FOR PUBLICATION
UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT
LEVI STRAUSS & COMPANY,
Plaintiff-Appellant, No. 09-16322
v.
D.C. No.
3:07-cv-03752-JSW
ABERCROMBIE & FITCH TRADING
COMPANY, OPINION
Defendant-Appellee.
Appeal from the United States District Court
for the Northern District of California
Jeffrey S. White, District Judge, Presiding
Argued and Submitted
June 16, 2010—San Francisco, California
Filed February 8, 2011
Before: Kenneth F. Ripple, Senior Circuit Judge,*
Pamela Ann Rymer and Raymond C. Fisher, Circuit Judges.
Opinion by Judge Ripple
*The Honorable Kenneth F. Ripple, Senior United States Circuit Judge
for the Seventh Circuit, sitting by designation.
2369
2372 LEVI STRAUSS v. ABERCROMBIE & FITCH
COUNSEL
Gregory S. Gilchrist, Townsend & Townsend & Crew, San
Francisco, California, for the plaintiff-appellant.
J. Michael Keyes, K & L Gates, Spokane, Washington, for the
defendant-appellee.
David H. Bernstein, Debevoise & Plimpton, New York, New
York, for amicus curiae International Trademark Association.
OPINION
RIPPLE, Senior Circuit Judge:
Levi Strauss & Co. (“Levi Strauss”) seeks review of a dis-
trict court judgment that Abercrombie & Fitch Trading Co.
LEVI STRAUSS v. ABERCROMBIE & FITCH 2373
(“Abercrombie”) did not dilute Levi Strauss’s trademarked
“Arcuate” design in violation of the Trademark Dilution Revi-
sion Act of 2006 (“TDRA”), 15 U.S.C. § 1125(c). Levi
Strauss maintains that the district court applied an incorrect
legal standard in evaluating its dilution claim, namely that the
junior mark be “identical or nearly identical” to the senior
one. We agree with Levi Strauss that the “identical or nearly
identical” standard did not survive Congress’s enactment of
the TDRA and that the district court’s use of the incorrect
standard was not harmless error. Accordingly, we reverse the
judgment of the district court and remand the case for further
proceedings consistent with this opinion.
BACKGROUND
A. The Stitched Designs
Levi Strauss created, and began selling, blue jeans in the
1870s. Since 1873, the company has stitched the back pocket
of its jeans with two connecting arches that meet in the center
of the pocket; Levi Strauss holds a federally registered trade-
mark on this “Arcuate” design. Sales of garments bearing the
Arcuate mark have accounted for more than ninety-five per-
cent of Levi Strauss’s revenue over the past thirty years, total-
ing roughly fifty billion dollars. Levi Strauss actively
monitors use of competing stitching designs and enforces its
trademark rights against perceived infringers.
In 2006, Abercrombie began using a stitching design on the
back pockets of its jeans that, according to Levi Strauss, “in-
corporates the distinctive arcing elements of the Arcuate
trademark.” Appellant’s Br. 4. Abercrombie’s “Ruehl” design
consists of two less-pronounced arches that are connected by
a “dipsy doodle,” which resembles the mathematical sign for
infinity. The design on the Abercrombie jeans sits lower on
the pocket than Levi Strauss’s Arcuate design.**
**Images of the “Arcuate” and “Ruehl” designs appear at the end of
this opinion.
2374 LEVI STRAUSS v. ABERCROMBIE & FITCH
B. District Court Proceedings
In 2007, Levi Strauss brought an action against Abercrom-
bie for trademark infringement, unfair competition and trade-
mark dilution under both federal and California state law.
Before trial, Levi Strauss dropped its state-law trademark
dilution claim, and it withdrew its claim for monetary relief
on its federal dilution claim. Consequently, Levi Strauss’s
federal trademark dilution claim, for which it sought only
injunctive relief, was tried before the court with advisory rul-
ings from the jury.
During trial, Levi Strauss presented the testimony of Dr.
Sanjay Sood. Dr. Sood conducted a survey to evaluate
whether women associated Abercrombie’s Ruehl design with
Levi Strauss. Dr. Sood’s “Confusion Survey” revealed that
“[a]pproximately 30% of all respondents identified the Ruehl
jeans as made, sponsored or endorsed by the same company
that made the LEVI’S® jeans, as compared to lower percent-
ages for the ‘control’ jeans.” Levi Strauss & Co. v. Abercrom-
bie & Fitch Trading Co., No. C 07-03752 JSW, 2009 WL
1082175, at *4 (N.D. Cal. Apr. 22, 2009). Dr. Sood testified
that “the Ruehl design would erode the distinctiveness of the
Arcuate mark, in light of what he opined was a significant
level of confusion between the Ruehl jeans and the [Levi
Strauss] jeans.” Id. Dr. Sood also testified concerning a “Rec-
ognition Survey” that he had conducted. According to Dr.
Sood, the survey revealed that “approximately 32% to 35% of
the respondents actually associated the Arcuate mark with
[Levi Strauss].” Id. at *5.
For its part, Abercrombie presented the testimony of Dr.
Gerald Ford, who had been engaged in commercial marketing
research for more than three decades and had offered expert
testimony in over 50 trademark cases. Dr. Ford testified that
Dr. Sood’s confusion survey was not conducted according to
generally accepted and standard practices; as a result, the sur-
vey was “ ‘so flawed that it render[ed] the survey results’ and
LEVI STRAUSS v. ABERCROMBIE & FITCH 2375
Dr. Sood’s conclusions ‘meaningless.’ ” Id. (internal citations
omitted). Nevertheless, Dr. Ford acknowledged that, “[i]f a
likelihood of confusion survey is conducted in a valid and
reliable manner, a finding that ten percent or more of the pop-
ulation is confused is problematic to a trademark owner.” Id.
Dr. Ford opined that Dr. Sood’s recognition survey was prob-
lematic; specifically, it “was flawed[ ] because it did not con-
trol for spurious recognition, used ambiguous questions,
lacked an appropriate control cell, and suffered from order
bias.” Id.
As part of its deliberations, the district court requested that
the jury provide advisory rulings on factual issues related to
Levi Strauss’s federal dilution claims. With respect to the ele-
ments of the dilution claim, the court asked the jury to answer
the following question: “Is Abercrombie’s Ruehl design iden-
tical or nearly identical to the Arcuate trademark?” ER 67.
Although the jury later determined that the Arcuate trademark
was famous and distinctive, it found that the marks were not
identical or nearly identical and also determined that the
Arcuate trademark was not likely to be diluted by the Ruehl
design.
C. Findings of Fact and Conclusions of Law
On April 22, 2009, the district court entered judgment in
favor of Abercrombie on Levi Strauss’s federal dilution
claim. In its findings of fact, the district court noted that,
“[w]ith the exception of visual depictions of the two designs,
Dr. Sood’s testimony regarding the results of the Confusion
Survey was [Levi Strauss]’s only evidence that the Ruehl
design was identical or nearly identical to the Arcuate mark.”
Levi Strauss & Co., 2009 WL 1082175, at *5 (emphasis
added). Although the district court acknowledged that Dr.
Sood’s survey suggested confusion among thirty percent of
the participants, the court determined that Dr. Ford’s critique
of Dr. Sood’s methodology was persuasive and that his testi-
mony was more reliable than that offered by Dr. Sood.
