THOIP v. The Walt Disney Company
Filing
154
OPINION AND ORDER The clerk of the court is directed to close these motions [Docket Nos. 138 and 141] and this case. SO ORDERED. (Signed by Judge Shira A. Scheindlin on 5/10/11) (jmi)
UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF NEW YORK
._--------------------------------------------------
)(
THOIP (A Chorion Limited Company),
OPINION AND ORDER
Plaintiff,
08 Civ. 6823 (SAS)
- against
THE WALT DISNEY COMPANY,
DISNEY CONSUMER PRODUCTS,
INC., and DISNEY DESTINATIONS,
LLC,
Defendants .
._-------------------------------------------------- )(
SHIRA A. SCHEINDLIN, U.S.D.J.:
I.
INTRODUCTION
This action arises out ofTHOIP's claim to rights to a family of
unregistered trademarks stemming from a collection of children's books featuring
the "Mr. Men" and "Little Miss" ("MMLM") cartoon characters. Under the
Lanham Act and the common law, THOIP alleges that its family of marks was
infringed by two lines ofT-shirts from The Walt Disney Company, Disney
Consumer Products, Inc., and Disney Destinations, LLC (collectively "Disney").
In an opinion dated August 13,2010, I granted Disney's motion for
summary judgment on the issue of "forward confusion," holding that there was
1
virtually no chance that a consumer seeing an allegedly infringing “Miss Disney
shirt”1 or “Little Miss Disney shirt”2 would think it came from, was affiliated with,
or was approved by THOIP.3 As to “reverse confusion,” I found that four of the
eight “Polaroid factors” weighed in THOIP’s favor, but reopened discovery,
allowing the parties the opportunity, if they wished, to conduct expert surveys on
reverse confusion.4 Both parties accepted the invitation. Disney now moves to
exclude THOIP’s survey, and both parties move for summary judgment on the
issue of reverse confusion. Disney’s motions are granted, and THOIP’s is denied.
II.
BACKGROUND
A.
The Ostberg Survey
1
See THOIP v. The Walt Disney Co., 736 F. Supp. 2d 689, 698-99
(S.D.N.Y. 2010) (“THOIP II”).
2
See id. at 699-701.
3
See id. at 715. In an earlier opinion considering the parties’ dueling
Daubert motions, I excluded THOIP’s survey (the “Ford Survey”) because it was
fundamentally flawed and, therefore, not a reliable indicator of consumer
confusion. See THOIP v. The Walt Disney Co., 690 F. Supp. 2d 218, 234-41
(S.D.N.Y. 2010) (“THOIP I”). Disney’s survey, which I determined was
sufficiently reliable to be admitted into evidence, found that virtually no one seeing
an allegedly-infringing Disney shirt thought that it was connected to THOIP or the
MMLM characters. See id. at 225-27, 241-42. This opinion assumes familiarity
with the facts, applicable legal standards, and analysis as set forth in THOIP I and
THOIP II.
4
See THOIP II, 736 F. Supp. 2d at 715.
2
In support of its motion, THOIP proffers a survey from its retained
expert Dr. Henry Ostberg that purports to examine whether consumers perceive the
“Little Miss THOIP shirts”5 at issue in this litigation to be “produced by or with the
permission or approval of [Disney].”6
1.
Design and Operation
Dr. Ostberg conducted two surveys based on the “array or Squirt
research model”7 – one for the Miss Disney shirts (“Miss Disney survey”) and
another for the Little Miss Disney shirts (“Little Miss Disney survey”). In brief,
respondents for each survey were asked a series of questions after being exposed to
two portfolios of products: one containing either Miss Disney or Little Miss
Disney shirts, and one containing Little Miss THOIP shirts.
The surveys were conducted in eighteen shopping malls throughout
5
See id. at 696-98.
6
See Expert Report of Dr. Henry Ostberg (“Ostberg Survey”), Ex. 1 to
Declaration of Martin Schwimmer, Counsel for THOIP, in Support of THOIP’s
Motion for Partial Summary Judgment (“Schwimmer Decl.”), at 6.
7
See id. at 7. See also 6 J. Thomas McCarthy, McCarthy on
Trademarks and Unfair Competition (“McCarthy on Trademarks”) § 32:173.50 at
32-370 to 32-373 (describing the Squirt format and its namesake SquirtCo. v.
Seven-Up Co., 628 F.2d 1086 (8th Cir. 1980)); id. § 32:177 at 32-377 to 32-378
(2008) (describing “line-up” survey methods, “[a] more subtle form of ‘Squirt’
survey”). Id. § 32:173.50 at 32-372.
3
the United States in the fall of 2010.8 Screening interviews located women who
indicated that they intended to purchase a T-shirt with an image of a cartoon
character, either for themselves or for a female child, during the next six months.9
The participants in each survey were divided (roughly evenly) into a “test group”
and a “control group”10 and were interviewed in research offices located in the
shopping malls.11
For the Miss Disney survey, test group respondents were first shown
five photographic arrays of three products each, displayed in an eleven- by
seventeen-inch portfolio:12
•
Three Bella Sara jigsaw puzzles (Master Pieces Puzzles)
•
Three “I Can Read!/Fancy Nancy” booklets (Harper-Trophy)
•
Three pajama pants with cartoon characters (Paul Frank Industries)
•
Three Dora cartoon T-shirts (Nickelodeon)
8
See Ostberg Survey at 15.
