 98, at 337-38; D.W. Feer Verkade, Unfair
Use of and Damage to the Reputation of Well-Known Trademarks, Trade Names and
Indications of Source—A Contribution from the Benelux, 17 Int’l Rev. Indus. Prop. &
Copyright L. 768, 771 (1986).
104. Pour Vous Madame [1984] B.I.E. 257.
105. Id., discussed in Feer Verkade, Unfair Use, supra note 103, at 772.
106. See, e.g., Lionel Bently & Brad Sherman, Intellectual Property Law 1005 (4th ed.
2014); Helen Norman, Intellectual Property Law 341 (2d ed. 2014); Tobias Cohen Jehoram,
Constant van Nispen & Tony Huydecoper, European Trademark Law 318 (2010).
107. See Simon Fhima, Trade Mark Dilution, supra note 9, at 179 n.74 (collecting cases).
See also Simon Fhima, Exploring, supra note 29, at 28 (noting that the U.K. courts have
“sought inspiration” from Claeryn/Klarein and suggesting that this is “understandable”
given that the EU anti-dilution provision does not turn on the existence of confusion).
108. Case C-408/01, Adidas-Salomon AG v. Fitnessworld Trading Ltd. [2003] E.C.R. I-
12537, para. 38 (A-G Jacobs).
Vol. 106 TMR 663

D. Summation
Dilution by tarnishment, as a stand-alone and actionable
claim in the absence of consumer confusion, has taken on a life of
its own since the passage of the first United States and Benelux
statutes. But its origins are murkier than are often recognized.
There is some irony in a statement by commentator Patrick
Harrington, writing immediately after passage of the 1947
Massachusetts statute, that “[a]ny legislative attempt to lay down
principles governing so complex a field as unfair competition runs
the risk that the statute will be too inflexible to furnish an
adequate solution for the varied situations which will arise.” 109
This brief historical survey reveals something of the opposite
problem. The statutes’ imprecise language gave courts the license
to interpret them more broadly than was arguably ever intended
(in the case of U.S. law), or without supplying a clear normative
basis for extending the law in the manner that occurred (in the
case of the Benelux law). It is to these normative issues that I will
now turn in Parts III and IV, in considering whether current laws
against tarnishment are justifiable.

III. CONSTRUCTING REPUTATION

Surveying the current form of tarnishment laws across
jurisdictions, two interesting points of convergence emerge. First,
these laws protect only those marks that have reached a particular
level of fame and, second, they render unlawful conduct that
harms the “reputation” of the plaintiff’s mark.
To reiterate, for a claim to succeed under United States
federal law, there must be a finding of an association arising out of
the similarity between marks that “harms the reputation of the
famous mark.” In the European Union, both the original and the
2015 Trade Marks Directives render unlawful the defendant’s use
of a sign identical with or similar to a mark that has a “reputation
in a Member State” and is “detrimental to . . . the repute of the
trade mark.” Notably, the term “reputation” is used in two
different ways in the EU provision: first, to indicate a threshold
level of fame; 110 and second, to indicate the “esteem” commonly
afforded the mark. These concepts are more closely intertwined
than even the common language suggests.

109. Harrington, supra note 38, at 491. Harrington was of the view that the 1947 statute
was only intended to align state unfair competition law with the more “liberal” position
under federal and New York law: id. at 492.
110. The European Court of Justice has interpreted this to mean that the mark must be
known among a “significant” proportion of the relevant public, which only need be in a
substantial part of the Member State: see Case C-375/97, Gen. Motors Corp. v. Yplon SA
[1999] E.C.R. I-5421, paras. 26 and 28. For Community/EU Trade Marks, the reputation
must subsist in a substantial part of the EU: see Case C-301/07, PAGO Int’l GmbH v.
Tirolmilch registrierte Genossenschaft mbH [2009] E.C.R. I-9429 (E.C.J.).
664 Vol. 106 TMR

The concept of “reputation” in the action for dilution by

tarnishment is one that tends to be glossed over by courts and
commentators, but it needs to be carefully unpacked. In this Part
III, I seek to analyze the way in which “reputation” has been
interpreted (or, more accurately, how courts have allowed it to be
constructed) in tarnishment actions. As will be seen, courts,
particularly in the United States, have tended to apply a highly
reductive notion of “reputation” that is ultimately closer to one of
“positive brand identity,” that is, an image said to have been
created by the mark owner and fostered through advertising
expenditure and investment. Such an approach has arguably been
fortified by the existence of the “fame” threshold, which allows
certain assumptions to be made about the value of those marks
that qualify for anti-dilution protection.

A. Heterogeneous Reputation versus Brand Identity

Reputation is inherently social and relational. In the case of
trademarks, a mark’s reputation depends on how it is perceived
and understood by consumers or society at large, a perception that
can be shaped, in part, by the mark’s owner. Such perception
shaping is not accomplished simply through the owner’s ability to
exercise control over the quality of the goods or services provided
under the mark, but also through the owner’s creation of a cluster
of associations in marketing its goods or services, which, in turn,
attempt to generate ongoing emotional and economic responses
among consumers. In marketing literature, these latter activities
would be called the creation of a “brand identity.” 111 This is a
concept that covers the construction of an image, containing a set
of attributes and qualities, that is transmitted to (passive)
consumers and that, it is hoped, will then “trigger predictable
affective and cognitive effects.” 112
Brand identity is not, however, coterminous with reputation.
Since, by definition, reputation refers to the qualities by which
something is known, assessing the reputation of a mark requires
an understanding of its reception. This cannot be assessed simply
by reference to sales figures or market share of goods or services
provided under the mark: it also requires a sophisticated
appreciation of how the mark has been received, interpreted, and
engaged with by consumers.
What this means is that the reputation of a mark, once it
enters the field of commerce, cannot be completely controlled. Nor

111. Shiva Nandan, An Exploration of the Brand Identity–Brand Image Linkage: A

Communications Perspective, 12 J. Brand Mgmt. 264 (2005).
112. Dev S. Gangjee, Property in Brands: The Commodification of Conversation, in
Property Concepts in Intellectual Property Law, 29, 31 (Helena R. Howe & Jonathan
Griffiths, eds., 2013).
Vol. 106 TMR 665

is it likely to be homogenous or easily defined. Indeed, the

reputations of the sort of well-known marks that the law privileges
with anti-dilution protection are likely to be highly multivalent.
The same mark might be understood in vastly different ways, and
generate a range of different responses among consumers. A
mark’s reputation will be the product of a wide range of factors,
including consumers’ personal experiences with products sold
under the mark, their knowledge of the owner’s practices, their
political preferences, and their engagement with the mark in
culture and as mediated by culture. The active interplay among
these factors is likely to mean that a mark’s reputation, rather
than being static, will be a dynamic and protean phenomenon, and
therefore difficult to reduce to a simple object of legal protection. 113
For a decision-making body to do justice to a socially
constructed, heterogeneous notion of “reputation,” in considering
an action seeking injunctive relief, or a proceeding opposing
registration or seeking invalidation of registration of a mark,
would be a challenging exercise. This is especially so given the
limited evidence likely to be proffered by the parties appearing
before it in a given case. 114 It is, nonetheless, extraordinary that
courts and other decision-making bodies have almost completely
disregarded the complexity of the issue of reputation in
tarnishment cases. Instead, they have been prepared to accept a
constructed, sanitized version of “brand identity” built up through
investment and marketing success as being synonymous with
“reputation,” 115 in a manner consistent with much of the pro-
dilution literature examining this issue. 116
The case law provides useful illustrations. In the long-running
Moseley litigation, once the case was remanded to the District
Court in the late 2000s, only one piece of evidence was cited by the
court in determining whether the reputation of the VICTORIA’S
SECRET mark had been harmed by the defendants’ use of

113. See generally Rosemary J. Coombe, The Cultural Life of Intellectual Properties:
Authorship, Appropriation, and the Law 269-70 (on the social practices through which
brand meanings are generated); Douglas B. Holt, How Brands Become Icons: The Principles
of Cultural Branding 3 (2004) (noting how brand meaning is constructed by a range of
stakeholders, including the culture industries, retailers and consumers as both individuals
and communities); Hope Jensen Schau, Albert M. Muñiz, Jr. & Eric J. Arnould, How Brand
Community Practices Create Value, 73(5) J. Mktg. Res. 30 (2009) (on consumer involvement
in brand value creation).
114. A similar problem exists in defamation law: see David Rolph, Defamation Law ch. 2
(2015).
115. This is acknowledged, albeit uncritically, in the opening paragraph of Simon
Fhima’s chapter on the topic: Simon Fhima, Trade Mark Dilution, supra note 9, at 159.
116. Two noteworthy examples are Steve Hartman, Brand Equity Impairment—The
Meaning of Dilution, 87 TMR 418, 419-20 (1997) and Patrick M. Bible, Defining and
Quantifying Dilution under the Federal Trademark Dilution Act of 1995: Using Survey
Evidence to Show Actual Dilution, 70 U. Colo. L. Rev. 295, 328 (1999). See also the examples
cited in Rierson, supra note 1, at 275-78.
666 Vol. 106 TMR

“Victor’s Little Secret” as the name of a store selling sex toys. A

Victoria’s Secret corporate officer stated that the company strives
to “maintain an image that is sexy and playful” and that the
company “avoids sexually explicit or graphic imagery.” 117 The
veracity and sufficiency of this evidence was uncritically accepted
by both the District Court 118 and the U.S. Court of Appeals for the
Sixth Circuit, with Judge Merritt of the Sixth Circuit noting the
“favorable symbolism of the famous mark.” 119 In Pfizer, Inc. v.
Sachs, a case in which Pfizer sought protection of its famous
VIAGRA mark against both trademark infringement and dilution
by tarnishment, the U.S. District Court dispensed with the issue of
Viagra’s reputation simply by quoting dicta of the U.S. Court of
Appeals for the Second Circuit to the effect that a trademark’s
“reputation and commercial value might be diminished ... because
the defendant’s use reduces the trademark’s reputation and
standing in the eyes of consumers as a wholesome identifier of the
owner’s products or services.” 120 The term “wholesomeness” was
also used to describe COCA-COLA in the early-1970s “Enjoy
Cocaine” case. 121
The examples above are striking because the reputations of
the plaintiffs’ marks are far more complex than the distilled
versions accepted by the courts. The first two cases are
particularly noteworthy. Victoria’s Secret has been the subject of
sustained criticism for decades over its representations of female
sexuality and what has been argued to be its disingenuous use of
tropes of soft-core pornography in its marketing. 122 Similarly, the
uncritical acceptance of VIAGRA as a “wholesome identifier”
ignores the way the mark has been received, not merely as a
byword for the restoration of male sexual performance, but as “a