2376 LEVI STRAUSS v. ABERCROMBIE & FITCH
Turning to its conclusions of law, the court first reviewed
the elements of a claim under the TDRA, 15 U.S.C.
§ 1125(c). The court stated:
To prevail on its dilution claim, [Levi Strauss] has
the burden of proving by a preponderance of evi-
dence the following elements: (a) that [Levi Strauss]
is the owner of a trademark that is famous; (b) that
the famous mark is distinctive, either inherently or
through acquired distinctiveness; (c) that [Abercrom-
bie] is making or has made use in commerce of an
identical or nearly identical trademark, in this case
the Ruehl design; (d) that [Abercrombie]’s use of its
Ruehl design began after [Levi Strauss]’s Arcuate
mark became famous; and (e) that [Abercrombie]’s
use of its Ruehl design is likely to cause dilution by
blurring of [Levi Strauss]’s Arcuate mark. 15 U.S.C.
§ 1125(c); Jada Toys, Inc. v. Mattel, Inc., 518 F.3d
628, 634 (9th Cir. 200[8]).
Levi Strauss & Co., 2009 WL 1082175, at *7 (emphasis
added). The district court noted that the advisory jury had not
found that the Ruehl design and the Arcuate mark were “iden-
tical or nearly identical,” a standard that required that “the
two marks . . . be similar enough that a significant segment
of the target group of customers sees the two marks as essen-
tially the same.” Id. (emphasis added).
After setting forth this standard for similarity, a standard
which “ ‘is more stringent than in the infringement context,’ ”
id. at *8 (quoting adidas-America, Inc. v. Payless Shoe
Source, Inc., 546 F. Supp. 2d. 1029, 1060 (D. Ore. 2008)), the
court observed that, although the evidence showed that Levi
Strauss had “expended significant amounts of money adver-
tising the Arcuate mark,” this was not “a case where the two
marks at issue involve only minor differences,” id. The dis-
trict court then reviewed aspects of the Arcuate mark and the
Ruehl design and observed that “[t]his evidence demonstrates
LEVI STRAUSS v. ABERCROMBIE & FITCH 2377
that a significant segment of the target group of customers
would not view the marks as essentially the same.” Id.
(emphasis added). It concluded that, consistent with the advi-
sory jury’s finding, Levi Strauss “has not established that
[Abercrombie] is making commercial use of a mark that is
identical or nearly identical to the Arcuate mark.” Id. at *9
(emphasis added). The court then concluded:
As noted, the test for similarity of the marks is
more stringent in the dilution context than for likeli-
hood of confusion purposes, and the two marks must
be “essentially the same mark.” For the reasons set
forth above, the Court finds that the Ruehl design
and the Arcuate mark are not visually similar. Fur-
thermore, as set forth above, the Court does not find
the results of Dr. Sood’s survey to be entitled to any
great weight. Thus, [Levi Strauss] has not put forth
persuasive evidence of actual association between
the Arcuate mark and the Ruehl design. Therefore,
these factors weigh in [Abercrombie]’s favor and
against a finding of likelihood of dilution.
....
Having considered all of the evidence, even if the
degree of inherent or acquired distinctiveness of the
Arcuate mark and the degree of recognition of the
mark weigh in [Levi Strauss]’s favor, the Court finds
that those factors do not outweigh the factors that
weigh in [Abercrombie]’s favor. Accordingly, the
Court concludes, consistent with the advisory jury’s
finding, that [Levi Strauss] has not met its burden to
establish by a preponderance of the evidence that the
Ruehl design is likely to cause dilution by blurring
of the Arcuate mark.
Id. (quoting Thane Int’l, Inc. v. Trek Bicycle Corp., 305 F.3d
894, 907 (9th Cir. 2002); footnote omitted; emphasis added).
2378 LEVI STRAUSS v. ABERCROMBIE & FITCH
The district court therefore entered judgment in favor of
Abercrombie.
DISCUSSION
Levi Strauss submits that the district court erred in requir-
ing it to establish that its mark was identical or nearly identi-
cal to the Ruehl design. Levi Strauss looks to the plain
language of 15 U.S.C. § 1125(c) and notes that the terms
“identical or nearly identical” appear nowhere in the language
of the statute. The statute, it maintains, does not require a
prima facie showing of substantial similarity before a district
court balances the dilution factors listed in § 1125(c)(2)(B).
Instead, “degree of similarity” is one of several factors that a
district court must balance in order to determine whether dilu-
tion has occurred and whether, therefore, a plaintiff is entitled
to injunctive relief.1
Abercrombie, by contrast, bases its argument not on the
language of the statute but, instead, on the case law of this
court. It argues that, even after the passage of the TDRA, this
court thrice has determined that a junior mark must be “identi-
cal or nearly identical” to that of the senior user in order for
a trademark-dilution plaintiff to be entitled to relief. See Per-
fumebay.com Inc. v. eBay, Inc., 506 F.3d 1165 (9th Cir.
2007); Jada Toys, Inc. v. Mattel, Inc., 518 F.3d 628 (9th Cir.
2008); Visa Int’l Serv. Ass’n v. JSL Corp., No. 08-15206,
2010 WL 2559003 (9th Cir. June 28, 2010).
To properly evaluate the parties’ contentions, we must look
at the origins of the “identical or nearly identical” standard,
how it has been employed and whether it remains viable after
the enactment of the TDRA.
1
The International Trademark Association, amicus in this appeal, agrees
with the interpretation of the TDRA proffered by Levi Strauss.
LEVI STRAUSS v. ABERCROMBIE & FITCH 2379
A. Origin of the “Identical or Nearly Identical”
Standard
The “identical or nearly identical” requirement traces its
roots in this circuit to Playboy Enterprises, Inc. v. Welles, 279
F.3d 796 (9th Cir. 2002). In that case, a former “Playmate of
the Year,” Terri Welles, had used this descriptive on a website
devoted to the sale of her pictures and to the promotion of her
services as a spokeswoman. Playboy sued Welles on various
theories, including a federal trademark dilution claim under
the prior statute.2 We determined that most of Welles’s use of
2
Specifically, the predecessor of the TDRA, the Federal Trademark
Dilution Act (“FTDA”), provided:
(c) Remedies for dilution of famous marks
(1) The owner of a famous mark shall be entitled, subject to the
principles of equity and upon such terms as the court deems rea-
sonable, to an injunction against another person’s commercial use
in commerce of a mark or trade name, if such use begins after the
mark has become famous and causes dilution of the distinctive
quality of the mark, and to obtain such other relief as is provided
in this subsection. In determining whether a mark is distinctive
and famous, a court may consider factors such as, but not limited
to—
(A) the degree of inherent or acquired distinctiveness of the
mark;
(B) the duration and extent of use of the mark in connection
with the goods or services with which the mark is used;
(C) the duration and extent of advertising and publicity of
the mark;
(D) the geographical extent of the trading area in which the
mark is used;
(E) the channels of trade for the goods or services with
which the mark is used;
(F) the degree of recognition of the mark in the trading areas
and channels of trade used by the marks’ owner and the per-
son against whom the injunction is sought;
(G) the nature and extent of use of the same or similar marks
by third parties; and
2380 LEVI STRAUSS v. ABERCROMBIE & FITCH
“Playmate of the Year” were nominative uses and, therefore,
excepted from coverage. However, Welles’s use of the abbre-
viation “PMOY” on the wallpaper of her site was not nomina-
tive, and, therefore was “not excepted from the anti-dilution
provisions.” Playboy Enters., 279 F.3d at 806. As a result, we
reversed the judgment in favor of Welles on the dilution issue
and gave the following instructions to the district court:
We note that if the district court determines that
“PMOY” is not entitled to trademark protection,
PEI’s claim for dilution must fail. The trademarked
term, “Playmate of the Year” is not identical or
nearly identical to the term “PMOY.” Therefore, use
of the term “PMOY” cannot, as a matter of law,
dilute the trademark “Playmate of the Year.”