9
See id. at 7, 17. Roughly half of the 351 participants in the Miss
Disney survey, and roughly half of the 338 participants in the Little Miss Disney
suvey, were 18 to 34 years old; the other half were 35 years of age or older. See id.
10
See id.
11
See id. at 17.
12
See id. at 8, 9.
4
•
Three Miss Disney shirts
The Miss Disney shirts included in the portfolio were “Miss Chatterbox,” “Miss
Fabulous,” and “Miss Attitude”13 (see following page):
13
See id. at 9. I discuss Dr. Ostberg’s basis for selecting the products
included in the arrays in Part IV.A.1.
5
Case 1:08-cv-06823-SAS Document 140-1
Filed 01/26/11 Page 38 of 51
Portfolio with Miss Disney Shirts
The Little Miss Disney test group participants were shown the same
photographic arrays, but with three Little Miss Disney shirts instead of the Miss
Disney shirts. The Little Miss Disney shirts were “Little Miss Perfect,” “Little
Miss Wicked,” and “Little Miss Sassy”:14
The order in which the three sets of products were included in the
portfolio was rotated among the different respondents.15 Each photographed
product purportedly displayed “the appropriate neck label or company indicia.”16
Respondents were then asked a series of questions about their
television viewing to provide a time interval and “a ‘distracter’ between what
14
See id. at 17.
15
See id. at 10.
16
Id. at 12.
7
respondents saw in the first and second portfolios, simulating the extraneous
conditions which may exist during a typical shopping visit.”17
Next, interviewers showed respondents a second portfolio, also
including five groups of three products each:
•
Three pajama pants with cartoon characters (Paul Frank Industries)
•
Three “I Can Read!/Fancy Nancy” booklets (Harper-Trophy)
•
Three 750-piece Jigsaw Puzzles (CEACO)
•
Three cartoon sweatshirts (Joe Boxer)
•
Three Little Miss THOIP shirts
The THOIP shirts in the second portfolio were “Little Miss Sunshine,” “Little Miss
Chatterbox,” and “Little Miss Trouble”18 (see following page):
17
18
Id. at 11.
See id. at 12. Dr. Ostberg claims that “[b]y having two sets of
products which were the same and two sets of products which were different in the
second portfolio than in the first portfolio, there was no suggestion as to whether
the THOIP t-shirts in the second portfolio were the same or different than [sic] the
[Miss or Little Miss] Disney shirts in the first portfolio.” Id. at 11.
8
Case 1:08-cv-06823-SAS Document 140-1
Filed 01/26/11 Page 41 of 51
Test Portfolio
Control group participants in each survey were first shown the same
first portfolio.19 In the second portfolio, however, photographs of the three Little
Miss THOIP shirts were replaced with three “control” T-shirts that “provide[d] an
appropriate and real-life control for determining the extent to which respondents
associate any product which has a cartoon character and wording including a name
and/or character trait with the Walt Disney Company.”20 As “controls,” Dr.
Ostberg selected three Crown Creative Co., Ltd. (“Crown Creative”) T-shirts
containing “an illustration of a ‘cute French bulldog’ together with wording that
included the bulldog’s name and/or character trait, either ‘Rebecca Bonbon,’ ‘I’m
Way Way Too Cool,’ or simply ‘Love:’”21
19
See id. at 13.
20
Id. at 14.
21
Id. at 13-14.
10
After respondents had viewed the second portfolio, the interviewer
then pointed to the THOIP (or Creative Crown) T-shirts and asked (twice), “Do
you think that these shirts – the ones on this page – are put out by the same
company as any of the products which you saw in the first portfolio?”22 If the
respondent gave an affirmative reply, she was then asked, “Which product or
products in the first portfolio do you think are put out by the same company as the
shirts you are looking at now?”23 If a respondent mentioned a company or product,
22
Id. at 12.
23
Id.
11
the interviewer asked why she thought that product was made by the same
company as the THOIP (or Creative Crown) shirts.24
To determine “whether or not respondents believed that there was
some other relationship between the THOIP [or Creative Crown] and Disney tshirts,” the interviewer again pointed to the photographs of the THOIP (or Creative
Crown) shirts and asked, “Do you think that the company that puts out these shirts
obtained – or needed to obtain – permission or approval from any other company
to market these shirts?”25 If the respondent gave an affirmative reply, she was
asked, “Who do you think gave – or needed to give – its permission or approval for
these shirts to be marketed?”26 Again, if a respondent mentioned a company or
product, the interviewer asked why she thought such permission or approval was
given or needed.27
Dr. Ostberg then subtracted “the percentage of respondents who
mentioned the Disney Company as being the source of the Crown Creative t-shirts
24
See id. at 13. I discuss the manner in which Dr. Ostberg coded the
respondents’ responses in Part IV.A.3, infra.
25
Id.
26
Id.
27
See id.
12
from the percentage who mentioned the Disney Company in their response to the
questions about THOIP t-shirts.”28
2.