117. This evidence was cited in V Secret Catalogue, Inc. v. Moseley, 605 F.3d 382, 391-92
(6th Cir. 2010).
118. V Secret Catalogue, Inc. v. Moseley, 558 F. Supp. 2d 734, 750 (W.D. Ky. 2008).
119. Moseley, 605 F.3d at 389.
120. Pfizer, Inc. v. Sachs, 652 F. Supp. 2d 512, 525 (S.D.N.Y. 2009) (emphasis added).
121. Coca-Cola Co. v. Gemini Rising, Inc., 346 F. Supp. 1183, 1189 (E.D.N.Y. 1972).
122. See Susan Faludi, Backlash: The Undeclared War Against American Women 173
(1991) (arguing that Victoria’s Secret demonstrates restriction and repression of female
sexuality through punitively restrictive clothing); Jane Juffer, A Pornographic Femininity?
Telling and Selling Victoria’s (Dirty) Secrets, 48 Soc. Text 27, 31 (1996) (“the company’s
insistence on the private and intimate maintains a proper image of femininity which
distances itself from pornography even as the catalog builds its profits on its pornographic
overtones”); see also id. at 32 (on how Victoria’s Secret functions as a trope of soft-core
pornography while escaping the label by operating discretely and privately (through mail-
order)); Marie D. Smith, Decoding Victoria’s Secret: The Marketing of Sexual Beauty, 25
Studies in Popular Culture 39 (2002) (Victoria’s Secret’s success is based on trading in male-
constructed image of female sexuality, entrenching the idea that female self-image ought to
be shaped by the physical features that men find desirable). See also Kenneth C.W.
Kammeyer, A Hypersexual Society: Sexual Discourse, Erotica, and Pornography in America
Today 3 (2008).
Vol. 106 TMR 667

euphemism for a pharmacotechnology that confirms hegemonic

masculinity.” 123 However, this concern extends well beyond cases
involving companies that trade off overt images or ideas of
sexuality being allowed to make somewhat specious claims of
“wholesomeness” in cases where their marks are associated with
sex-related goods or imagery. The COCA-COLA illustration is
particularly interesting because of the extraordinarily complex
reputation of the mark. Goods sold under the COCA-COLA mark
clearly enjoy colossal popularity and the mark might well be
associated with ideals of youth, freedom, and global success; but
can the mark not also be said to have been recoded, to some extent,
as a symbol of cultural imperialism, controversial labor practices,
or poor nutrition? 124 The bland adjective “wholesome” does not
come close to encapsulating the reputation of such a mark.
This does not mean that all marks have contentious
reputations, or that some marks cannot have overwhelmingly
“positive” ones. The concern is that, to the extent that courts or
pro-dilution commentators address the question of “reputation,”
they tend to accept unreflectively the notion that brands have
unequivocal, unidirectional meanings, and that “reputation”
consists of only positive attributes, all of which have been
generated and cultivated by the mark owner. This raises problems
at both a normative and doctrinal level.
The normative issue—whether the law ought to intervene to
protect the advertising or investment functions of marks in the
absence of consumer confusion—has been well canvassed in the
literature. 125 But one particular problem that stands out in this
context is the assumption that only the mark owner plays a role in
the creation of a mark’s reputation: that its investment alone
determines the mark’s image and expressive dimensions, and
entitles the owner to reap the full value of its reputation. 126 This
idea of a mark owner as “author” is highly reductive, as is the
underlying assumption that consumers are passive recipients of
pre-generated brand qualities rather than active co-creators of

123. Adele E. Clarke, Laura Mamo & Jennifer Ruth Fosket, Biomedicalization:
Technoscience, Health, and Illness in the US 161-63 (2010). See also Meika Loe, The Rise of
Viagra: How the Little Blue Pill Changed Sex in America 19-21 (2004) (on the drug Viagra
re-establishing “normal” ideas about male and female sexuality and sexual dysfunction, and
“appearing to reinforce long-standing gender stereotypes and thus social inequalities”: id. at
21).
124. For an overview, see Mark Pendergrast, For God, Country and Coca-Cola: The
Definitive History of the Great American Soft Drink and the Company That Makes It (3d
rev. ed. 2013).
125. For two recent examples, see Gangjee & Burrell, supra note 1; Katya Assaf, Brand
Fetishism, 43 Conn. L. Rev. 83 (2010).
126. See generally Gangjee, Property in Brands, supra note 112.
668 Vol. 106 TMR

brand value. 127 The doctrinal issue is more subtle. To elide

reputation and brand identity in deciding a tarnishment action
unduly privileges the mark’s owner by oversimplifying the
ultimate enquiry that must be made: namely, whether the
defendant’s conduct causes an association likely to damage the
mark’s reputation. If reputation is interpreted in a simplistic way,
focusing only on positive attributes as curated by the brand owner,
it is much easier to assume that “negative” associations will cause
the requisite harm. This is a point to which I return in Part IV in
considering the issue of “harm” in more detail.

B. How “Fame” Acts as a Proxy for Brand Identity

The argument that the cause of action for dilution by
tarnishment is concerned more with protecting constructed brand
identity than with protecting “reputation,” properly understood, is
reinforced by the fact that the action is available only to owners of
“famous” marks.
This limitation does not have strong normative underpinnings.
Logically, if the law wishes to provide a cause of action to prevent
associations likely to harm a mark’s reputation, there seems to be
no reason for this to be limited to cases where the mark has
crossed a particular threshold of fame. To return to the Advocate
General’s example cited in the Introduction, why should the Coca-
Cola Company have a right to prevent the use of “Coca-Cola” on
“low-grade engine oils or cheap paint strippers,” but a small,
relatively new business that makes locally sourced organic juice
under a distinctive mark not be able to bring a similar action
against a junior user of the mark on such chemical products?
Presumably the nature of the likely harm in both cases is similar,
albeit of a different degree—something that should logically go to
the scope of the remedy. Indeed, might it not be said that a
dilution by tarnishment action is needed more by those businesses
with only slight or emerging reputations, given that, as the mark’s
reputation grows, so, too, do its owner’s chances of being able to
bring traditional infringement proceedings against a junior user of
the mark in relation to non-competing goods? 128

127. Steven Wilf, Who Authors Trademarks?, 17 Cardozo Arts & Ent. L.J. 1 (1999); Lux,
supra note 1, at 1076-78. The idea has also been problematized in marketing literature: see,
e.g., C.K. Prahalad & Venkat Ramaswamy, Co-opting Customer Competence, 78 Harv. Bus.
Rev. 79 (2000); C.K. Prahalad & Venkat Ramaswamy, Co-creation Experiences: The Next
Practice in Value Creation, 18 J. Interactive Mktg. 5 (2004).
128. See Graeme B. Dinwoodie & Mark D. Janis, Dilution’s (Still) Uncertain Future, 105
Mich. L. Rev., First Impressions 98, 100 (2006); Katya Assaf, Magical Thinking in
Trademark Law, 37 L. & Soc. Inquiry 595, 613-14 (2012). See also Robert Burrell & Michael
Handler, Reputation in European Trade Mark Law: A Re-examination, 17 E.R.A. Forum
(forthcoming 2016).
Vol. 106 TMR 669

Pragmatic and historical factors seem to offer the most obvious

explanation for the “fame” threshold in both U.S. and EU anti-
dilution law. Simon Fhima makes a sound case that the
“reputation in a Member State” requirement was incorporated into
the original Trade Marks Directive as a compromise, with the
Benelux delegation arguing in favor of the wholesale adoption of
Article 13A(2) of the BMW into the Directive, and the other
Member States objecting to something so broad. 129 In the United
States, some state courts in cases from the 1980s involving
allegations of blurring had started to interpret their anti-dilution
statutes as applying only in cases of famous marks, 130 and an
explicit “fame” threshold was recommended by the USTA in a 1987
report 131 and was incorporated into the USTA’s 1992 revisions to
its Model State Trademark Bill. 132 These developments must have
influenced Congress in incorporating a similar threshold in the
TRDA only three years later. In both the European Union and the
United States, however, there is very little evidence of any detailed
consideration of why a requirement of “fame” or “sufficient
reputation” was thought to be an appropriate precondition for an
action to prevent dilution. 133 At best, a superficial case could be
made that a “fame” requirement is a logical extension of the long-
standing idea that only “highly distinctive” or “strong” marks
ought to be entitled to protection against blurring. 134 However,
fame and strength do not always overlap and, in any event, it
remains unclear why the tarnishment action should be limited
even to “highly distinctive” or “strong” marks. A good case can,
instead, be made that the “fame” limitation has been deployed by
proponents of anti-dilution laws as a strategic device to deflect
criticism of the potential breadth of such laws. 135 It becomes more
difficult to argue that a dilution statute is a “rogue law” that
makes a trademark a “potentially anti-competitive weapon” if it