Id. (emphasis added; footnote omitted).
Our authority for the “identical or nearly identical” lan-
guage employed in Playboy Enterprises was Luigino’s, Inc. v.
Stouffer Corp., 170 F.3d 827, 832 (8th Cir. 1999), and J.
Thomas McCarthy, McCarthy on Trademarks and Unfair
Competition § 24:90.2 (4th ed. 2001). See Playboy Enters.,
279 F.3d at 806 n.41. In Luigino’s, the Eighth Circuit held
that, “[t]o support an action for dilution by blurring, ‘the
marks must at least be similar enough that a significant seg-
ment of the two target groups of customers sees the two
marks as essentially the same.’ ” 170 F.3d at 832 (quoting J.
Thomas McCarthy, McCarthy on Trademarks and Unfair
Competition § 24:90.1, at 24-145 (4th ed. 1998)). In support
of this proposition, the Eighth Circuit not only relied on
McCarthy on Trademarks and Unfair Competition, but also
(H) whether the mark was registered under the Act of March
3, 1881, or the Act of February 20, 1905, or on the principal
register.
15 U.S.C. § 1125(c)(1) (2005).
LEVI STRAUSS v. ABERCROMBIE & FITCH 2381
Mead Data Central, Inc. v. Toyota Motor Sales, U.S.A., Inc.,
875 F.2d 1026, 1029 (2d Cir. 1989).
Tracing the ancestry of the standard a step further, Mead
Data Central involved a trademark dilution claim brought
under a New York state statute. According to the Second Cir-
cuit, the legislative history of that act stated “the purpose of
the statute as preventing ‘the whittling away of an established
trade-mark’s selling power and value through its unauthorized
use by others upon dissimilar products.’ ” Mead Data Cent.,
875 F.2d at 1028 (quoting 1954 N.Y. Legis. Ann. 49; empha-
sis added by the Second Circuit). The court continued:
If we were to interpret literally the italicized word
“its”, we would limit statutory violations to the
unauthorized use of the identical established mark.
. . . However, since the use of obvious simulations
or markedly similar marks might have the same
diluting effect as would an appropriation of the origi-
nal mark, the concept of exact identity has been
broadened to that of substantial similarity. Neverthe-
less, in keeping with the original intent of the statute,
the similarity must be substantial before the doctrine
of dilution may be applied.
Id. at 1028-29 (internal citations omitted). The court stopped
short of requiring that “the marks in question . . . be suffi-
ciently similar that confusion may be created as between the
marks themselves.” Id. at 1029. Instead, it held that “the
marks must be ‘very’ or ‘substantially’ similar and that,
absent such similarity, there can be no viable claim of dilu-
tion.” Id.3
3
In a concurring opinion, one member of the panel articulated a differ-
ent approach. He believed that there were two elements necessary to estab-
lish a dilution claim under state law: “an extremely strong mark” and “a
likelihood of dilution.” Mead Data Cent., Inc. v. Toyota Motor Sales,
U.S.A., Inc., 875 F.2d 1026, 1032 (2d Cir. 1989) (Sweet, J., concurring).
2382 LEVI STRAUSS v. ABERCROMBIE & FITCH
[1] Thus, the requirement of identity, or substantial simi-
larity, pre-dates the adoption of the FTDA in 1996 and has its
origins in state dilution law, specifically that of the State of
New York.
B. Case Law Employing the “Identical or Nearly
Identical” Standard
After Playboy Enterprises, we again employed the “identi-
cal or nearly identical” formulation in Thane International,
Inc. v. Trek Bicycle Corp., 305 F.3d 894 (9th Cir. 2002). In
Thane, we grappled with the issue whether Thane’s use of
“OrbiTrek” diluted Trek’s trademarked name. In considering
the “Identity of the Marks,” the court gave a more thorough
explanation for its adoption of the “identical or nearly identi-
cal” standard:
We begin from the recently-established require-
ment that for a dilution claim to succeed, the mark
used by the alleged diluter must be identical, or
nearly identical, to the protected mark. Playboy
Enterprises, 279 F.3d at 805. Such a requirement
comports with the statutory language, the four-part
dilution test derived from that language outlined in
Panavision Int’l L.P. v. Toeppen, 141 F.3d 1316,
Likelihood of dilution, he explained, required a factual inquiry, made on
a “case-by-case” basis, with courts considering the following factors “1)
similarity of the marks 2) similarity of the products covered by the marks
3) sophistication of consumers 4) predatory intent 5) renown of the senior
mark 6) renown of the junior mark.” Id. at 1035. Turning to the “similarity
of the marks,” the concurring judge observed that “[d]ilution is likely only
where the junior mark is similar to the senior mark.” Id. However, “[e]xact
identity between the marks is not required, but the more similar the marks
the higher the likelihood of dilution.” Id. (internal citations omitted).
These factors are very similar to the six enumerated factors set forth by
Congress in the TDRA for courts to employ in determining whether a
mark or trade name is likely to cause dilution. See 15 U.S.C.
§ 1125(c)(2)(B)(i)-(vi).
LEVI STRAUSS v. ABERCROMBIE & FITCH 2383
1324 (9th Cir. 1998) and Avery Dennison [Corp. v.
Sumpton, 189 F.3d 868, 874 (9th Cir. 1999)],[4] and
with the statute’s legislative history and purposes.
The statute establishes that the junior user, to be
liable for dilution, must use “a mark or trade name
. . . after the mark has become famous.” § 1125(c)(1)
(emphasis added). As articulated in Panavision and
Avery Dennison, the test for dilution similarly pro-
vides that to make out an antidilution cause of
action, a plaintiff must show that “its mark is
famous” and “the defendant is making commercial
use of the mark in commerce.” Avery Dennison, 189
F.3d at 874 (emphasis added); see Panavision, 141
F.3d at 1324; see also Mattel, Inc. v. MCA Records,
Inc., 296 F.3d 894, 903 (9th Cir. 2002) (“ ‘Dilution’
refers to the ‘whittling away of the value of a trade-
mark’ when it’s used to identify different products.”)
(emphasis added) (citation omitted). These locutions
indicate that the defendant must use essentially the
same mark, not just a similar one.
Id. at 905 (emphasis in first paragraph added). Thus, Thane
tied the requirement for identity or near identity to the lan-
guage of the then-governing FTDA and to the tests that we
had developed in interpreting the FTDA.