Results
Based on his coding, Dr. Ostberg found that 16 percent of respondents
in the Miss Disney test group believed the THOIP and Miss Disney shirts “were
‘put out’ by the same company or that the THOIP shirts needed permission or
approval from the Disney Company.”29 Meanwhile, 8 percent of respondents in
the control group believed the THOIP and Crown Creative “control” shirts were
“put out” by the same company, or that the Crown Creative shirts required
permission or approval from Disney,30 creating a net likelihood of reverse
confusion between the Little Miss THOIP and Miss Disney shirts of 8
percent.31
28
Id. at 14.
29
Id. at 19 (emphasis added). In particular, 15 percent believed the
shirts were produced by the same company, primarily because of the “similarity of
the words or lettering” (9 percent), “the characters on the shirts” (5 percent), and
the “overall design or layout” (2 percent). Id. One percent believed the company
producing the THOIP shirts obtained or needed to obtain permission or approval
from “the Disney Company.” Id.
30
See id. at 20-21. In particular, 5 percent believed the Crown Creative
shirts were produced by the same company and mentioned the Disney Company,
while 3 percent thought the Crown Creative shirts needed Disney’s permission or
approval. See id. at 21.
31
See id. at 21.
13
As for the Little Miss Disney test group, 24 percent believed the
THOIP and Little Miss Disney shirts were “put out” by the same company or that
the company producing the THOIP shirts needed permission or approval from the
Disney Company.32 Five percent of respondents in the control group expressed
similar views about the THOIP and Crown Creative “control” shirts,33 resulting in
a net likelihood of reverse confusion between the Little Miss THOIP and Little
Miss Disney shirts of 19 percent.34
3.
Rebuttal to Ostberg Survey
Disney proffers a report and declaration from its retained expert, Dr.
Itamar Simonson, to rebut the Ostberg Survey.35 Dr. Simonson’s report is
addressed below in Part IV.A.
B.
The Helfgott Survey
32
See id. at 22. Twenty percent believed the former (primarily because
of “similarity of the words or lettering” (15 percent), “the characters on the shirts”
(3 percent), and “the overall look” (3 percent)). Id. Four percent believed that
“permission or approval from the Disney Company was needed.” Id. at 23.
33
See id. at 23. All 5 percent believed the Crown Creative and Little
Miss Disney shirts were produced by the same company. See id. at 23-24.
34
35
See id. at 24.
See Expert Report of Dr. Itmar Simonson (“Simonson Report”), Ex. 1
to Declaration of Dr. Itamar Simonson in Support of Defendants’ Motion in Limine
and for Summary Judgment (“Simonson Decl.”).
14
Disney proffers a survey from Dr. Myron J. Helfgott that, like Dr.
Ostberg’s, purports to “determine whether or not T-shirts put out by THOIP are
likely to cause consumers to think they are put out by, in association with, or with
the permission of Disney.”36
1.
Design and Operation
Dr. Helfgott conducted a so-called “Eveready” or “monadic” survey.37
Respondents were shown either a Little Miss THOIP shirt or a control shirt, and
then were queried about source, association, and permission. The two THOIP
shirts were (1) “Little Miss Bossy” and (2) “Little Miss Chatterbox”:38
36
Expert Report of Dr. Myron J. Helfgott (“Helfgott Survey”), Ex. 2 to
Simonson Decl., at 5.
37
See McCarthy on Trademarks § 32:174 at 32-367 to 32-369
(describing the Eveready format and its namesake Union Carbide Corp. v. EverReady, Inc., 531 F.3d 366 (7th Cir. 1976)). The six hundred respondents who
participated in the survey were screened and divided by age similarly to those who
participated in the Ostberg surveys. See Helfgott Survey at 6, 9.
38
See Helfgott Survey at 7. Both THOIP shirts bore the name “THOIP”
in very small print at the side or bottom of the cartoon illustration on the shirts, as
part of the trademark statement. See id. However, since all THOIP T-shirts carry a
neck label bearing the name “Junk Food” (the manufacturer of the THOIP Tshirts), Dr. Helfgott “decided that maximum benefit would be accrued by showing
the Junk Food label in half of each exhibit group” in order to measure precisely
“the degree to which a likelihood of reverse confusion score is affected by the
presence (or absence) of the neck label.” Id. at 8. He found that it had no effect on
reverse confusion scores. See id. at 12.
15
The two control T-shirts were a (1) “Got Green” and (2) “Berry Sweet” T-shirt
bearing those names and a cartoon illustration.39
During their interviews, each respondent was shown one of the four
t-shirts and asked to look it over “as you would in a store and were considering
39
See id. at 7. According to Dr. Helfgott, “[n]either of the control
cartoons [is] reminiscent of either the Little Miss THOIP cartoon characters or the
Disney cartoon characters.” Id.
16
buying it.”40 The interviewer then asked respondents the following three likelihood
of confusion questions:
•
“What company do you think puts out this T-shirt, or don’t you know?”
•
“Do you think the company that puts out this T-shirt puts it out themselves,
or in association with some other company, or don’t you know?”
•
“Do you think the company that puts out this T-shirt got permission from
some other company, did not get permission from some other company, or
don’t you know?”41
If the respondent answered any of the above questions with a response other
than “don‘t know,” a follow-up “reasons” question was asked: “What in particular
about this T-shirt makes you think (that)?”42 For each interview in which at least
one name was mentioned in response to the previous questioning, a final question
was added for each company name: “Please tell me all of the products and brands
you know of which this company puts out.” Probe “any others?”43 In order to
avoid omitting any possible indication of Disney reverse confusion, Dr. Helfgott
40
Id. at 10.