129. Simon Fhima, Trade Mark Dilution, supra note 9, at 24.

130. See, e.g., Mead Data Central, Inc. v. Toyota Motor Sales, USA, Inc., 875 F.2d 1026
(2d Cir. 1989).
131. See United States Trademark Association Trademark Review Commission and
Recommendations to USTA President and Board of Directors, supra note 25, at 456, 459.
132. See Burstein, supra note 6, at 1194.
133. In the United States, the Congressional record is thin. One reason that was given is
that anti-dilution protection for famous marks was needed to ensure compliance with the
TRIPS Agreement: see H.R. Rep. No. 104-374, p. 4, accepted by Simon Fhima, Trade Mark
Dilution, supra note 9, at 24-25. However, this reason reflected a misunderstanding of the
scope of art. 16.3 of the TRIPS Agreement: see Handler, supra note 13, 308-09. For further
criticism of the “fame” requirement, see Klieger, supra note 40, at 846.
134. On the protection of “highly distinctive” and “strong” marks under pre-FTDA state
law, see Welkowitz, Trademark Dilution, supra note 25, at 34-50.
135. Simon Fhima, Trade Mark Dilution, supra note 9, at 22.
670 Vol. 106 TMR

makes confusion-plus protection available only to a limited subset

of marks. 136
But it is what is shared by this “limited subset” of marks that
is of particular importance and might help provide the most
convincing reason for the adoption and entrenchment of the “fame”
requirement in anti-dilution laws. In order to have achieved a
particular level of fame, it is almost inevitable that the mark
owner will have needed to have engaged in significant investment
and expenditure in promoting its mark. What this means is that
by setting a “fame” threshold to qualify for protection, something
that can be established simply by quantitative evidence
demonstrating how well known the mark is, a court or decision-
maker can take for granted that if a mark qualifies for protection,
such investments in the creation of a protectable brand identity
have already been made. “Fame” thus becomes a proxy for positive
brand identity. This, in turn, makes it easier to find that certain
unauthorized associations will be likely to cause the requisite
harm, without requiring the plaintiff to prove the second element
of “reputation” as something conceptually distinct from its fame. 137
One of the clearest examples of the elision of brand identity and
fame is contained in the Advocate General’s opinion in Intel Corp.
Inc. v. CPM United Kingdom Ltd., where it was said that:
the best known trade marks . . . present a powerful image of
quality, exclusivity, youth, fun, luxury, adventure, glamour or
other reputedly desirable lifestyle attributes, not necessarily
associated with specific products but capable of presenting a
strong marketing message in itself. 138
Not only is this statement striking in its assumption that famous
marks convey only inherently desirable qualities, but also in its
suggestion that these qualities are present by virtue of the fact
that the mark is well-known. This approach, equating fame with
“repute” under EU law, can also be seen in decisions of the General
Court 139 and has received the support of some European

136. J. Thomas McCarthy, McCarthy on Trademarks and Unfair Competition, § 24:104

(Westlaw ed. 2016).
137. See Martin Senftleben, The Trademark Tower of Babel—Dilution Concepts in
International, US and EC Trademark Law, 40 IIC: Int’l Rev. Intell. Prop. & Competition L.
45, 72-73 (2009) (criticizing the primarily quantitative approach taken under EU law to
establishing “reputation”).
138. Case C-252/07, [2008] E.C.R. I-8823, at para. 8 (A-G Sharpston).
139. See Case T-215/03, Sigla SA v. Office for Harmonization in the Internal Market
(Trade Marks and Designs) (OHIM) [2007] E.C.R. II-711, para. 35 (“a mark also acts as a
means of conveying other messages concerning, inter alia, the qualities or particular
characteristics of the goods or services which it covers or the images and feelings which it
conveys, such as, for example, luxury, lifestyle, exclusivity, adventure, youth. To that effect
the mark has an inherent economic value which is independent of and separate from that of
the goods and services for which it is registered. The messages in question which are
conveyed inter alia by a mark with a reputation or which are associated with it confer on
Vol. 106 TMR 671

commentators. 140 It also seems to be implicit in some U.S. cases in

which tarnishment was alleged, including Pfizer, Inc. v. Sachs,
that the “fame” requirement does the work of establishing a
positive reputation. 141 This is further evidence, building on the
points made in Part III.A, of how little this area of the law is
interested in the impact of non-confusing uses on “reputation,”
properly understood, and is instead more concerned with the
impairment of the atmospherics of brand identity.

IV. DOES NON-CONFUSING USE

HARM REPUTATION?
Assuming that the dilution by tarnishment action is
ultimately intended to protect self-constructed brand identity, a
much more complex and contentious issue remains, going to the
very heart of the action—the question of what, if any, harm is
caused by a non-confusing use of a famous mark.
Courts’ explanations of what dilution by tarnishment is, and
how it arises, have tended to take a fairly similar form around the
world. In the United States, in two key cases from the mid-1990s,
the U.S. Court of Appeals for the Second Circuit held that “[t]he
sine qua non of tarnishment is a finding that plaintiff’s mark will
suffer negative associations through defendant’s use” 142 and that
dilution by tarnishment may occur if the plaintiff’s mark is “linked
to products of shoddy quality, or is portrayed in an unwholesome
or unsavory context” such that “the public will associate the lack of
quality or lack of prestige in the defendant’s goods with the
plaintiff’s unrelated goods.” 143 In the EU, the European Court of
Justice has held that tarnishment occurs where the defendant’s
use:
may be perceived by the public in such a way that the trade
mark’s power of attraction is reduced. The likelihood of such
detriment may arise in particular from the fact that the goods
or services offered by the third party possess a characteristic

that mark a significant value which deserves protection, particularly because, in most cases,
the reputation of a mark is the result of considerable effort and investment on the part of its
proprietor”).
140. Simon Fhima, Trade Mark Dilution, supra note 9, at 177 (“[o]ne might expect an
automatic assumption that a trade mark, particularly a well-known one, has a positive
reputation”).
141. This also appears to be implicit in Starbucks Corp. v. Wolfe’s Borough Coffee, Inc.,
588 F.3d 97, 111 (2d Cir. 2009) (appearing to be willing to accept Starbucks’ claim that its
coffee was of very high quality).
142. Hormel Foods Corp. v. Jim Henson Prods., Inc., 73 F.3d 497, 507 (2d Cir. 1996)
(emphasis added).
143. Deere & Co. v. MTD Prods., Inc., 41 F.3d 39, 43 (2d Cir. 1994) (emphasis added).
672 Vol. 106 TMR

or a quality which is liable to have a negative impact on the

image of the mark. 144
What is common to these formulations is that more than a link
between the marks is needed. There must be some impact on the
famous mark; a transfer of negative associations that causes
consumers to think differently about the plaintiff’s mark and the
goods or services it provides under that mark, with adverse
consequences for the plaintiff. 145 Given that the tests under both
United States federal law and European Union law look to the
likelihood that such harm will occur (rather than a demonstration
of past tarnishment), this makes it all the more important to
assess the question of when and how, exactly, this negative impact
might occur.
Here, it is worth distinguishing between two types of conduct
by defendants. The first involves use of the famous mark on what
we might call “dissonant” goods or services—these might be poor
quality goods or services, or goods or services that convey
meanings or symbolism antagonistic to the owner’s reputation. In
both cases, the defendant’s use is not intended to make any direct
comment on the mark’s owner. The second category involves
situations where the defendant’s goods might not be dissonant, but
where the plaintiff’s mark has been consciously used (including in
modified form) to refer to and comment on the mark or its owner,
for example by way of critique or parody. The first type of use will
be explored in Part IV.A below to assess the fundamental question
of whether harm is likely to be caused, while the second type of use
more squarely raises the issue of whether there are adequate
safeguards in place to ensure that freedom of expression is not
unduly stifled, which will be addressed in Part IV.B.

A. Use on Dissonant Goods:

Do Harmful Negative Feedback Effects Occur?
It is easy enough to accept that harm to reputation might
occur where the defendant’s use of a mark on inferior quality
goods, or goods antithetical to the mark owner’s business, causes
confusion, whether as to source, sponsorship or affiliation or some
other form of commercial connection between the parties. If such
confusion exists, then it seems probable that a consumer will think
that the owner is in some way responsible for that inferior quality,
or has engaged in a form of “self-tarnishment” by producing a new

144. Case C-487/07, L’Oréal SA v. Bellure NV [2009] E.C.R. I-5185, para. 40 (emphasis
added).
145. This has notably been described as “negative contagion”: Assaf, Magical Thinking,
supra note 128, at 614-17.
Vol. 106 TMR 673

line of goods that conflicts with the defendant’s core business. 146
Anglo-American trademark law has long recognized reputational
harm in such circumstances. For example, in the United Kingdom
and British Commonwealth countries, a passing-off action has long
been available where the injury alleged is damage to reputation,
consequent on a misrepresentation, in cases involving both similar
and dissimilar goods. 147 The same idea can be said to underpin the
British Commonwealth “defensive trade mark” regime, which was
established in the United Kingdom in 1938 and still exists in
countries such as Australia today. Under this scheme, an owner of
a registered mark can separately register that mark for an
extended range of goods, without any intention of using the marks
on those goods, if it can establish that a third party’s use of the
mark on those “extended” goods would indicate a commercial
connection with the owner. 148 In the United States, returning to
the history discussed in Part II.B, courts recognized as early as the
1920s that a major type of harm that flowed from the defendant’s
confusion-generating use of a mark on non-competing goods was
harm to reputation.
To the extent that it encompasses conduct involving confusion
as to source, sponsorship or affiliation, the dilution by tarnishment
action adds nothing to the action for trademark infringement or
passing off. A much more controversial question, therefore, is
whether harm to reputation is likely to occur beyond these
situations. The issue here is whether the “negativity” associated
with or generated by the defendant’s use will be transferred to the
plaintiff’s mark and adversely affect its reputation—that is,
whether there will be negative feedback effects even where
consumers would not consider the plaintiff’s and defendant’s goods
to be connected. It is easy to assert that this might happen and, as
indicated in the Introduction, the idea has strong intuitive appeal.
But is there, in fact, evidence that this is likely to occur?
It might have been expected that there would have been
sustained engagement with this issue in the case law. Instead,
perhaps because the majority of cases in which tarnishment has
had independent work to do in the United States (that is, where
there was no claim for trademark infringement, or where such a
claim failed) have involved marks being associated by defendants
with sexual content or drug use, we see courts assuming the