4
In Panavision International, L.P. v. Toeppen, 141 F.3d 1316, 1324 (9th
Cir. 1998), we articulated the following test under the FTDA:
In order to prove a violation of the Federal Trademark Dilution
Act, a plaintiff must show that (1) the mark is famous; (2) the
defendant is making a commercial use of the mark in commerce;
(3) the defendant’s use began after the mark became famous; and
(4) the defendant’s use of the mark dilutes the quality of the mark
by diminishing the capacity of the mark to identify and distin-
guish goods and services. 15 U.S.C. § 1125(c).
Id. at 1324; see also Avery Dennison Corp. v. Sumpton, 189 F.3d 868, 874
(9th Cir. 1999) (same).
2384 LEVI STRAUSS v. ABERCROMBIE & FITCH
In adopting this standard, we also were persuaded by the
legislative history of the FTDA and by the McCarthy trea-
tise’s explanation of the nature of a dilution claim. The legis-
lative history, although “not definitive on the issue,”
“suggest[ed] that the marks must be identical or close there-
to.” Id. Specifically, the Senate Report had given the follow-
ing examples of dilution: Kodak being used for pianos and
Buick being used for aspirin. Id. at 906 (citing S. Rep. No.
100-515, at 7 (1988), as reprinted in 1988 U.S.C.C.A.N.
5577, 5583). Furthermore, referencing the McCarthy treatise,
we observed that “a dilution claim alleges a form of appropri-
ation.” Id. We continued:
Appropriation implies the adoption of the mark
itself, not the use of a similar mark. As discussed
previously, infringement is designed to protect
against consumer confusion about the source of a
product that may arise, inter alia, because a com-
pany uses a similar mark. Dilution, on the other
hand, protects the distinctiveness of a particular
mark whether or not the products compete or con-
sumer confusion exists. § 1127. Because dilution and
likelihood of confusion tests are directed at different
actions, it does not make sense to import the rela-
tively subjective similarity of the marks test from the
likelihood of confusion context into the dilution con-
text. See 4 J. McCarthy, Trademarks and Unfair
Competition, § 24:90.2 (4th ed. 2001).
Id.
Finally, we observed, other courts similarly had adopted an
“identical or nearly identical” requirement. Indeed, Playboy
Enterprises had borrowed this standard from the Eighth Cir-
cuit’s decision in Luigino’s, and other circuits had employed
equally stringent standards for similarity.
Thus, although the identical or nearly identical standard
had its roots in pre-FTDA state dilution law, we determined
LEVI STRAUSS v. ABERCROMBIE & FITCH 2385
in Thane that our adoption of the standard was rooted in the
language of the FTDA, the legislative history and purpose of
that statute, our prior interpretations of the FTDA, and the
policies we believed were embodied in that statute.
C. The Moseley Decision and Adoption of the TDRA
After our decision in Thane, the Supreme Court handed
down a decision that greatly impacted many courts of
appeals’ interpretations of the FTDA. In Moseley v. V Secret
Catalogue, Inc., 537 U.S. 418 (2003), the Court held, contrary
to the approach that had been taken by this court, as well as
the courts of appeals for the Second and Sixth Circuits, that
the text of the FTDA “unambiguously requires a showing of
actual dilution, rather than a likelihood of dilution.” Id. at 433
(emphasis added).
However, this requirement of actual dilution was not long-
lived. In 2006, largely in response to the Moseley decision,
Congress enacted the TDRA. In doing so, Congress did not
simply alter the language on which the Court in Moseley had
relied;5 instead, Congress replaced the FTDA with a more
detailed statute. The TDRA did provide relief for “likely,” as
opposed to actual, dilution. 15 U.S.C. § 1125(c)(1). However,
it also explicitly provided relief for dilution “by blurring” or
by “tarnishment,” id., and defined both types of dilution, id.
§ 1125(c)(2). Section 1125(c)(1) of Title 15 now states in rel-
evant part:
5
The FTDA had provided: “The owner of a famous mark shall be enti-
tled . . . to an injunction against another person’s commercial use in com-
merce of a mark or trade name, if such use begins after the mark has
become famous and causes dilution of the distinctive quality of the mark,
and to obtain such other relief as is provided in this subsection.” 15 U.S.C.
§ 1125(c) (2005) (emphasis added). In Moseley v. V Secret Catalogue,
Inc., 537 U.S. 418 (2003), the Court held that the highlighted language
required a showing of actual dilution.
2386 LEVI STRAUSS v. ABERCROMBIE & FITCH
Subject to the principles of equity, the owner of a
famous mark that is distinctive, inherently or through
acquired distinctiveness, shall be entitled to an
injunction against another person who, at any time
after the owner’s mark has become famous, com-
mences use of a mark or trade name in commerce
that is likely to cause dilution by blurring or dilution
by tarnishment of the famous mark, regardless of the
presence or absence of actual or likely confusion, of
competition, or of actual economic injury.
15 U.S.C. § 1125(c)(1). Subsection (c)(2) defines “dilution by
blurring” accordingly:
(B) For purposes of paragraph (1), “dilution by blur-
ring” is association arising from the similarity
between a mark or trade name and a famous mark
that impairs the distinctiveness of the famous mark.
In determining whether a mark or trade name is
likely to cause dilution by blurring, the court may
consider all relevant factors, including the following:
(i) The degree of similarity between the mark or
trade name and the famous mark.
(ii) The degree of inherent or acquired distinctive-
ness of the famous mark.
(iii) The extent to which the owner of the famous
mark is engaging in substantially exclusive use of
the mark.
(iv) The degree of recognition of the famous mark.
(v) Whether the user of the mark or trade name
intended to create an association with the famous
mark.
LEVI STRAUSS v. ABERCROMBIE & FITCH 2387
(vi) Any actual association between the mark or
trade name and the famous mark.
15 U.S.C. § 1125(c)(2)(B).
[2] Several aspects of the TDRA are worth noting. The
first, as mentioned previously, is that Congress did not merely
make surgical linguistic changes to the FTDA in response to
Moseley. Instead, Congress created a new, more comprehen-
sive federal dilution act. Furthermore, any reference to the
standards commonly employed by the courts of appeals—
“identical,” “nearly identical,” or “substantially similar”—are
absent from the statute. The TDRA defines “dilution by blur-
ring” as the “association arising from the similarity between
a mark or a trade name and a famous mark that impairs the
distinctiveness of the famous mark.” Id. § 1125(c)(2)(B)
(emphasis added). Moreover, in the non-exhaustive list of
dilution factors that Congress set forth, the first is “[t]he
degree of similarity between the mark or trade name and the
famous mark.” Id. § 1125(c)(2)(B)(i). Thus, the text of the
TDRA articulates a different standard for dilution from that
which we utilized under the FTDA.
D. Post-TDRA, Ninth Circuit Case Law
Abercrombie claims that, despite the absence of any refer-
ence in the TDRA to the “identical or nearly identical stan-
dard,” this standard nonetheless survived the passage of the
TDRA. It points to three cases from this court to support that
claim. Two of these cases, Perfumebay.com Inc. v. eBay, Inc.,
506 F.3d 1165 (9th Cir. 2007), and Jada Toys, Inc. v. Mattel,
Inc., 518 F.3d 628 (9th Cir. 2008), were addressed by the par-
ties in their briefs; Abercrombie maintains that Levi Strauss’s
“entire appeal is foreclosed” by these cases. See Appellee’s
Br. 12. The third case, Visa International Service Association
v. JSL Corp., No. 08-15206, 2010 WL 2559003 (9th Cir. June
28, 2010), was decided after oral argument in this appeal.