41
Id.
42
Id.
43
Id.
17
“decided to accept any reference to Disney as a likelihood of reverse confusion
response.”44
2.
Results
Six respondents out of 400 associated a THOIP test shirt with Disney
(1.5 percent of the sample).45 Three control cell respondents out of 200 identified
the control exhibits as Disney’s (1.5 percent).46 “The fact that the test cell
measurement of reverse confusion has the same incidence of Disney mentions as
the control cell measurement of noise, and not higher, indicates that there is no
residual evidence of reverse confusion.”47
III.
APPLICABLE LAW
A.
Reverse Confusion
Reverse confusion “is the misimpression that the junior user is the
source of the senior user’s goods.”48 “A ‘reverse confusion’ situation exists where
44
Id. at 11. Thus, an interview was counted as indicating a likelihood of
reverse confusion if answers to any of the three confusion questions included any
reference to Disney.
45
See id. at 12.
46
See id.
47
Id. (emphasis added).
48
Banff, Ltd. v. Federated Dep’t Stores, 841 F.2d 486, 490 (2d Cir.
1988) (“Banff II”). Accord Patsy’s Brand, Inc. v. I.O.B. Realty, Inc., 317 F.3d 209,
218 (2d Cir. 2003) (upholding district court’s “weighing the actual confusion
18
the junior user is able to amass such trademark strength in its imitative mark that
the senior user’s products become associated with the junior user in the minds of
consumers.”49 In determining whether there is a likelihood of reverse confusion,
courts within the Second Circuit apply the eight-factor balancing test set forth in
Polaroid Corporation v. Polarad Electronics Corporation,50 just as they do to
factor in favor of [the senior user’s] [b]rand” where “[a]lthough no customer
bought a jar of the Defendants’ sauce thinking it was made by the Plaintiff (normal
confusion), in a few instances, a customer of the Defendants’ restaurant bought, or
at least saw, a jar of the Plaintiff’s sauce, thinking it came from the Defendants
(reverse confusion)”); Sterling Drug, Inc. v. Bayer AG, 14 F.3d 733, 742 (2d Cir.
1994) (“This case primarily involves reverse confusion, where consumers mistake
Sterling’s products as originating with Bayer AG.”); W.W.W. Pharm. Co., Inc. v.
Gillette Co., 984 F.2d 567, 571 (2d Cir. 1993), narrowed on other grounds by
Deere & Co. v. MTD Prods., Inc., 41 F.3d 39, 46 & n.10 (2d Cir. 1994) (“Reverse
confusion has been thought to injure the reputation of the prior user of the mark by
causing potential customers to consider it a trademark infringer.”) (citing Banff II,
841 F.2d at 490); id. at 574 (“In this case, where reverse confusion is alleged, the
allegation is that a subsequent user’s selection of a trademark “is likely to cause
consumers to believe, erroneously, that the goods marketed by the prior user are
produced by the subsequent user.”) (citing Lang v. Retirement Living Pub. Co.,
Inc., 949 F.2d 576, 583 (2d Cir. 1991)); Lang, 949 F.2d at 583 (citing Banff II, 841
F.2d at 490); Yarmuth-Dion, Inc. v. D’ion Furs, Inc., 835 F.2d 990 (2d Cir. 1987)
(reverse confusion is “an instance in which a person acquainted with the junior
user’s name becomes confused when confronted with the senior user’s name”).
But see Black Diamond Sportswear, Inc. v. Black Diamond Equip., Ltd., No.
06-3508-cv, 2007 WL 2914452, at *4 (2d Cir. Oct. 5, 2007) (quoting Freedom
Card, Inc. v. JPMorgan Chase & Co., 432 F.3d 463, 473 (3d Cir. 2005), for the
proposition that “[t]he offender in a reverse confusion case will typically exploit
confusion to push the senior user out of the market.”).
49
Star Indus., Inc. v. Bacardi & Co. Ltd., 412 F.3d 373, 388 n.3 (2d Cir.
2005).
50
287 F.2d 492, 495 (2d Cir. 1961).
19
assess claims for forward confusion.51 The Polaroid factors are: (1) the strength
of plaintiff’s mark; (2) the similarity of plaintiff’s and defendant’s marks; (3) the
proximity of the products; (4) the likelihood that plaintiff will “bridge the gap”; (5)
actual confusion between products; (6) defendant’s good or bad faith in adopting
the mark; (7) the quality of defendant’s product; and (8) the sophistication of the
buyers.52
IV.
DISCUSSION
A.
Motion in Limine to Exclude the Ostberg Survey
Disney argues that the Ostberg Survey suffers from a number of flaws
that individually and collectively require its exclusion under Federal Rules of
Evidence 702 and 403. I agree.
1.
The Ostberg Survey Failed to Replicate Actual Marketplace
Conditions
First, the Ostberg Survey failed to replicate actual marketplace
conditions.53 In THOIP I, this Court found the Ford Survey inadmissible in part
51
See, e.g., Plus Prods. v. Plus Discount Foods, Inc., 722 F.2d 999,
1004 n.6 (2d Cir. 1983).
52
See Polaroid, 287 F.2d at 495. See also THOIP II, 736 F. Supp. 2d at
705-06 (elaborating on the likelihood of confusion analysis).