146. Although Mark McKenna has made persuasive arguments that even accepting
these types of confusion as giving raise to actionable trademark infringement is
problematic: see Mark P. McKenna, Testing Modern Trademark Law’s Theory of Harm, 96
Iowa L. Rev. 63 (2009); Mark P. McKenna, A Consumer Decision-Making Theory of
Trademark Law, 98 Va. L. Rev. 67 (2012).
147. See Bently & Sherman, supra note 106, at 875-76 and the cases cited therein;
Robert Burrell & Michael Handler, Australian Trade Mark Law ch. 12 (2d ed. 2016).
148. See Burrell & Handler, Dilution and Trademark Registration, supra note 1.
674 Vol. 106 TMR

existence of the likelihood of the very harm that needs to be

established.
Moseley is an excellent case study in this regard. The only
evidence proffered by Victoria’s Secret that was potentially
relevant to its claim of tarnishment was a statement by an army
colonel who had come across the defendants’ sex store. He claimed
that he was “offended” by the defendants’ use of a close variant of
the VICTORIA’S SECRET trademark to promote “unwholesome,
tawdry merchandise” and was “dismayed” by the defendants’
“effort to associate [themselves] with, trade off the image of, and in
fact denigrate a store,” namely Victoria’s Secret, frequented by his
family. 149 Without addressing this evidence, the U.S. District
Court assumed the likelihood of harm based on the nature of the
defendants’ business. 150 Judge Merritt of the U.S. Court of Appeals
for the Sixth Circuit went a step further. Drawing on a line of
cases in which associations with sexual subject matter had been
found to constitute tarnishment (but without mentioning that most
of these involved consumer confusion), and misreading the
adoption in the TDRA of a “likelihood of dilution” standard, the
judge indicated there was now a “rebuttable presumption, or at
least a very strong inference, that a new mark used to sell sex-
related products is likely to tarnish a famous mark if there is a
clear semantic relationship between the two.” 151 Although it was
noted that the tarnishing effect of the defendants’ mark was
“somewhat speculative,” the absence of evidence put forward by
the defendants meant that they were unable to rebut the
presumption of tarnishment. 152
The introduction of a rebuttable presumption imposing a
burden on the defendant to adduce evidence of the absence of
something that in cases for injunctive relief is likely to be
speculative (that is, evidence of a lack of likely harm to the
plaintiff’s reputation) has been rightly criticized. 153 There is a
tension here with the Supreme Court’s earlier decision in eBay Inc.
v. MercExchange, LLC, holding that irreparable harm should not
be presumed in cases where a plaintiff seeks injunctive relief for

149. V Secret Catalogue, Inc. v. Moseley, 605 F.3d 382, 391 (6th Cir. 2010).
150. V Secret Catalogue, Inc. v. Moseley, 558 F. Supp. 2d 734, 750 (W.D. Ky. 2008).
151. Moseley, 605 F.3d at 388.
152. Id. at 389.
153. See, e.g., Rierson, supra note 1, at 258; Jennifer E. Rothman, Sex Exceptionalism in
Intellectual Property, 23 Stan. L. & Pol’y Rev. 119, 135-36 (2012); McCarthy, supra note 136,
§ 24:89 (calling it “wildly misguided”). Even commentators otherwise sympathetic to
dilution have been critical: see, e.g., Theodore H. Davis, Jr., United States Jurisprudence
Following the Enactment of the Trademark Dilution Revision Act of 2006, in International
Trademark Dilution, 453, 489 (Daniel R. Bereskin, ed., 3d ed. 2015) (noting the shift in
burden was “without apparent precedent”).
Vol. 106 TMR 675

patent infringement, 154 a decision that has since been applied by

the U.S. Courts of Appeals for the Third and Ninth Circuits in
cases involving preliminary injunctions to restrain trademark
infringement. 155 The majority’s finding in Moseley is doubly
problematic because the Court’s opinion did not consider the
extent to which Victoria’s Secret’s “reputation” was not simply that
of a purveyor of goods marketed in a “sexy and playful” manner, as
it sought to describe itself, but was instead integrally connected to
its commodification of sex. This raises a serious question about
how the defendant’s use on sex-related goods could tarnish such a
reputation, properly understood. 156
These and similar issues as to the operation of the
tarnishment action in sex-related cases are important, 157 but
should not distract us from the more fundamental problem with
Moseley, or indeed other cases in which the harm of tarnishment
has been assumed by the “unwholesome” nature of the defendant’s
goods or services. 158 Laura Bradford is right at a descriptive level
in stating that “[j]udges are willing to forbid uses of marks that are
likely to incite disgust or fear out of concern that such emotions,
once stimulated, will be difficult to put aside even when the
consumer knows the use is unauthorized.” 159 But, at a normative
level, this simply begs the very question that needs to be
answered. How do we know that those feelings of disgust, once
generated, will be transferred and impact on the consumer’s
perception of the famous mark? 160

154. 547 U.S. 388 (2006). See also Winter v. Natural Res. Def. Council, Inc., 555 U.S. 7,
20 (2008) (noting the irreparable harm requirement in preliminary injunction cases).
155. See, respectively, Herb Reed Enters. v. Fla. Entm’t Mgmt., 736 F.3d 1239, 1248-49
(9th Cir. 2013), cert. denied, 135 S. Ct. 57 (2014) (considering that eBay and Winter “cast
doubt on the validity of this court’s previous rule that the likelihood of ‘irreparable injury
may be presumed from a showing of likelihood of success on the merits of a trademark
infringement claim’”); Ferring Pharm., Inc. v. Watson Pharm., Inc., 765 F.3d 205, 216 (3d
Cir. 2014) (following Herb Reed). This development was anticipated by Sandra L. Rierson,
IP Remedies After eBay: Assessing the Impact on Trademark Law, 2 Akron Intell. Prop. L.J.
163 (2008), who argued that it should also apply under the TDRA: id. at 182-84. My thanks
go to Sandy Rierson for drawing my attention to this issue.
156. See Rothman, supra note 153, at 137-38 (also criticizing Pfizer, Inc. v. Sachs, 652 F.
Supp. 2d 512 (S.D.N.Y. 2009) on the basis that “[i]t is confounding how encouraging people
to have sex tarnishes the image of a drug intended to facilitate sex”); Mark Bartholomew,
Trademark Morality, 55 Wm. & Mary L. Rev. 85, 139 (2013) (making a similar criticism
about the Dallas Cowboy Cheerleader case).
157. Rothman, for example, has queried the general assumption that an association
with sex necessarily harms the selling power of a mark, when so much that we know about
advertising suggests otherwise: Rothman, supra note 153, at 138.
158. See, e.g., Rolls-Royce Motor Cars Ltd. v. Davis, Civ. No. 15-0417 (D.N.J. Mar. 11,
2016).
159. Laura R. Bradford, Emotion, Dilution and the Trademark Consumer, 23 Berkeley
Tech. L.J. 1227, 1285 (2008).
160. See Rebecca Tushnet, Stolen Valor and Stolen Luxury: Free Speech and Exclusivity,
in The Luxury Economy and Intellectual Property: Critical Reflections 121, 139 (Haochen
676 Vol. 106 TMR

Judge Moore, in her dissent from the Court’s opinion in

Moseley, recognized the problem. After noting that there was no
evidence that the army officer, or anyone else, had changed their
opinion of Victoria’s Secret as a result of the defendants’ conduct,
the judge stated:
it is simply no more probable that Victoria’s Secret will suffer
reputational harm as a result of the Moseleys’ use of “Victor’s
Little Secret” than it is probable that those who are offended
by “Victor’s Little Secret” will limit their negative impressions
to the Moseleys and refrain from projecting those negative
associations upon Victoria’s Secret. 161
If anything, the army officer’s evidence showed the latter was more
likely. 162 What is particularly interesting about Judge Moore’s line
of reasoning is that it suggests that in any tarnishment case, the
plaintiff should be required to put forward evidence, going beyond
simple speculation, as to why a consumer’s negative impressions of
the defendant would in fact be projected onto the plaintiff or its
mark, when the consumer knows the two parties are unrelated.
That this might be exceptionally difficult for a plaintiff to establish
raises questions about whether it is even plausible that
reputational damage will occur in such situations. 163
Courts in other jurisdictions have attempted to explain in
more detail the reputational damage putatively occurring in cases
of “dissonant” goods. Returning to the Claeryn/Klarein decision,
discussed in Part II.C, the Benelux Court of Justice held that a
finding of damage could be based upon the use of a mark that
appeals to “sensations of the public” in such a way that it
interferes with the mark’s “attraction” and its “capacity to
stimulate the desire to buy the kind of goods for which it is
registered.” 164 This is an extremely broad formulation, because it
goes beyond situations where the defendant’s goods are of inferior
quality or are of a tawdry or unwholesome nature. It contemplates
as a type of cognizable harm the memory of unpleasant
“sensations” that may arise when the consumer re-encounters the
plaintiff’s goods. Or, as Dutch lawyer Charles Gielen memorably

Sun, Barton Beebe and Madhavi Sunder, eds., 2015) (noting the ease with which dislike of a
practice can, problematically, translate into a finding of harm).
161. Moseley, 605 F.3d at 393.
162. Id. at 393-94.
163. See also the empirical work of Buccafusco, Heald & Bu, supra note 10 and
discussed in detail infra Part IV.B, which calls into question the extent to which
unauthorized sexual associations are harmful.
164. Claeryn/Klarein (1976) 7 I.I.C. 420, 423.
Vol. 106 TMR 677

put it: “[i]t is not funny to drink a good glass of Claeryn Dutch gin
while thinking of a cleaning agent at the same time!” 165
The critical analysis of the harm said to be caused by the use
of a mark on dissonant goods tends to stop at that point. As noted
in Part II.C, Claeryn/Klarein continues to be cited as a paradigm
of the tarnishment action under current EU law, and Gielen’s
statement has been referred to, uncritically, as an explanation of
the potential harm that was avoided in that case. 166 It is also
reflected in the European Union Intellectual Property Office’s
current practice guidelines and case law on the assessment of
tarnishment. It appears that the Office will look for uses that
“evoke undesirable or questionable mental associations which
conflict with the associations or image generated by legitimate use
of the reputed trade mark by its proprietor.” 167 In a 2010 decision
with notable echoes to Claeryn/Klarein, the Office rejected an
application for registration of a mark featuring the element “SPA”
for “[s]couring and polishing preparations and substances” in the
face of an earlier, similar “Spa” device mark for mineral water, on
the basis that the “pleasant notions” conveyed by the latter goods
did “not mix seamlessly” with the defendant’s goods. 168 More
remarkably, the applicant’s mark was also rejected for “pot pourri;
incense; incense sticks; room fragrances and articles for perfuming
rooms” because mineral water was “not pleasantly associated” with
such goods. 169
However, as with the U.S. jurisprudence, there are two major
gaps in the reasoning that has been employed to reach the
conclusion that these are situations that do, in fact, involve
reputational damage. The first is that the reasoning involves
similar assumptions that these “unpleasant associations” would