Abercrombie submitted the case as further authority in sup-
2388 LEVI STRAUSS v. ABERCROMBIE & FITCH
port of its position, see Fed. R. App. P. 28(j), and Levi Strauss
similarly submitted its comments on the case’s application to
the matter presently before the court.
[3] Our review of these cases leads us to conclude that the
issue that we must decide today—whether, to establish dilu-
tion by blurring under the TDRA, the junior mark must be
“identical or nearly identical” to the senior mark—was not
presented or squarely resolved in these prior cases.
1.
The first case on which Abercrombie relies is Perfume-
bay.com. In Perfumebay.com, eBay brought a dilution claim
under California state law. In analyzing the state-law dilution
claim, we observed that the state claim was subject to the
same analysis as a federal claim, which required that “[t]he
mark used by the alleged diluter . . . be identical, or nearly
identical, to the protected mark for a dilution claim to suc-
ceed.” Perfumebay.com, 506 F.3d at 1180 (internal quotation
marks and citation omitted). Although applying California
dilution law, we noted that our “emphasis on the extent to
which the marks are nearly identical and the strength of the
senior mark is bolstered by Congress’ passage of the Trade-
mark Dilution Revision Act of 2006,” which explicitly
included “ ‘[t]he degree of similarity between the mark or
trade name and the famous mark’ ” and “ ‘[t]he degree of
inherent or acquired distinctiveness of the famous mark’ ” in
its dilution analysis. Id. at 1181 n.9 (quoting 15 U.S.C.
§ 1125(c)(2)(B)).
We do not believe these observations ought to be consid-
ered a binding determination that the “identical or nearly
identical” standard survives the passage of the TDRA. We
have held that, “where a panel confronts an issue germane to
the eventual resolution of the case, and resolves it after rea-
soned consideration in a published opinion, that ruling
becomes the law of the circuit, regardless of whether doing so
LEVI STRAUSS v. ABERCROMBIE & FITCH 2389
is necessary in some strict logical sense.” United States v.
Johnson, 256 F.3d 895, 914 (9th Cir. 2001) (en banc). How-
ever, in Perfumebay.com, we did not “confront” the issue of
the applicable standard under the TDRA. Perfumebay.com
involved a dilution claim under California state law. See Per-
fumebay.com, 506 F.3d at 1180 n.8 (“eBay’s dilution claim is
made pursuant to Cal. Bus. & Prof. Code § 14330.”). We pre-
viously had held that California’s state dilution law was
equivalent to the FTDA. Id. (“eBay’s ‘state law dilution claim
is subject to the same analysis as its federal claim.’ ” (quoting
Panavision Int’l, 141 F.3d at 1324)). At the time that Per-
fumebay.com was decided, the California legislature had not
altered its dilution statute to reflect the changes in the newly
enacted TDRA.6 Because the California statute had not been
altered in the same manner as the federal statute, there simply
was no opportunity for us to consider whether a change in the
language of the federal statute effected a change in the stan-
dard applicable to federal claims. Indeed, our reference to the
TDRA in Perfumebay.com was limited to our observation in
footnote 9 that our emphasis on the degree of similarity and
the strength of the senior mark was “bolstered” by Congress’s
passage of the TDRA. Id. at 1181 n.9. In sum, Perfume-
bay.com did not present an opportunity for the court to
address squarely the question currently before this panel. See
Starbucks Corp. v. Wolfe’s Borough Coffee, Inc., 588 F.3d 97,
109 n.4 (2d Cir. 2009) (referencing Perfumebay.com and stat-
ing that “[t]he Ninth Circuit has not resolved whether this
‘essentially the same’ standard also applies to federal dilution
claims post-TDRA”).
6
“In June, 2007, the International Trademark Association amended the
anti-dilution provisions of the Model Bill to reflect the Trademark Dilu-
tion Revision Act of 2006. The 2007 version of the model anti-dilution
provisions generally track the 2006 federal law with changes necessary to
adopt the law to a state-wide regime. . . . In 2007, California was one of
the first states to enact the 2007 version of the Model State Trademark
Bill.” J. Thomas McCarthy, McCarthy on Trademarks and Unfair Compe-
tition § 22:6.25 (4th ed. 2010) (footnotes omitted).
2390 LEVI STRAUSS v. ABERCROMBIE & FITCH
2.
The same is true of our decision in Jada Toys. Jada Toys
had been using the name HOT RIGZ in conjunction with its
line of toy trucks; Mattell believed that HOT RIGZ was dilut-
ing its HOT WHEELS trademark and therefore sued Jada
Toys under state and federal law. Both at the time that Mattel
instituted its action in 2004, and at the time the district court
granted summary judgment against Mattel in 2005, the FTDA
was the governing federal law. Similarly, when the action was
briefed to this court, the parties framed their arguments under
the FTDA, not the TDRA. Thus we originally decided the
case under the FTDA and later amended the opinion in
response to the TDRA.7
The reason for amending the opinion was to apply the new
“likelihood of dilution standard” incorporated into the then
recently enacted TDRA. See Jada Toys, 518 F.3d at 634 n.2.
Beyond that change, the amended Jada Toys opinion did not
confront any of the other differences between the TDRA and
the FTDA. This is unsurprising because the plaintiff-appellant
Mattel was entitled to a remand both under Thane’s interpre-
tation of the FTDA as well as under the TDRA’s multifactor
inquiry.
In both Jada Toys opinions, we stated that Mattel had to
“show that (1) the mark is famous and distinctive; (2) the
defendant is making use of the mark in commerce; (3) the
defendant’s use began after the mark became famous; and (4)
the defendant’s use of the mark is likely to cause dilution,”
changing only the fourth element in the amended opinion to
“likely to cause dilution” instead of “dilutes the quality of the
mark.” Compare Jada Toys, 518 F.3d at 634, with Jada Toys,
Inc. v. Mattel, Inc., 496 F.3d 974, 980-81 (9th Cir. 2007). We
made no change to the statement that:
7
See Jada Toys, Inc. v. Mattel, Inc., 496 F.3d 974 (9th Cir. 2007),
amended and superseded by Jada Toys, Inc. v. Mattel, Inc., 518 F.3d 628
(9th Cir. 2008).
LEVI STRAUSS v. ABERCROMBIE & FITCH 2391
[F]or a plaintiff to establish that the mark is being
used in commerce (as per step two), “the mark used
by the alleged diluter must be identical, or nearly
identical, to the protected mark.” Thane Int’l, 305
F.3d at 905(noting that this circuit’s description of
dilution by blurring and by tarnishment requires a
defendant to use the plaintiff’s actual mark) (citation
omitted). In order to be nearly identical, two marks
“must be ‘similar enough that a significant segment
of the target group of customers sees the two marks
as essentially the same’ ” Id. at 906 (quoting Playboy
Enters., 279 F.3d at 806 n. 41).
Compare Jada Toys, 518 F.3d at 634, with Jada Toys, 496
F.3d at 981. Applying Thane’s standard, Jada Toys found a
triable issue of fact regarding whether HOT WHEELS was
“nearly identical” to HOT RIGZ. See Jada Toys, 518 F.3d at
635.