53
See THOIP I, 690 F. Supp. 2d at 230-40 (discussing the criteria to
consider in assessing the validity and reliability of a survey, and discussing the
Ford Survey’s failure to replicate actual marketplace conditions).
20
because Dr. Ford tested the THOIP and Disney products using a sequential array
methodology – despite the absence of evidence that such products would be
encountered in close proximity in the actual marketplace.54 Yet, in testing for
reverse confusion, Dr. Ostberg designed a similar sequential array survey in which
(1) respondents were exposed to specific sets of three Miss Disney/Little Miss
Disney shirts and then three THOIP shirts and (2) the THOIP and Disney shirts
were presented amongst an array of extraneous products. Moreover, Dr. Ostberg
selected the Disney and THOIP shirts used in his survey because they “were
believed to be the top sellers for each of the two companies”55 – not because he
had any evidence of how the shirts were actually sold. “Unlike books, however,
54
See id. at 239-40. I also criticized the Ford Survey for failure to use
hang tags and neck labels clearly. See id. at 239. Disney now argues that the
photographs used in Dr. Ostberg’s arrays “oftentimes partially obscured the neck
labels on the shirts and, in any case, made them smaller and less prominent than
they were in real life, making it difficult for respondents to notice or read the neck
labels on the shirts.” Disney’s Memorandum of Law in Support of Motion in
Limine and for Summary Judgment (“Disney Mem.”) at 10. However, after
reviewing all thirty photographs used in the Ostberg Survey, I conclude that the
photographic arrays adequately displayed the shirts’ neck labels. Indeed, some of
the respondents cited “Junk Food,” the THOIP shirt licensee, as the source of some
of the T-shirts. See Statistical Tables, App. A to Ostberg Report, Tables 102 and
105. In any event, Dr. Helfgott found that the presence (or absence) of the Junk
Food neck label, in his survey, had no effect on reverse confusion scores. See
Helfgott Survey at 12.
55
Ostberg Survey at 10. Accord id. at 4 nn.1-2. By comparison, Dr.
Ford coupled THOIP and Disney shirts based on resemblance. See THOIP I, 690
F. Supp. 2d at 237.
21
where the best-sellers are often sold together on the same ‘best-seller’ rack at a
bookstore, the fact that a particular T-shirt in a line of shirts is a top seller does not
mean that it will be sold in close proximity with other best-selling shirts.”56
Instead, THOIP argues that “[t]he evidence . . . shows not only that the t-shirts
flowed through the same trade channels, but also that Disney and THOIP shipped
to the same distributors,” and that “[b]y selecting the best-selling designs, Dr.
Ostberg reasonably approximated the most likely proximate encounters.”57 It calls
Disney’s “assertion” that it will never market its various products side-by-side with
THOIP’s “implausible,” directing the Court to photographs of (1) “Disney
trademarked merchandise [‘Hannah Montana’ children’s underwear] displayed
side-by-side with THOIP’s products [‘Little Miss Sunshine’ children’s
underwear]”58 and (2) a sale rack at Walmart displaying THOIP and Disney T56
Disney’s Memorandum in Opposition to THOIP’s Motion for Partial
Summary Judgment (“Disney Opp. Mem.”) at 7-8. THOIP has not submitted to
the Court any documents showing the actual sales figures of the THOIP or Disney
shirts. According to Disney, this is also a “telling omission as the actual sales
show that the three ‘best-selling’ Miss Disney shirts, in a line of four shirts,
represented a relatively small number of sales, making it all the more unlikely that
even the most popular Miss Disney T-shirts would be encountered with the THOIP
T-shirts.” Id. at 8.
57
THOIP’s Memorandum of Law in Opposition to Disney’s Motion in
Limine and for Summary Judgment (“THOIP Opp. Mem.”) at 5.
58
Id. at 6 (citing Photograph, Ex. 13 to Schwimmer Decl.) (originally an
exhibit to the Declaration of Dale M. Cendali, counsel for Disney, in Support of
Motion in Limine to Exclude Expert Testimony of Dr. Gary Ford [Docket No. 68]).
22
shirts – though not the T-shirts at issue in this case – sold side-by-side.59 Finally, it
directs the Court to Junk Food’s website, where both Disney and Little Miss
THOIP T-shirts are sold “one click away from each other.”60
I reject THOIP’s argument. Evidence of three instances (one virtual61)
in which Disney products were sold alongside Little Miss THOIP products says
nothing about the likelihood of consumers encountering the Little Miss Disney or
Miss Disney shirts together with the Little Miss THOIP shirts (or Little Miss
THOIP products, for that matter) in the marketplace. It does not overcome the lack
of evidence that such products were sold side-by-side in the past62 and it does not
prove that Disney will sell the products side-by-side in the future.
Thus, notwithstanding that many of the same retailers sold THOIP
and Miss Disney shirts, and despite the evidence that Disney shirts
are at times sold side-by-side non-Disney shirts, such general
information does not justify [Dr. Ostberg’s] specific choices. . . .
In addition, with respect to the Little Miss Disney line, even
assuming arguendo the soundness of THOIP’s position that a
sequential array survey is justified where the products are found
59
Id. at 7 (citing Photograph, Ex. 15 to Schwimmer Decl.).