165. Charles Gielen, Harmonisation of Trade Mark Law in Europe: The First Trade
Mark Harmonisation Directive of the European Council, 1992 Eur. Intell. Prop. Rev. 262,
267.
166. See, e.g., Simon Fhima, Trade Mark Dilution, supra note 9, at 162.
167. Office for Harmonization in the Internal Market (Trade Marks and Designs),
Guidelines for Examination in the Office for Harmonization in the Internal Market (Trade
Marks and Designs), Part C: Opposition, Section 5, Trade Marks With a Reputation, Article
8(5) CTMR, 61 (Aug. 1, 2015), at [Link]
cument_library/contentPdfs/law_and_practice/trade_marks_practice_manual/WP/Part-C/05-
part_c_opposition_section_5_trade_marks_with_reputation_article_8_5_ctmr/part_c_opposit
ion_section_5_trade_marks_with_reputation_article_8_5_ctmr_en.pdf. See also id. at 63-65,
giving examples of decisions where applications for registration were rejected on the basis of
detriment to repute (such as “Kappa” for tobacco products, on the basis that it would
“prompt negative mental associations with the respondent’s earlier [‘Kappa’ device marks
for sports clothing] or associations conflicting with and detrimental to their image of a
healthy lifestyle”: id. at 63).
168. Case R 417/2008-1, SPACE NK Ltd. v. SA Spa Monopole, Compagnie fermière de
Spa, en abrégé SA Spa Monopole NV (OHIM First Board of Appeal, July 22, 2010), para.
101.
169. Id. at para. 103.
678 Vol. 106 TMR

actually occur, and that any quality signals conveyed by the senior
mark would be sufficiently disrupted so as to warrant legal
intervention. Looked at the other way, it assumes, for example,
that an ordinary Claeryn drinker would be unable to
compartmentalize the associations conveyed by the two marks,
keeping in mind the vastly different contexts in which he or she is
likely to encounter and experience the two sets of goods. The
second gap is that even if the “unpleasant associations” do arise
and transfer, it is assumed that they inevitably translate into
something more, in the sense of causing the consumer to shift his
or her perceptions of the senior mark in a manner that has the sort
of economic consequences that would justify the law intervening to
prevent use on the junior goods.
To what extent, then, are these gaps in reasoning filled by
evidence of what we know about the way in which consumers are
likely to respond in such situations? The orthodox approach to
tarnishment in the United States and European Union would
carry much more weight if there were strong support in the
cognitive science literature for the existence of “negative feedback
effects” occurring in the absence of source confusion between
branded goods. However, such support tends to be thin.
In an article from the early 1990s, relied on by a number of
commentators, Alexander Simonson suggested that transfers are
likely to occur because of “stimulus generalization” (that is, a
“mechanical-type” reaction among consumers, triggered when a
negative stimulus appears in a different context) and “schema
theory” (in which a “script” or “schema” generated by the use of a
brand will be evoked when the consumer sees the same brand
again, or a sufficiently similar variant, even in a different
commercial context). 170 Putting aside the problem with the idea
that undiluted brands trigger only the positive set of responses
desired by the mark owner and can therefore be so easily disrupted
by “negative” stimuli (as canvassed in Part III.A), what is missing
from this analysis is any sense that trademarks are perceived by
consumers in particular commercial and social contexts. The
context in which the consumer is likely to re-experience the famous
mark significantly complicates the idea that simply seeing or
thinking of the famous mark will “mechanically” trigger the
negative association. This is not only because the famous mark is
likely to be used in the context of goods or services different from
those of the junior user, but also because the consumer is likely to
re-encounter the famous mark in conjunction with an array of
other signs, information, and stimuli that are designed to reaffirm
consumers’ prior understandings of the famous mark and suppress

170. Alexander F. Simonson, How and When Do Trademarks Dilute: A Behavioral

Framework to Judge “Likelihood” of Dilution, 83 TMR 149, 160-61 (1993).
Vol. 106 TMR 679

any competing “meanings” the mark might generate. 171 To put it

another way, a consumer seeing the “Tiffany’s” striptease joint
might have an immediate response that brings the TIFFANY
jewelry business to mind, and form a new mental association
between the word “Tiffany’s” and sleaze. But this falls well short of
proving that when the consumer next encounters the famous
TIFFANY mark in a commercial setting that negative stimuli
would even be generated, let alone disrupt any existing
associations the consumer might have with the famous mark.
A further problem is that while Simonson posited “stimulus
generalization” and “schema theory” as explaining tarnishment in
the absence of source confusion, the only published research that
was relied on in support related to authorized brand extensions,
where the evidence of negative feedback effects was described as
being equivocal at best. 172 There is, in fact, a growing body of more
recent literature on authorized brand extensions and feedback
effects that suggests that consumers are much better than might
be expected at cabining negative reactions to problematic brand
extensions. 173 For example, it has been argued that although a
brand extension that is congruent with the original brand can
result in positive feedback effects, where an authorized brand
extension is thought by consumers to be incongruent with core
brand beliefs, a negative evaluation of the extension will not affect
evaluation of the core brand. 174 Similarly, it has been suggested
that consumers with high self-brand connections maintain
favorable evaluations of those brands, despite the subsequent
receipt of negative information about them. 175 This is consistent
with earlier work that suggests that consumers learn brand
associations that later block the learning of new predictive
attribute associations, 176 or, as Mark McKenna has succinctly put
it, “that well-known brands are quite resistant to change.” 177 If this

171. This draws on Tushnet’s work on context effects in Gone in Sixty Milliseconds,
supra note 1, at 529-32.
172. Simonson, supra note 170, at 166-67.
173. For an overview, see McKenna, Testing, supra note 146, at 104-05; Tushnet, Gone
in Sixty Milliseconds, supra note 1, at 543-44.
174. Henrik Sjödin & Fredrik Törn, When Communication Challenges Brand
Associations: A Framework for Understanding Consumer Responses to Brand Image
Incongruity, 5 J. Consumer Behav. 32 (2006). See also Joseph W. Chang, Will a Family
Brand Image Be Diluted by an Unfavorable Brand Extension? A Brand Trial-Based
Approach, 29 Advances in Consumer Res. 299, 303 (2002) (arguing that dilution is unlikely
at least in cases of “high involvement” core brands).
175. Shirley Y.Y. Cheng, Tiffany Barnett White & Lan Nguyen Chaplin, The Effects of
Self-Brand Connections on Responses to Brand Failure: A New Look at the Consumer-Brand
Relationship, 22 J. Consumer Psychol. 280 (2012).
176. Stephen J. Hoch, Product Experience Is Seductive, 29 J. Consumer Res. 448 (2002).
177. McKenna, Testing, supra note 146, at 105. See also Intel Corp. Inc. v. CPM United
Kingdom Ltd. [2007] EWCA Civ 431, para. 29 (where Lord Justice Jacob stated: “[i]f a trade
680 Vol. 106 TMR

holds true for authorized brand extensions, where the consumer

knows that the brand owner has responsibility for the (inferior)
quality of the junior goods, this must seriously call into question
the extent to which “stimulus generalization” or “schema theory”
provide any real insight into what is likely to happen where
consumers do not believe the junior use is authorized. 178 If
anything, it suggests that consumers’ attitudes toward TIFFANY
jewelry, COCA-COLA beverages, KAPPA sportswear and SPA
mineral water will remain unaffected after having been confronted
with these marks or variants used in relation to a striptease joint,
low grade engine oil, tobacco products, and scouring preparations,
respectively.
This final point receives support from Christo Boshoff’s recent
empirical work on consumers’ emotional responses to brand
tarnishment. 179 Boshoff conducted experiments in which forty
consumers, from 25 to 50 years old, were exposed to static, on-
screen print advertisements featuring twelve well-known brands
and twelve “tarnished” versions of them, with the order in which
participants were exposed to the two variations of each brand
being randomized. 180 Using electroencephalography (EEG) to
measure brain activity and electromyography (EMG) to measures
changes in the electrical activity of facial muscles, Boshoff found
that, at an aggregate level, the tarnished brands elicited only
neutral responses. 181 More interestingly, when the participants
were exposed to the tarnished brand first, their EEG responses to
the untarnished brand were more positive than in cases where they
were exposed to the untarnished version first. 182 These results cast
doubt on the existence of negative feedback effects and “provide
some support for the view that well-known trademarks/brands are
practically immune to dilution.” 183
Even if it is accepted that some negative feedback effects
might occur in the absence of source confusion, in that consumers

mark for particular goods or services is truly inherently and factually distinctive it will be
robust enough to withstand a mere passing bringing to mind when it or a similar mark is
used for dissimilar goods or services. The average consumer is a reasonably sensible
individual. He is used to lots of trade marks in different fields—some of which may resemble
trade marks for other fields”).
178. See Tushnet, Gone in Sixty Milliseconds, supra note 1, at 544.
179. Christo Boshoff, The Lady Doth Protest Too Much: A Neurophysiological Perspective
on Brand Tarnishment, 25 J. Prod. & Brand Mgmt. 196 (2016).
180. Id. at 200-01. The brands were VISA, PANASONIC, SONY ERICSSON, SHELL,
LOGITECH, CROPP VILLAGE (a Polish retailer), GAP, GUCCI, WWF, PIZZA HUT,
STARBUCKS, and KLEENEX: id. The “tarnishment” appeared to involve the use of the
brand in a humorous, but not unsavory, context: id. at 205.
181. Id. at 201-02.
182. Id. at 201-03.
183. Id. at 204.
Vol. 106 TMR 681