We went on, after finding the marks nearly identical, sepa-
rately to determine that a reasonable jury could find that HOT
RIGZ was likely to dilute Mattel’s HOT WHEELS trademark
based on the six factors enumerated in the TDRA. See Jada
Toys, 518 F.3d at 635-36. Because Mattel survived Jada Toys’
motion for summary judgment under the standard articulated
in Thane, we were not confronted with the discrepancy
between Thane’s standard and the textual changes embodied
in the TDRA.8 In revising the Jada Toys opinion to be consis-
tent with the TDRA’s shift to a “likelihood” of dilution stan-
dard, which was the principal focus of the case, we never
considered that other parts of the TDRA might affect our stan-
dard of analysis for the requisite degree of similarity.
8
Recall that Thane’s standard considered only the marks’ visual similar-
ity and the senior mark’s distinctiveness, see Perfumebay.com, 506 F.3d
at 1180-81 (quoting Thane, 305 F.3d 907 n.7), whereas the TDRA consid-
ers all relevant factors and enumerates six of particular relevance, see 15
U.S.C. § 1125(c)(2)(B).
2392 LEVI STRAUSS v. ABERCROMBIE & FITCH
[4] As with Perfumebay.com, we do not believe that Jada
Toys binds us to apply the “identical or nearly identical” stan-
dard for purposes of assessing a dilution by blurring claim
under the TDRA. Jada Toys was tried under the FTDA, and
the FTDA was the law employed by the parties in their brief-
ing before this court. We resolved the case without any dis-
cussion of whether Thane’s standard survived the change in
law. Indeed, we copied the FTDA analysis of “near identity”
into the TDRA opinion without mentioning the material
changes in the statute’s language regarding similarity. The
parties never briefed the applicability of the TDRA or the sig-
nificance of its revisions to the antidilution law. Nor did the
facts of the case require us to confront the possible implica-
tions of a less stringent similarity standard. In short, our use
of the “identical or nearly identical” standard in Jada Toys
cannot represent a definitive resolution of the issue before us
“after reasoned consideration,” and, therefore, does not estab-
lish the standard for dilution under the TDRA.
3.
Finally, the parties have asked us to consider our recent
decision in Visa International. In that case, Visa International
sued JSL Corporation, which operates eVisa, “a ‘multilingual
education and information business that exists and operates
exclusively on the Internet,’ at www.evisa.com.” Visa Int’l
Serv. Ass’n, 2010 WL 2559003, at *1. Visa International
claimed that eVisa was likely to dilute the Visa trademark.
The district court agreed with Visa International and granted
summary judgment in its favor. We affirmed. Both parties
rely on the same passage of the opinion:
A plaintiff seeking relief under federal anti-
dilution law must show that its mark is famous and
distinctive, that defendant began using its mark in
commerce after plaintiff’s mark became famous and
distinctive, and that defendant’s mark is likely to
LEVI STRAUSS v. ABERCROMBIE & FITCH 2393
dilute plaintiff’s mark. See Jada Toys, Inc. v. Mattel,
Inc., 518 F.3d 628, 634 (9th Cir. 2008). . . .
There are two types of dilution, but here we are
concerned only with dilution by blurring, which
occurs when a mark previously associated with one
product also becomes associated with a second. See
15 U.S.C. § 1125(c)(2)(B); Mattel, Inc. v. MCA
Records, Inc., 296 F.3d 894, 903-04 (9th Cir. 2002).
This weakens the mark’s ability to evoke the first
product in the minds of consumers. “For example,
Tylenol snowboards, Netscape sex shops and Harry
Potter dry cleaners would all weaken the ‘commer-
cial magnetism’ of these marks and diminish their
ability to evoke their original associations.” Mattel,
296 F.3d at 903. . . .
. . . Congress has enumerated factors courts may
use to analyze the likelihood of dilution, including
the similarity between the two marks and the distinc-
tiveness and recognition of the plaintiff’s mark. 15
U.S.C. § 1125(c)(2)(B)(i), (ii), (iv); see also Per-
fumeBay.com, Inc. v. eBay, Inc., 506 F.3d 1165,
1181 n.9 (9th Cir. 2007). And, in an appropriate
case, the district court may conclusively determine
one or more of these factors before trial.
The marks here are effectively identical; the only
difference is the prefix “e,” which is commonly used
to refer to the electronic or online version of a brand.
That prefix does no more to distinguish the two
marks than would the words “Corp.” or “Inc.” tacked
onto the end. See Horphag Research Ltd. v. Garcia,
475 F.3d 1029, 1036 (9th Cir. 2007) (use of identical
mark provides “circumstantial evidence” of dilu-
tion).
And Visa is a strong trademark. “In general, the
more unique or arbitrary a mark, the more protection
2394 LEVI STRAUSS v. ABERCROMBIE & FITCH
a court will afford it.” Nutri/System, Inc. v. Con-Stan
Indus., Inc., 809 F.2d 601, 605 (9th Cir. 1987).
Id. at *1-2.
In its 28(j) letter submitted on July 1, 2010, Abercrombie
sets forth three reasons why “the opinion conclusively estab-
lishes the applicability of the ‘identical or nearly identical’
standard under the TDRA.” First, Abercrombie notes that
Visa International relied on cases that employ the “identical
or nearly identical” standard. Second, Abercrombie believes
that the “Court’s emphasis on the ‘one mark, two products’
paradigm shows that the ‘identical or nearly identical’ stan-
dard provides the appropriate legal framework under the
TDRA.” Third, it continues, “in assessing the similarity
between the two marks at issue, the Court applied the identi-
cal or nearly identical standard.”
We cannot accept Abercrombie’s view of Visa Interna-
tional. We believe it clear that our reference to prior cases and
our use of the “one mark, two products” paradigm were of lit-
tle significance to our analysis. Nor can we agree that Visa
International “applied the identical or nearly identical stan-
dard.” In evaluating the district court’s judgment in Visa
International, we stated that “Congress has enumerated fac-
tors courts may use to analyze the likelihood of dilution,
including the similarity between the two marks and the dis-
tinctiveness and recognition of the plaintiff’s mark.” Visa Int’l
Serv. Ass’n, 2010 WL 2559003, at *1 (emphasis added). We
then went on to evaluate the evidence presented according to
those statutory standards. With respect to similarity, we
observed that “[t]he marks here are effectively identical” with
the “only difference” being “the prefix ‘e.’ ” Id. at *2. Our
reference to “effectively identical” was a factual assessment
of the similarity of the two marks with which we were pre-
sented. There simply is no basis for asserting that we required
a showing of identity or near identity in evaluating Visa Inter-
LEVI STRAUSS v. ABERCROMBIE & FITCH 2395
9
national’s claim under the TDRA. With respect to the factors
of the distinctiveness and recognition of the Visa mark, we
observed that Visa had “introduced uncontroverted evidence
that Visa is the world’s top brand in financial services and is
used for online purchases almost as often as all other credit
cards combined.” Id.
[5] Having reviewed all of the authorities provided by
Abercrombie, we conclude that we are the first panel to
address the issue whether the “identical or nearly identical”
standard survives Congress’s adoption of the TDRA. We turn
now to that question.
E. Interpretation of the TDRA
1.