60
Id. (directing the Court to http://www.junkfoodclothing.com/#).
61
Indeed, “the Internet is a big place [where] millions of T-shirts are
available for sale . . . across a countless number of websites. Moreover, if
particular products truly are sold in proximity over the Internet, a proper survey
would approximate that online purchasing environment by having respondents
actually navigate the Internet.” Disney Opp. Mem. at 7.
62
See THOIP I, 690 F. Supp. 2d at 237.
23
in different stores within short distances of each other, THOIP has
not shown a reasonable likelihood that consumers would have
proximately encountered its shirts and the Little Miss Disney
shirts in a critical number of real world situations. Though
THOIP’s shirts were available in thousands of stores nationwide,
the Little Miss THOIP and Little Miss Disney shirts were on sale
in a small number of nearby stores in at most three geographic
locations (New York City, Anaheim, and Orlando). In these
circumstances, there is not a reasonable likelihood that consumers
would have encountered in close proximity Little Miss THOIP
and Little Miss Disney shirts, let alone those pairs specifically
tested by Dr. [Ostberg].63
What is more, products other than the Miss Disney shirts “were selected, not
because they were necessarily sold next to the Disney shirts, but to permit
respondents to identify/describe the products in the first portfolio (if any) they
thought were made by or were related to the company which puts out the THOIP
shirts.”64 Indeed, Dr. Ostberg admitted in his deposition that he had no basis to
believe that any of these products used in his portfolio – including puzzles, books,
and pajama pants – were ever sold alongside any of the shirts at issue.65 “[A]
63
Id. at 237-38 (footnotes omitted).
64
Ostberg Survey at 10.
65
See Deposition of Dr. Henry Ostberg (“Ostberg Dep.”), Ex. 3 to
Simonson Decl., at 191:5-23 (“[T]he only purpose of showing other products that
might be somewhat similar in some stores is to create a context so as to avoid
undue focus on one product. And it really doesn’t matter what the other products
are that I included in the portfolio.”).
24
legally-probative estimation of consumer confusion must be tethered to
marketplace conditions,”66 but Dr. Ostberg’s was not.
In this vein, while I rejected Disney’s argument in THOIP I that the
Ford Survey
failed to approximate marketplace conditions because it
improperly created artificial awareness of THOIP’s claimed mark
among survey respondents, and, relatedly, “tested ‘conditional
probability,’ i.e., only the potential rate of confusion amongst the
limited subset of people who were already aware of THOIP’s
claimed mark, not the actual likelihood of confusion among the
far broader universe of potential consumers who are not[,]”67
Disney’s argument carries more weight in the context of a survey testing for
reverse confusion – especially where THOIP has proffered no evidence showing
that Disney advertised or promoted the Miss and Little Miss Disney lines in the
past, or has plans to do so in the future.68 I agree with Disney that “THOIP cannot
66
THOIP I, 690 F. Supp. 2d at 237.
67
Id. at 239 (citations omitted).
68
See discussion infra pp. 43-46. See also McCarthy on Trademarks §
23:10 at 23-73:
a survey cannot be run in a reverse confusion case prior to the
junior user’s saturation of the market with its mark because, until
that time, consumers have not been exposed to the relatively large
advertising and promotion of the junior user that is the hallmark
of a reverse confusion case.
Cf. THOIP I, 690 F. Supp. 2d at 240 (justifying the Ford Survey’s creation of
artificial awareness of the THOIP shirts because “though public awareness of
THOIP’s name and that of its parent company is low, there is an appreciable
awareness of the Little Miss – and related Mr. Men – characters”).
25
point to the ‘Disney’ name or the Disney characters to argue reverse confusion.
Rather, THOIP must demonstrate that the allegedly infringing shirts – not just
Disney in general – have swamped or are likely to swamp THOIP’s brand based on
the admissible[] record.”69 Otherwise, “[i]f such a survey were enough to support a
finding of reverse confusion . . . Disney could be held liable for ‘reverse confusion’
even if it put out only a single shirt with no promotion.”70
2.
The Ostberg Survey Lacked a Proper Control
“To fulfill its function, a control should ‘share[ ] as many
characteristics with the experimental stimulus as possible, with the key exception
of the characteristic whose influence is being assessed.’”71 THOIP argues that the
French bulldog shirts were proper controls because they “share all but the
infringing characteristics” with the Disney shirts – namely, they “have a cartoon
character (but not one of the parties’ characters)” and “wording,”72 but not (1) the
words “LITTLE MISS,” (2) not words describing a characteristic of the character,
69
Disney Opp. Mem. at 2 n.1.
70
Id. at 5.
71
THOIP I, 690 F. Supp. 2d at 240 (quoting Shari Seidman Diamond,
Reference Guide on Survey Research, in Reference Manual On Scientific Evidence
at 258 (Federal Judicial Center 2d ed. 2000)).
72
This Court faulted the Ford Survey for failing to use words in control
products. See id.