might feel less positively about a famous brand after seeing it used
on dissonant goods, 184 a further, arguably even more important,
question needs to be asked: is this is a “harm” that rises to a level
that warrants legal intervention? If all that can be said is that a
consumer now has to contend with a dissonant association being
triggered when he or she is confronted with the famous mark
(something which is likely to suppressed by other cues on re-
encountering the famous mark), it is not clear that we are very
much further along the path from this being a sort of “imagination
cost” imposed on consumers. This is something that has been
criticized as being a wholly inadequate explanation of the alleged
“harm” of dilution by blurring. 185
The issue would seem to be whether the generation of that
negative feedback effect is likely to rise to a level where it will
have some impact on consumers’ behavior. The Benelux Court of
Justice in Claeryn/Klarein held that for there to be damage, there
must be an impairment of the “capacity of the mark to stimulate
the desire to buy.” The generation of the negative association
might go part of the way to establishing this, but it does not
inexorably follow—and cannot be uncritically assumed—that a
consumer’s altered perception of the famous mark will make it any
less likely to purchase the goods in question. In other words, we
should demand some evidence that any disgust or frustration felt
by a consumer encountering a brand in a dissonant context would
translate into a course of economic behavior in relation to that
brand. It is not at all self-evident that a Claeryn drinker who has
to contend with feelings of mild discomfort generated by a mental
association with a detergent whenever he or she wishes to drink
his or her gin, which he or she presumably enjoys for its quality,
would shun the brand on that basis. The idea that a “Spa” mineral
water drinker would similary shun the brand because of the
creation of an indelible mental association between the word and
potpourri is even less plausible. This line of reasoning also receives
support from the brand extension literature. McKenna has drawn
attention to research where it has been shown that there is only
likely to be a correlation between a consumer’s attitudes and the
consumer’s behavior in situations of direct product experience
(that is, the purchase of goods, direct testing, sampling, or other
forms of evaluation), but not in cases of mere advertising

184. Cf. Claudiu V. Dimofte & Richard F. Yalch, The Mere Association Effect and Brand
Evaluations, 21 J. Consumer Psychol. 24 (2011) (arguing that negative associations
conveyed by particular words (such as animals, numbers or natural phenomena) can, when
a consumer is primed with that word, impact on a consumer’s perception of an unrelated
brand, but not considering situations akin to those where a trademark dilution action would
potentially be available).
185. See Tushnet, Gone in Sixty Milliseconds, supra note 1; Austin, supra note 1.
682 Vol. 106 TMR

exposure. 186 When read in light of studies that show that

consumers also often inaccurately predict their future emotional
states, by “overestimat[ing] the intensity and duration of their
emotional reactions to future events,” 187 this raises real questions
about the likelihood of a consumer’s changed attitude as a result of
an external negative stimulus impacting on his or her future
purchasing decisions. 188
Recognizing that what is sought to be prevented is, ultimately,
an economic harm, 189 a much more convincing basis for legal
intervention to prevent the likelihood of dilution by tarnishment
would therefore require:
evidence of a change in the economic behaviour of the average
consumer of the goods or services for which the earlier mark
was registered consequent on the use of the later mark, or a
serious likelihood that such a change will occur in the
future. 190
This is the current law in the European Union in relation to the
dilution by blurring action. There would be nothing stopping the
“detriment to repute” ground in the Trade Marks Directive being
interpreted in a similar way, 191 or a United States court
interpreting the TDRA to find that this is what is required to
demonstrate a likelihood of tarnishment. 192 That this evidence,
which would need to go beyond speculative claims as to

186. Chang, supra note 174, at 299 n.2.

187. Timothy D. Wilson & Daniel T. Gilbert, Affective Forecasting: Knowing What to
Want, 14 Current Directions in Psychol. Sci. 131, 131 (2005).
188. See McKenna, Testing, supra note 146, at 95-96. See also Boshoff, supra note 179,
at 204 (arguing his results “do not suggest a strong likelihood of severe economic harm due
to negative emotional responses to brand tarnishment among consumers”).
189. See Alexander Dworkowitz, Ending Dilution Doublespeak: Reviving the Concept of
Economic Harm in the Dilution Action, 20 Tex. Intell. Prop. L.J. 25, 60-61 (2011).
190. Case C-252/07, Intel Corp. Inc. v. CPM United Kingdom Ltd. [2008] E.C.R. I-8823,
para. 77 (E.C.J.), affirmed by the Court of Justice in Case C-383/12 P, Environmental Mfg.
LLP v. Office for Harmonization in the Internal Market (Trade Marks and Designs) (OHIM)
(Nov. 14, 2013), para. 37 (“The concept of ‘change in the economic behaviour of the average
consumer’ lays down an objective condition. That change cannot be deduced solely from
subjective elements such as consumers’ perceptions. The mere fact that consumers note the
presence of a new sign similar to an earlier sign is not sufficient of itself to establish the
existence of a detriment or a risk of detriment to the distinctive character of the earlier
mark . . . in as much as that similarity does not cause any confusion in their minds”).
191. It is accepted that the situation is complicated in the European Union by the
presence of the additional “unfair advantage” ground, which has been more generously
interpreted by the Court of Justice and which is likely to catch cases of “free-riding” in cases
where the (more difficult to prove) change in consumers’ economic behavior cannot be
established. See further infra Part V.
192. See also McCarthy, supra note 136, § 24:115 (“judges should demand persuasive
evidence that dilution is likely to occur. Even the probability of dilution should be proven by
evidence, not just by theoretical assumptions about what possibly could occur or might
happen”).
Vol. 106 TMR 683

consequences said to flow from the existence of asserted negative

feedback effects generated by the junior use, would likely be
extremely difficult for mark owners to provide seems entirely
justified, given the highly elusive nature of the harm in question.
Indeed, if this meant that the action were only available in cases
where consumers would be confused as to the origin of the two sets
of goods or services, this would not be a problem. As John Golden
has persuasively argued, redundancy in the law has value if this
works to contain the operation of a vague, uncertain standard (in
this case, dilution by tarnishment) that sits alongside a more
stable one (confusion-based trademark infringement), such that
the contained standard will produce different outcomes only in the
most exceptional cases. 193

B. Use for Parody, Critique, or Comment:

Is There Tarnishment, and
Are There Adequate Safeguards?
Moving away from use on dissonant goods, a famous mark
could also be used in what might be termed a more “targeted”
manner—that is, a junior user might wish to adopt a famous mark
to critique, comment on, parody, or in some way refer more directly
to the mark or its owner. Here, we might feel more comfortable in
saying that such conduct could potentially harm the mark’s
reputation, especially if that commentary makes negative claims
about the mark: in such instances we would seem to be much
closer to situations analogous to those dealt with under the laws of
defamation or injurious falsehood.
The fact that the junior user is seeking to engage in some form
of critique or commentary on the famous mark raises the issue of
the extent to which allowing the mark owner to maintain a
tarnishment action impinges unduly on the junior user’s freedom
of expression. 194 The usual response to this claim is therefore to
point to the variety of safeguards in place in local laws to ensure
that free speech concerns are adequately accommodated. The
United States provides a strong example in this regard. Current
U.S. federal anti-dilution law exempts from liability “[a]ny fair
use, including a nominative or descriptive fair use, or facilitation of
such fair use, of a famous mark by another person other than as a
designation of source for the person’s own goods or services” (which

193. John M. Golden, Redundancy: When Law Repeats Itself, 94 Tex. L Rev. 629, 708-09
(2016).
194. This is not to suggest that freedom of expression concerns arise only in relation to
this second category of “targeted” use: see, e.g., LaFrance, supra note 1. However, the impact
of the tarnishment action on freedom of expression is more squarely raised in this sort of
case, and worth exploring in the context of an assessment of the internal “safeguards” of the
tarnishment action around the world.
684 Vol. 106 TMR

specifically includes uses for parody, critique or commentary), as

well as all forms of news reporting and commentary, and non-
commercial uses. 195 A recent illustration of the operation of this
defense is the U.S. Court of Appeals for the Fourth Circuit’s
decision in Radiance Foundation, Inc. v. N.A.A.C.P. 196 In this case,
the National Association for the Advancement of Colored People
brought an action against an organization that used the phrase
“National Association for the Abortion of Colored People” in an
online article criticizing the NAACP’s stance on abortion. There
was no challenge to the District Court’s finding that there was a
prima facie case of dilution by tarnishment. 197 However, the
appellate court overturned the district court, finding that Radiance
was able to make out both a fair use defense (given that it was
commenting on what it considered to be the NAACP’s policy stance
on an issue), and a defense of non-commercial use (given that the
article was not an advertisement and did not refer to Radiance’s
services). 198
A broadly similar position had been reached under the FTDA,
which did not contain equivalent defenses. For example, in Smith
v. Wal-Mart Stores, Inc., 199 the defendant’s sale of T-shirts bearing
designs and slogans incorporating the word “Walocaust” was held
not to have tarnished the plaintiff’s “Wal-Mart” marks on the basis
that the defendant was engaged in non-commercial speech. The
judge cited the U.S. Court of Appeals for the Ninth Circuit’s
decision in Mattel, Inc. v. Walking Mountain Prods., 200 a case in
which an artist who created a series of photographs of nude Barbie
dolls being threatened by household appliances was able to defeat