Our method of statutory interpretation is well established:
Statutory interpretation begins with the language
of the statute. See United States v. Ron Enters., Inc.,
489 U.S. 235, 241 (1989). When the plain meaning
of a statutory provision is unambiguous, that mean-
ing is controlling. Id. at 242. To determine the plain
meaning of a statutory provision, we examine not
only the specific provision at issue, but also the
structure of the statute as a whole, including its
object and policy. See Green v. Commissioner, 707
F.2d 404, 405 (9th Cir. 1983). If ambiguity exists,
we may use legislative history as an aid to interpreta-
tion. See id.; Mt. Graham Red Squirrel v. Madigan,
954 F.2d 1441, 1453 (9th Cir. 1992).
9
Indeed, as we shall discuss later, see infra at 2399-2400, we believe
that the manner in which the panel in Visa International Service Associa-
tion v. JSL Corp., No. 08-15206, 2010 WL 2559003 (9th Cir. June 28,
2010), applied the TDRA is consistent with the interpretation of the statute
upon which we base today’s decision.
2396 LEVI STRAUSS v. ABERCROMBIE & FITCH
Children’s Hosp. & Health Ctr. v. Belshe, 188 F.3d 1090,
1096 (9th Cir. 1999) (parallel citations omitted). Here, more-
over, the language of the new statute, when compared to the
language of the former statute, provides a firm basis for deci-
sion.
[6] Beginning with subsection (c)(1) of 15 U.S.C. § 1125,
Congress provided that “the owner of a famous mark . . . shall
be entitled to an injunction against another person who . . .
commences use of a mark or trade name in commerce that is
likely to cause dilution.” 15 U.S.C. § 1125(c) (emphasis
added). When referring to the junior mark, Congress did not
authorize an injunction against another person who com-
mences use of “the” mark; use of the definite article “the”
clearly would have signaled that the junior mark had to be the
same as the senior. Instead, Congress employed the indefinite
article “a,” which indicates that any number of unspecified,
junior marks may be likely to dilute the senior mark.
[7] Turning to the language of subsection (c)(2)(B), the
TDRA defines “dilution by blurring” as the “association aris-
ing from the similarity between a mark and a trade name and
a famous mark that impairs the distinctiveness of the famous
mark.” Id. § 1125(c)(2)(B) (emphasis added). Congress did
not require an association arising from the “substantial” simi-
larity, “identity” or “near identity” of the two marks. The
word chosen by Congress, “similarity,” sets forth a less
demanding standard than that employed by many courts under
the FTDA.
This analysis of the language of the statute, and our com-
parison of this language with the now-repealed statute, are
further supported by Congress’s decision to employ, in sub-
section (c)(2)(B), a non-exhaustive list of relevant factors to
determine when dilution has occurred. Congress’s implemen-
tation of such a methodology is simply not compatible with
a determination that identity, near identity or substantial simi-
larity are necessary to constitute a threshold showing for relief
LEVI STRAUSS v. ABERCROMBIE & FITCH 2397
under § 1125(c). Indeed, Congress chose instead to make the
“degree of similarity between the mark or trade name and the
famous mark,” id. § 1125(c)(2)(B)(i) (emphasis added), to be
the first of the six (or more) relevant factors to be considered.
No doubt, similarity has a special role to play in the imple-
mentation of the new statute’s multifactor approach. After all,
dilution by blurring is defined by the statute as an “association
arising from the similarity between a mark . . . and a famous
mark.” Id. § 1125(c)(1)(B). It is also the first factor listed in
the multifactor approach. Nevertheless, Congress’s decision
to make “degree of similarity” one consideration in a multi-
factor list strongly suggests that it did not want “degree of
similarity” to be the necessarily controlling factor.
Finally, we believe that it is significant that, in adopting the
TDRA, Congress decided to re-write 15 U.S.C. § 1125(c), as
opposed to altering discrete wording or subsections. This
action suggests that Congress did not wish to be tied to the
language or interpretation of prior law, but instead crafted a
new approach to our consideration of dilution-by-blurring
claims.10
10
Abercrombie urges us to look to the legislative history of the TDRA
as support for its claim that the “identical or nearly identical” standard sur-
vives adoption of that statute. However, “ ‘we do not resort to legislative
history to cloud a statutory text that is clear.’ ” Cumbie v. Woody Woo,
Inc., 596 F.3d 577, 581 n.11 (9th Cir. 2010) (quoting Ratzlaf v. United
States, 510 U.S. 135, 147-48 (1994)). The TDRA contains no language
that suggests that identity, near identity or substantial similarity is required
in order to show dilution. Furthermore, use of legislative history is particu-
larly inappropriate in circumstances where, as here, the “legislators’ pub-
lished statements do not squarely address the question presented.” Oregon
v. Ashcroft, 368 F.3d 1118, 1136 (9th Cir. 2004). Although the House
Committee cited its earlier report on the FTDA to help explain the TDRA,
see H.R. Rep. No. 109-23 (2005), reprinted in 2006 U.S.C.C.A.N. 1091,
the legislative history is not precise enough to overcome the clarity of the
text. We need not decide the extent of continuity between the FTDA and
TDRA because, insofar as it is relevant to this case, there is none — Thane
is plainly inconsistent with the TDRA’s text.
2398 LEVI STRAUSS v. ABERCROMBIE & FITCH
[8] Thus, the plain language of 15 U.S.C. § 1125(c) does
not require that a plaintiff establish that the junior mark is
identical, nearly identical or substantially similar to the senior
mark in order to obtain injunctive relief. Rather, a plaintiff
must show, based on the factors set forth in § 1125(c)(2)(B),
including the degree of similarity, that a junior mark is likely
to impair the distinctiveness of the famous mark.
Our interpretation of the TDRA is compatible with the case
law of the only other court of appeals to have addressed
squarely the question whether the requirement of identity or
substantial similarity survives the TDRA. In Starbucks, the
company brought a federal dilution claim against Wolfe’s
based on its use of the term “Charbucks” in connection with
one of its coffee blends. The district court found that “the
marks were not substantially similar” and “that ‘[t]his dissimi-
larity alone is sufficient to defeat [Starbucks’s] blurring
claim.’ ” Starbucks, 588 F.3d at 107. The Second Circuit dis-
agreed. It acknowledged that, pre-TDRA, it had required
plaintiffs to show that the marks were “ ‘very’ or ‘substan-
tially similar’ ” before they could prevail on a federal dilution
claim. Id. (quoting Playtex Prods., Inc. v. Georgia-Pacific
Corp., 390 F.3d 158, 167 (2d Cir. 2004)). This was no longer
the case under the TDRA, the court explained:
The post-TDRA federal dilution statute, however,
provides us with a compelling reason to discard the
“substantially similar” requirement for federal trade-
mark dilution actions. The current federal statute
defines dilution by blurring as an “association aris-
ing from the similarity between a mark . . . and a
famous mark that impairs the distinctiveness of the
famous mark,” and the statute lists six non-
exhaustive factors for determining the existence of
an actionable claim for blurring. 15 U.S.C.
§ 1125(c)(2)(B). Although “similarity” is an integral
element in the definition of “blurring,” we find it sig-
nificant that the federal dilution statute does not use
LEVI STRAUSS v. ABERCROMBIE & FITCH 2399
the words “very” or “substantial” in connection with
the similarity factor to be considered in examining a
federal dilution claim.