26
and (3) not words in block letters.73 I disagree. Dr. Ostberg’s control – “t-shirts
bearing an image of a French bulldog and no descriptive term with hearts,
splattered paint, or stars”74 – shared very few similarities with the THOIP or
Disney shirts:
Not only did the font, typeface, layout and size of the character in
relation to the wording on the control shirts differ, but the control
shirts were cluttered, giving them a very different overall look and
feel from the THOIP or Disney shirts.75
Indeed, the “overall look and feel” of Dr. Ostberg’s “control” T-shirts was much
more comparable to that of the “Dora” T-shirts – notably the only other T-shirts
besides the Disney shirts contained in the first portfolio.76 Dr. Helfgott’s “Got
Green” and “Berry Sweet” control shirts, by contrast, “were as similar as possible
to the test shirts without the protectible elements.”77 For these reasons, even
controlling for confusion based on protectible elements, “fewer respondents were
likely to ‘match’ the control shirts with the Disney shirts [than were likely to
73
THOIP Opp. Mem. at 10. Accord Ostberg Report at 14.
74
Disney Mem. at 12.
75
Id.
76
Compare supra page 11 (Crown Creative shirts) with supra page 6
(Dora shirts) (fourth set of photographs).
77
THOIP Opp. Mem. at 10 (citing Louis Vuitton Malletier v. Dooney &
Bourke, Inc., 525 F. Supp. 568, 595 (S.D.N.Y. 2007)) (purporting to describe the
Crown Creative shirts).
27
‘match’ the Little Miss THOIP shirts with the Disney shirts], improperly inflating
the results.”78
3.
Dr. Ostberg’s Improper Coding Inflated “Confusion” Rates
Dr. Ostberg also inflated his “confusion” rates by counting
respondents as confused even if their responses were not necessarily indicative of
reverse confusion. For example, Dr. Ostberg counted as confused respondents who
said that the two shirts (shown about a minute apart) displaying the word “Miss” or
“Little Miss” were put out by the “same company,” even though these respondents
did not mention Disney.79
But if respondents did not identify “Disney” or at the very least
one of Disney’s famous characters by name in identifying the
source of the T-shirts (instead saying “Little Miss” or “Miss”
something), this shows mere matching and not whether any of the
respondents believed that both sets of shirts were put out by
Disney.80
Moreover, Dr. Ostberg also counted as confused those respondents who responded
that the THOIP shirts were put out by the same company as a “Little Miss
78
Id.
79
See Simonson Decl. ¶ 10.
80
Id. See also Franklin Res. Inc. v. Franklin Credit Mgmt. Corp., 988 F.
Supp. 322, 335 (S.D.N.Y. 1997) (“Surveys which do nothing more than
demonstrate the respondents’ ability to read are not always probative on the issue
of likelihood of confusion.”).
28
Sunshine” T-shirt.81 But Disney did not put out a “Little Miss Sunshine” shirt –
THOIP did – and thus there was no Disney “Little Miss Sunshine” shirt in the first
array.82 Although respondents may have been confusing THOIP’s Little Miss
Sunshine T-shirt with the 2006 movie by that name, “such irrelevant confusion
should not have been counted towards Dr. Ostberg’s results.”83
Finally, Dr. Ostberg also counted as confused those respondents who
picked the Disney shirts for reasons such as: “same character/character
resemblance,” “the characters are theirs,” “targeted to the same age group,” and
“bright colors.”84 But such answers are not indicative of reverse confusion because
they are not based on the purportedly protectable elements of THOIP’s mark;
references to cartoon characters or colors are not references to the mark.85 THOIP
81
See Simonson Decl. ¶ 10.
82
See Disney Mem. at 13 (citing Simonson Decl. ¶ 10).
83
Id. According to Dr. Simonson, recalculating the results just to
exclude mentions of “Little Miss Sunshine,” the confusion rate would be 6.3
percent for the Miss Disney shirts and 16 percent for the Little Miss Disney shirts.
See Simonson Decl. ¶ 10. Recalculating the results to exclude all responses merely
mentioning the words on the shirts (including the Little Miss Sunshine mentions),
the confusion rate is further reduced to 1.2 percent for the Miss Disney shirts and
6.8 percent for the Little Miss Disney shirts. See id.
84
85
See Simonson Decl. ¶ 11.
See id. According to Dr. Simonson, recalculating the results to
exclude such responses, and netting out the control answers not based on these
same characteristics, the confusion rates decrease for the Miss Disney shirts to 6.8
29
argues that because Dr. Ostberg coded the control responses the same way, any
bias in the results is eliminated.86 However, because the Ostberg survey lacked a
proper control – one that differed markedly from the THOIP and Disney shirts – I
cannot agree.
4.
The Ostberg Survey Suffered from Demand Effects
Disney argues that the Ostberg Survey also suffered from demand
effects. Respondents were first shown a portfolio of photographs of five sets of
products, only two of which were T-shirts – the Little Miss or Miss Disney shirts
and shirts bearing the “well known children’s character Dora the Explorer” – while
the other products in the portfolio consisted of “completely disparate products.”87
They were then specifically pointed to the page in the portfolio with the three
THOIP (or Creative Crown) T-shirts and asked leading questions about “shirts”
instead of neutral questions about “products.”88 Thus, Disney contends,
respondents were skewed to “match the THOIP ‘shirts’ that they were pointed to
with the only ‘shirts’ in the first array that looked anything like the THOIP shirts –
percent and the Little Miss Disney shirts to 9.7 percent. See id.
86
See THOIP Opp. Mem. at 11.
87
Disney Mem. at 11.
88
See id.