195. 15 U.S.C. § 1125(c)(3). It is also that case that for liability to arise at all under §
1125(c)(1), the defendant must have commenced “use of a mark or trade name in
commerce.” It has been persuasively argued that this language incorporates a requirement
that the defendant be using an indication of source, rather than merely using a term in a
commercial setting: see Burstein, supra note 6, at 1224-27. It has also been suggested that
the effect of this requirement is that some cases under pre-FTDA law, such as those
involving “Enjoy Cocaine” and the Dallas Cowboy Cheerleaders, would not be actionable
under the TRDA: Barton Beebe, A Defense of the New Federal Trademark Dilution Revision
Act, 16 Fordham Intell. Prop. Media & Ent. L.J. 1143, 1172 (2006). However, it is possible to
point to tarnishment cases under the TDRA, such as Pfizer, Inc. v. Sachs, 652 F. Supp. 2d
512 (S.D.N.Y. 2009), Louis Vuitton Malletier, SA v. Hyundai Motor America, 2012 WL
1022247 (S.D.N.Y. 2012) and Radiance Foundation, Inc. v. N.A.A.C.P., 25 F. Supp. 3d 865
(E.D. Va. 2014), where the requirement that the defendant have “commence[d] use of a
mark or trade name in commerce” has been glossed over. Consequently, it appears more
likely that a defendant that has used a sign in a prima facie dilutive manner will only be
able to escape liability it can bring itself within one of the defenses in § 1125(c)(3).
196. 786 F.3d 316 (4th Cir. 2015).
197. See Radiance Foundation, Inc. v. N.A.A.C.P., 25 F. Supp. 3d 865, 894-96 (E.D. Va.
2014).
198. Radiance, 786 F.3d at 331-32.
199. 537 F. Supp. 2d 1302 (N.D. Ga. 2008).
200. 353 F.3d 792 (9th Cir. 2003).
Vol. 106 TMR 685

a tarnishment action, where it was said that “tarnishment caused

merely by an editorial or artistic parody which satirizes a product
or its image is not actionable because of the free speech protections
of the First Amendment.” 201
Such legislative and judicial recognitions of the importance of
free speech and the need to ensure that dilution laws are
adequately constrained are clearly important. However, to return
to the points made in Parts III and IV.A, it is crucial when
thinking about cases of criticism and commentary, especially in
cases involving commercial speech, not to downplay the central
question of whether there is a likelihood of the harm of
tarnishment occurring at all. If all that is being safeguarded by the
action is the owner’s “brand identity,” it is relatively easy to
assume that critique or parody might mean that the plaintiff’s
mark “will suffer negative associations.” 202 But given that the legal
test is whether the defendant’s conduct harms the reputation of
the mark, this would seem to require a much more nuanced
analysis of whether such harm is likely to occur. This is especially
the case if there is a substantial disparity between the owner’s
brand identity and its reputation, properly understood, and where
the defendant’s conduct can be seen to be drawing attention to an
aspect of the plaintiff’s reputation that it might wish to suppress.
For example, in Walking Mountain, a plausible case could be made
that the artist’s use of the Barbie doll was intended to shed light
on the brand’s reputation as the embodiment of “the conventional
beauty myth and the societal acceptance of women as objects.” 203 It
might well be the case that the junior user wishes to draw
attention to some negative conduct of the mark owner that is not
especially well known or has been largely forgotten in
contemporary discourse but can still be said to form part of its
reputation. 204 In cases involving targeted parodies of or critical
engagement with what might loosely be described as “pre-
tarnished” marks, there must be real doubt about whether the
threshold requirement of “harm” to reputation in fact exists, 205

201. Id. at 812, cited in Smith, 537 F. Supp. 2d at 1339. For further discussion of the
treatment of non-commercial use under the FTDA, see Lisa P. Ramsey, Increasing First
Amendment Scrutiny of Trademark Law, 61 S.M.U. L. Rev. 381, 402-03 (2008).
202. This being the language used to describe the harm in Hormel Foods Corp. v. Jim
Henson Prods., Inc., 73 F.3d 497, 507 (2d Cir. 1996).
203. Mattel, 353 F.3d at 797 (quoting the artist’s declaration in support of his motion for
summary judgment). For recent discussion, see Rebecca Tushnet, Make Me Walk, Make Me
Talk, Do Whatever You Please: Barbie and Exceptions, in Intellectual Property at the Edge:
The Contested Contours of IP, 405 (Rochelle Cooper Dreyfuss & Jane C. Ginsburg, eds.,
2014). See also Dan Hunter and F. Gregory Lastowka, BarbieTM, 18 Tul. J. Tech. & Intell.
Prop. 133 (2015).
204. See Lux, supra note 1, at 1066-67.
205. In this regard, it is worth noting that there is already a body of scholarship that
queries whether a blurring action ought to be available where the owner has engaged in
686 Vol. 106 TMR

which would seem to be a better way of resolving the issue than on

the basis of the vexed question of whether the defendant’s speech
is or is not commercial. 206
Further, even where the reputation of the famous mark is
uncontentious, it cannot be taken for granted that parodies or
similar adoptions of or references to well-known marks in
“unwholesome” contexts give rise to the harm of tarnishment at
all. In recent empirical work undertaken by Christopher
Buccafusco, Paul Heald and Wen Bu, two experiments were
conducted in which subjects were shown thirty pairs of posters of
famous movies and asked which of each pair a theatre should show
to maximize its profits, with one movie in each of the last ten pairs
designated by the authors to be a “target” movie. Some subjects
were exposed to posters of unauthorized parodic pornographic
versions of some of the target movies in the first twenty pairs
before they were exposed to the posters for the target movies in the
final ten pairs. The authors’ hypotheses were that subjects exposed
to the pornographic posters would have more negative attitudes to
the target movies afterwards than those subjects in the control
group and that these tarnishment effects would be greater for
women and for socially conservative subjects but would be weaker
for those who had already seen the target movies (given the
already positive associations generated). 207 Their data did not
support these predictions: those exposed to the pornographic
poster were, in fact, more likely to choose the target movie when
exposed to it, with no difference between the responses of men and
women, and with prior exposure to the famous movie not affecting
the outcome. 208 In the second experiment, subjects were, in
addition, shown eight pairs of posters of famous movies and asked
which of each they would like to see a sequel of, with some subjects
being exposed to posters of pornographic versions earlier in the
experiment. The data showed no significant difference between
those exposed to the pornographic versions and those who were
not, with tarnishing effects being evident only among the most
socially conservative subjects. 209 This empirical research is
consistent with the arguments made in Part IV.A: it calls into
question the extent to which associating a brand with sexual
content has any impact on consumers’ perceptions of, or desire to

substantial brand diversification: see generally Sara Stadler Nelson, The Wages of Ubiquity
in Trademark Law, 88 Iowa L. Rev 731 (2003). It might be possible to see an “already
tarnished” argument developing along similar lines.
206. For recent analysis, see Jennifer E. Rothman, Commercial Speech, Commercial Use,
and the Intellectual Property Quagmire, 101 Va. L. Rev. 1929 (2015).
207. Buccafusco, Heald & Bu, supra note 10, at 34.
208. Id. at 40-41, 46.
209. Id. at 46-49.
Vol. 106 TMR 687

acquire goods or services under, the brand and, “put[s] the ball
back in the court of tarnishment theorists to produce empirical
support for their claims.” 210
Assuming, however, that most courts are likely to continue to
protect the mark owner’s “brand identity,” and are not likely to
apply a rigorous test in determining whether the harm of
tarnishment or detriment to repute exists, this raises the question
of whether the safeguards contained in legal systems designed to
accommodate free speech interests in commercial contexts are
adequate. Despite the success of the defendants in the United
States actions outlined earlier in this Part IV.B, it is important not
to be too sanguine about the idea that legislative carve-outs or
sensitive decision-making by courts applying constitutional
scrutiny can do the work of protecting free speech in all cases of
“targeted” use of famous marks, particularly given local differences
in trademark laws and constitutions.
Looking first at the issue of legislative carve-outs, the key
problem is the potential under-inclusivity of defenses. This is a
problem even with something as broad as a “fair use” defense.
Under U.S. law, the defense contained in the TDRA does not apply
where the defendant’s use is as a designation of source. This not
only forces decision-makers into the difficult position of having to
draw fine distinctions between source-identifying and non-source-
identifying uses but also, more importantly, necessarily means
that some “targeted” uses will never be able to qualify as fair
uses. 211 It is not immediately clear why source-identifying uses
ought to be proscribed, when closely similar conduct that might be
said to cause the same harm is not. While it could be argued that
the U.S. Court of Appeals for the Fourth Circuit found a way
around this problem in Louis Vuitton Malletier SA v. Haute
Diggity Dog, LLC 212 by incorporating the fact that the defendant’s
CHEWY VUITON brand was operating as a parody of the
plaintiff’s famous mark into the assessment of whether there was
dilution by blurring in the first place, it is difficult to see how this
could be factored into an assessment of tarnishment. This is
particularly so in cases involving more critical parodies or
comment than the spoof product in Haute Diggity Dog, which are
more likely to impact on the famous mark’s reputation.
The problem is more acute in other jurisdictions. No such “fair
use” defense applies under European law. Instead, under the anti-
dilution provisions in both the original and the 2015 Trade Marks

210. Id. at 53 (emphasis added).

211. A similar problem exists under the Trade Marks Act 1998 (Singapore). Although
section 55A(3) specifies a number of defenses that incorporate free speech concerns, these
too are limited in their scope in that they do not appear to cover commercial uses for parody
or criticism.
212. 507 F.3d 252 (4th Cir. 2007).
688 Vol. 106 TMR

Directives, the only real carve-out of any potential relevance is for

use in the course of trade that is not “without due cause.” The
European Court of Justice has recently considered the application
of this proviso in two specific contexts: use of a mark in keyword
advertising, 213 and good-faith use antedating registration of the
famous mark. 214 From those decisions, it seems that a defendant is
not required to show “necessity” or a compelling need to use the
mark for it to be with due cause (as was the case under former
Benelux law, following Claeryn/Klarein 215), and it has been
suggested that national courts in the European Union are now
more likely to look to whether the junior user has a legitimate
interest in using the mark. 216 It is not yet clear how far this might
extend, or the extent to which free speech concerns, such as the
interest in using the famous mark to engage in parody or critique,
can be accommodated within this proviso. And even this limited
possibility might be unavailable in other jurisdictions with anti-
dilution laws: for example, in New Zealand, whose anti-dilution
provisions are largely modelled on post-Directive United Kingdom
laws, the “without due cause” limitation does not even exist. 217
Looking at the issue of the ability of decision-makers in cases
involving critique or parody to prioritize free speech concerns
based on constitutional considerations, the case that seems to
provide the most support for this idea is Laugh It Off Promotions
CC v. South African Breweries Int’l (Finance) BV. 218 In that case
the Constitutional Court of South Africa found in favor of a seller
of T-shirts bearing a variation on the mark owner’s CARLING
BLACK LABEL device mark with “Black Label” replaced by “Black
Labour,” “Carling Beer” replaced by “White Guilt,” and “America’s
lusty lively beer” replaced by “Africa’s lusty lively exploitation
since 1652.” The Constitutional Court overturned the lower court’s
finding that the defendant’s use was likely to be detrimental to the
repute of the plaintiff’s mark under the Trade Marks Act 1993, and
that there was no separate justification based on the constitutional
guarantee of freedom of expression. Finding this approach to be in
error, the Constitutional Court considered the “detriment to
repute” provision in the Act needed to be read in light of, and as
being compatible with, the constitutional guarantee, and thus set