Id. at 108. Furthermore, the court continued, in addition to
leaving out any modifier for “similarity,” Congress also
employed specific language that cannot be reconciled with a
requirement of substantial similarity; the court stated:
[O]ne of the six statutory factors informing the
inquiry as to whether the allegedly diluting mark
“impairs the distinctiveness of the famous mark” is
“[t]he degree of similarity between the mark or trade
name and the famous mark.” 15 U.S.C.
§ 1125(c)(2)(B)(i) (emphasis added). Consideration
of a “degree” of similarity as a factor in determining
the likelihood of dilution does not lend itself to a
requirement that the similarity between the subject
marks must be “substantial” for a dilution claim to
succeed.
Id. (internal citations omitted); see also Tiffany (NJ) Inc. v.
eBay, Inc., 600 F.3d 93, 111 n.18 (2d Cir. 2010) (“We have
recently explained that under the [TDRA] the similarity
between the famous mark and the allegedly blurring mark
need not be ‘substantial’ in order for the dilution by blurring
claim to succeed.”).11
Although we did not speak directly to the issue in Visa
International, we believe that our decision in that case is con-
sonant with the interpretation of the TDRA that we adopt
here. In Visa International, we reviewed the district court’s
11
We have held that the factors a court may consider under
§ 1125(c)(2)B)(i-vi) are not necessarily entitled to the same weight. A suf-
ficiently strong showing of similarity can overcome all other relevant fac-
tors. See Visa Int’l, 610 F.3d at 1090 (affirming summary judgment on the
basis of a junior mark “effectively identical” to a “strong” senior mark).
2400 LEVI STRAUSS v. ABERCROMBIE & FITCH
grant of summary judgment in favor of Visa on its trademark
dilution claim. We noted that “likelihood of dilution” is a fac-
tual inquiry “generally not appropriate for decision on sum-
mary judgment.” Visa Int’l, 2010 WL 2559003, at *1. We
pointed out that Congress had “enumerated factors courts may
use to analyze the likelihood of dilution, including the similar-
ity between the two marks and the distinctiveness and recog-
nition of the plaintiff’s mark.” Id. (citing 15 U.S.C.
§ 1125(c)(2)(B)(i), (ii), (iv)). Evaluating these factors, we
determined that “[t]he district court was quite right in granting
summary judgment to Visa.” Id. at *4. In essence, we consid-
ered the evidence in light of the factors set forth in
§ 1125(c)(2)(B). We identified “similarity” as one of the fac-
tors a court should consider and, notably, did not mention a
requirement that the marks be “identical or nearly identical.”
Most importantly, we took the degree of similarity into
account along with the other § 1125(c)(2)(B) factors in deter-
mining whether dilution was likely to occur. In short,
although not addressing the issue now before us head-on in
Visa International, we applied the methodology that we have
described here and made no reference to a requirement of
“identity or near identity” of the marks.12
F. The District Court’s Error Was Not Harmless
Abercrombie also submits that, even if the district court
erred in applying the “identical or nearly identical” require-
ment, the judgment nonetheless should be affirmed because
any error was harmless. According to Abercrombie, the dis-
trict court determined the Arcuate mark and the Ruehl design
were not visually similar. Because “similarity” is the correct
12
Although we hold that a particular degree of similarity is not a thresh-
old, similarity is the necessary predicate for dilution analysis. See 15
U.S.C. § 1125(c)(2)(B) (“ ‘[D]ilution by blurring’ is association arising
from the similarity between a mark or trade name and a famous mark
. . . .”). And greater degrees of similarity manifestly are more likely to
support a finding of dilution.
LEVI STRAUSS v. ABERCROMBIE & FITCH 2401
standard under the TDRA, Levi Strauss suffered no prejudice
as a result of the district court’s application of the incorrect
standard. Abercrombie also notes that, in assessing the likeli-
hood of dilution, the district court not only considered
whether the marks were identical, but also took into account
the other five factors set forth in § 1125(c)(2)(B). Abercrom-
bie maintains that, because the court engaged in a weighing
of all of the factors, the “similarity” requirement played only
an insignificant role in its determination.
Abercrombie presents a very strained view of the district
court’s opinion. In its order entering judgment for Abercrom-
bie, the court noted that “[t]he advisory jury found that the
Ruehl design and the Arcuate mark were not identical or
nearly identical.” Levi Strauss & Co., 2009 WL 1082175, at
*7. It also observed that the test for similarity in a dilution
context “is more stringent than in the infringement context.”
Id. (internal quotation marks omitted). After evaluating the
visual depictions of the designs, the court concluded that
“[t]his evidence demonstrates that a significant segment of the
target group of customers would not view the marks as essen-
tially the same.” Id. at *8 (emphasis added). The court then
reviewed Dr. Sood’s testimony and concluded that, given the
survey shortcomings, there was insufficient evidence that tar-
get customers would “see the Ruehl design and the Arcuate
mark as essentially the same.” Id. (emphasis added). In sum-
mary, the court stated: “[T]he Court concludes, consistent
with the advisory jury’s finding, that [Levi Strauss] has not
established that [Abercrombie] is making commercial use of
a mark that is identical or nearly identical to the Arcuate
mark.” Id. at *9. Use of the “identical or nearly identical”
standard permeated the court’s analysis and provided the basis
upon which the court evaluated the evidence.
This standard also played a pivotal role in the court’s deter-
mination that the Ruehl design was not likely to dilute the
Arcuate mark. With respect to the similarity of the marks, the
court observed that the two marks must be essentially the
2402 LEVI STRAUSS v. ABERCROMBIE & FITCH
same and that, “[f]or the reasons set forth above, the Court
finds that the Ruehl design and the Arcuate mark are not visu-
ally similar.” Id. In other words, the court equated similarity
with sameness and employed the latter, more stringent defini-
tion when entering its findings of fact.
[9] Finally, our review of the district court’s balancing of
the relevant factors convinces us that application of the incor-
rect standard affected its dilution determination. According to
the district court, degree of similarity was only one of three
factors that weighed in Abercrombie’s favor. The district
court assumed, without deciding, that Levi Strauss also had
two factors—acquired distinctiveness and degree of
recognition—that weighed in its favor. Thus, application of
the correct, less-demanding standard could have tipped the
balance in favor of Levi Strauss. The degree of similarity
between the Ruehl and Arcuate marks may be insufficient to
support a likelihood of dilution, but that conclusion can come
only after consideration of the degree of similarity in light of
all other relevant factors and cannot be determined conclu-
sively by application of an “essentially the same” threshold.
[10] Given the relative balance of the parties’ positions, we
cannot say, with any confidence, that the district court would
have reached the same result absent the legal error. See Gal-
damez v. Potter, 415 F.3d 1015, 1025 (9th Cir. 2005)
(“presum[ing] prejudice where civil trial error is concerned”
and shifting the burden to the defendant “to demonstrate that
‘it is more probable than not that the jury would have reached
the same verdict’ had it been properly instructed” (quoting
Obrey v. Johnson, 400 F.3d 691, 701 (9th Cir. 2005))). We
therefore must reverse the judgment of the district court with
respect to Levi Strauss’s claim under the TDRA.
REVERSED and REMANDED.
LEVI STRAUSS v. ABERCROMBIE & FITCH 2403
APPENDIX
Levi Strauss’s “Arcuate” design.
2404 LEVI STRAUSS v. ABERCROMBIE & FITCH
Abercrombie’s “Ruehl” design.