213. Case C-323/09, Interflora Inc. v. Marks & Spencer plc. [2011] E.C.R. I-8625.
214. Case C-65/12, Leidseplein Beheer BV v. Red Bull GmbH [2014] E.T.M.R. 24.
215. See supra note 97.
216. Bently & Sherman, supra note 106, at 1011.
217. Although, as explained supra note 20, infringement requires the defendant’s use to
be “as a trade mark,” which does provide a degree of protection for non-commercial speech.
In relation to the similar provision under Australian law, see Patricia Loughlan, Protecting
Culturally Significant Uses of Trade Marks (Without a First Amendment), 2000 Eur. Intell.
Prop. Rev. 328.
218. 2006 (1) SA 144 (CC).
Vol. 106 TMR 689

up a test under which the trademark owner had to show a

likelihood of substantial economic detriment to succeed. 219 The
mere fact that the defendant’s racial slur might have been
“unsavory” fell well short of evidence of likely economic harm. 220
Laugh It Off has rightly been lauded as a significant case in
reconciling a constitutional guarantee of freedom of expression
with property interests. 221 However, the extent to which it might
provide a model for other jurisdictions is not clear, especially given
that the defendant’s speech in that case was commercial. In the
United States, Ramsey has argued convincingly that although
courts could potentially apply constitutional scrutiny to find anti-
dilution law to be an unconstitutional restriction of commercial
speech that does not cause consumer confusion, there are
significant disadvantages to this approach that are likely to
prevent it from occurring in practice. 222 Robert Burrell and Dev
Gangjee have shown that in the European Union, despite the
existence of Article 10 of the European Convention on Human
Rights, courts have shown themselves reluctant to prioritize free
speech principles in interpreting the scope of trademark rights, at
least outside cases of political or polemical expression. 223 The
situation would be different again in a country such as Australia,
where the implied constitutional guarantee of freedom of political
communication does not operate as a free-standing device that
courts can use to guide them in the interpretation of statutes;
rather, it operates only as a check on legislative power. 224 As such,
if Australia were to adopt U.S.-style anti-dilution legislation, this
constitutional guarantee could provide no protection for a junior
user of a famous mark in the type of situation that confronted the
court in Laugh It Off.
This discussion demonstrates that laws aimed at preventing
dilution by tarnishment, even when read in light of safeguards
designed to protect free speech, are still likely to have an adverse
effect on certain types of critique and comment seeking to alter
consumers’ perceptions of famous marks. As a practical matter,
more could be done to improve the scope of these safeguards.
However, to focus only on strengthening speech-based limits to the
scope of the dilution action has the potential disadvantage of

219. Id. at paras. 48 and 56.

220. Id. at para. 55.
221. Laurence R. Helfer & Graeme W. Austin, Human Rights and Intellectual Property:
Mapping the Global Interface 300 (2011).
222. Ramsey, supra note 201, at 450-53.
223. Robert Burrell & Dev Gangjee, Trade Marks and Freedom of Expression: A Call for
Caution, 41 IIC: Int’l Rev. Intell. Prop. & Competition L. 544, 559-64 (2010).
224. For recent confirmation, see Unions N.S.W. v. New South Wales (2013) 252 C.L.R.
530, 554 (High Court of Australia).
690 Vol. 106 TMR

drawing attention away from the more fundamental question I

have sought to explore in this article: whether the so-called harm
of tarnishment ever rises to a level that warrants legal
intervention. Indeed, one of the notable outcomes of Laugh It Off,
which could have been reached in the absence of the constitutional
guarantee of freedom of expression, was the setting up of a test for
tarnishment that requires plaintiffs to demonstrate a likelihood of
substantial economic detriment. This chimes with the argument I
made in Part IV.A that plaintiffs should be required to adduce
cogent evidence of a change in the economic behavior of the
average consumer. That plaintiffs may find this nearly impossible
in turn raises the question of whether it is appropriate to have
laws against dilution by tarnishment at all. Notwithstanding the
nature of any test for dilution, or the scope of any defenses or
limitations, the mere existence of these laws on the books may well
have a substantial chilling effect on commercial and non-
commercial expression alike. 225 This is particularly the case in
those jurisdictions where cease and desist letters can be sent
without the possibility of actions for “unjust threats” being brought
in response. It is hard to see how the maintenance of a dilution by
tarnishment action is justifiable when the harm of tarnishment, in
situations not covered by a traditional trademark infringement
action, is so elusive. As Burrell and Gangjee have put it, “perhaps
the answer is not to carve out freedom of expression exceptions to
dilution actions, but rather to rethink dilution.” 226

V. CONCLUDING THOUGHTS:
TOWARD MISAPPROPRIATION?
Dilution by tarnishment, to the extent it encompasses conduct
that does not result in consumer confusion, is a far more
problematic type of action than might first appear. A plausible
case can be made that the first “tarnishment” statutes in the
United States and Europe did not intend to provide such extensive
protection and, as the case law developed, no attempt was made to
explain exactly what was being protected, or how damage could be
inflicted by use of a mark in the absence of confusion. A study of
that case law shows that far from being designed to safeguard
“reputation,” properly understood, the action has come to protect a
limited form of constructed “brand identity,” entrenching
simplistic notions of a brand as repository of owner-generated
positive attributes, transmitted to and uncritically absorbed by

225. For recent consideration, see Regina Schaffer-Goldman, Cease-and-Desist:

Tarnishment’s Blunt Sword in Its Battle Against the Unseemly, the Unwholesome, and the
Unsavory, 20 Fordham Intell. Prop. Media & Ent. L.J. 1241 (2010); Irina D. Manta, Bearing
Down on Trademark Bullies, 22 Fordham Intell. Prop. Media & Ent. L.J. 853 (2012).
226. Burrell & Gangjee, supra note 223, at 547.
Vol. 106 TMR 691

consumers. Even more of a concern is that the case law provides

almost no basis to believe that reputational damage is likely to
occur in cases that do not involve source confusion. Judges have
assumed the existence of negative feedback effects and, except in
rare cases, have failed to demand evidence of the likelihood of the
junior use leading to changes in consumers’ economic behavior.
When read alongside the substantial body of scholarship
criticizing the notion that “blurring” is a cognizable harm, 227 this
raises the question of whether both limbs of the dilution action,
although framed in terms of preventing harmful effects, are really
about preventing “unfair” conduct. Is it, in fact, the case that the
entire dilution action is ultimately more about enforcing moral
standards than regulating economic behavior? 228 There is much to
be said for Barton Beebe’s recent argument that dilution in the
United States is little more than a fig leaf for a “misappropriation”
action. 229 Writing in response to Beebe, Graeme Dinwoodie has
argued that European trademark law is more overt about its
purpose of protecting against misappropriation, in particular by
providing a cause of action, separate from dilution by blurring or
tarnishment, that enables the trademark owner to seek a remedy
against a defendant who is attempting to take “unfair advantage”
of the distinctive character or repute of that mark. 230 Rather than
seeing this as a normatively hollow concept, 231 Dinwoodie suggests
that an action for misappropriation could be a useful supplement
to a trademark system geared around regulating economic
behavior, but cautions that:
[t]he challenges that arise from injecting fairness into the
calculus are: (1) the supposed indeterminacy of fairness-based
standards in an economic context; (2) the overbreadth of the
concept of misappropriation if truly limitless; and (3) if not
limitless, the problems of incommensurability that come from
weighing fairness with the economic concerns that heavily
structure the trademark regime alongside which sits unfair
competition. 232

227. See supra note 1.

228. See Bartholemew, supra note 156, at 138 (but limiting his analysis to tarnishment
by use on sex-related goods).
229. Beebe, Suppressed Misappropriation Origins, supra note 1.
230. Graeme B. Dinwoodie, Dilution as Unfair Competition: European Echoes, in
Intellectual Property at the Edge: The Contested Contours of IP, 81 (Rochelle Cooper
Dreyfuss & Jane C. Ginsburg, eds., 2014).
231. Cf. Moorgate Tobacco Co. Ltd. v. Philip Morris Ltd. (1984) 156 C.L.R. 414, 445-46
(High Court of Australia) (where Justice Deane dismissed the idea that a tort of unfair
competition should be introduced into Australian law, calling it a “cause of action whose
main characteristic is the scope it allows, under high-sounding generalizations, for judicial
indulgence of idiosyncratic notions of what is fair in the market place”).
232. Dinwoodie, supra note 230, at 100-01.
692 Vol. 106 TMR

These challenges are for the future, and it is worth noting that the
European Court of Justice, in considering the “unfair advantage”
provision of the original Trade Marks Directive, has shown little
interest in explaining when a defendant’s appropriation of a
famous mark without due cause is “unfair,” such that might be
said to constitute a misappropriation. 233 For now, it is enough to
note that there are real dangers in maintaining a normatively
hollow cause of action based on the harms “blurring” or
“tarnishment” of a famous mark, given that it is not at all clear
that, in the absence of confusion, a third-party use of such a mark
is ever likely to cause such harms at a level that warrants legal
intervention.

233. See Case C-487/07, L’Oréal SA v. Bellure NV [2009] E.C.R. I-5185, para. 49 (E.C.J.)
(“where a third party attempts, through the use of a sign similar to a mark with a
reputation, to ride on the coat-tails of that mark in order to benefit from its power of
attraction, its reputation and its prestige, and to exploit, without paying any financial
compensation and without being required to make efforts of his own in that regard, the
marketing effort expended by the proprietor of that mark in order to create and maintain
the image of that mark, the advantage resulting from such use must be considered to be an
advantage that has been unfairly taken of the distinctive character or the repute of that
mark”), an approach confirmed in Case C-323/09, Interflora Inc. v. Marks & Spencer plc.
[2011] E.C.R. I-8625, para. 89. For extended criticism, see Gangjee & Burrell, supra note 1.
Note, however, the more recent attention that has been given to the “without due cause”
element of the action as justifying some types of conduct: supra notes 213-216 and
accompanying text.