30
the Disney shirts.”89 However, non-T-shirt distractors were also used in the control
arrays, so that any spurious confusion would have been caused there as well, and
eliminated when the control group positives were subtracted out of the final results.
Nonetheless, I agree with Disney that “the Disney shirts ‘stood out
like a bearded man in a lineup with four clean-shaven men’” in the test surveys:90
Respondents were not likely to match the THOIP shirts with the
puzzle boxes bearing images of winged horses in magical settings,
the “Fancy Nancy” booklets depicting scenes from the stories
entitled “My Family History,” “The Delectable Cupcakes,” and
“Pajama Day,” or the pajama pants bearing small repeating
images of skulls or monkey heads with no words. . . . Nor were
respondents likely to match the THOIP shirts with the Dora shirts;
not only is Dora a very well-known character, but also the Dora
shirts presented in the array looked nothing like the THOIP shirts.
Rather, they included the distinct Dora character with the name
“Dora” without a descriptive term, or included other nondescriptive terms such as “totally” or “imagine” in disparate fonts
amidst a clutter of other images, including hearts and flowers.
Given the slanted structure of Dr. Ostberg’s array, the survey
“test[s] nothing more than the memory and common sense of a
respondent” to match the “Little Miss” or “Miss” Disney t-shirts
with THOIP’s “Little Miss” t-shirts.91
89
Id.
90
Id. (quoting Sunbeam Corp. v. Equity Indus. Corp., 635 F. Supp. 625,
634 (E.D. Va. 1986)).
91
Id. at 11-12 (quoting Simon Prop. Group L.P. v. Mysimon, Inc., 104 F.
Supp. 2d 1033, 1051 (S.D. Ind. 2000), and citing Starter Corp. v. Converse, Inc.,
170 F.3d 286, 297 (2d Cir. 1999) (affirming exclusion of survey that “was little
more than a memory test, testing the ability of the participants to remember the
names of the shoes they had just been shown”)).
31
Although the control survey likely suffered from some of these demand effects as
well, as mentioned earlier the control T-shirt was much more cluttered than the
THOIP or Miss Disney and Little Miss shirts, thereby impeding their ability to
control for any such “noise” in the survey design – especially given their
resemblance to the Dora shirts.
In sum, because the Ostberg Survey failed to replicate actual
marketplace conditions, lacked a proper control, improperly counted certain
responses as indicating confusion, and suffered from demand effects, it is not a
reliable indicator of consumer confusion, and is inadmissible.
B.
Motions for Summary Judgment
THOIP and Disney move for partial summary judgment on the issue
of reverse confusion.
1.
Likelihood of Confusion
Applying the Polaroid factors in THOIP II,92 I concluded that the
strength of THOIP’s mark (factor one), the “similarity of the allegedly-infringing
elements of Disney’s shirts to THOIP’s mark” (factor two), the competitive
proximity of the parties’ shirts (factor three), and their equal quality (factor seven)
92
Before analyzing the likelihood of confusion under the Polaroid
factors in THOIP II, I first concluded that THOIP’s mark was entitled to trademark
protection without a showing of “acquired meaning.” 736 F. Supp. 2d at 708.
32
“all weigh in THOIP’s favor [on the issue of reverse confusion].”93 However, I
held that “[w]hile Disney almost undoubtedly intended to copy THOIP’s shirts,” “a
trier of fact could not view the evidence as demonstrating that Disney intended to
‘exploit the goodwill and reputation’ of THOIP by creating shirts ‘with the intent
to sow confusion between two companies’ products’”94 – factor six.95 Factor four,
I held, was “irrelevant”96 and factor eight was “neutral.”97 I reopened discovery to
93
Id. at 715.
94
Id. at 714 (quoting Star Indus., 412 F.3d at 388).
95
Discussing factor six in the context of reverse confusion, Judge
Denise Cote of this Court has opined that
it is unlikely that a larger and better known junior user intends to
trade on the reputation of the lesser-known plaintiff. This factor
is therefore less relevant in a reverse confusion inquiry, although
any finding that the defendant adopted its mark with an intent to
cause any form of confusion should weigh in favor of the plaintiff.
First Nat’l Bank of Omaha, Inc. v. Mastercard Intern. Inc., Nos. 03 Civ. 707, 02
Civ. 3691, 2004 WL 1575396, at *12 (S.D.N.Y. July 15, 2004) (citing A & H
Sportswear, Inc. v. Victoria’s Secret Stores, Inc., 237 F.3d 198, 232 (3d Cir.
2000)). In any event, the record evidence here also does not “support a reasonable
inference that [Disney] intended to promote such confusion, for example, by
flooding the market with its [Miss and Little Miss] products and causing
consumers to believe erroneously that [THOIP’s] goods were produced by
[Disney].” Black Diamond Sportswear, 2007 WL 2914452, at *4 (affirming
district court’s determination that defendant did not act in bad faith). Instead, in
developing the Little Miss Disney and Miss Disney shirts, Disney believed that it
was “following a trend” that it perceived in the marketplace. THOIP II, 736 F.
Supp. 2d at 713 (quotation marks omitted).
96
THOIP II, 736 F. Supp. 2d at 712.
97
Id. at 715.
33
afford THOIP the opportunity to submit evidence of actual reverse confusion – the
fifth Polaroid factor.98 Before turning to an analysis of that factor in light of the
Ostberg and Helfgott Surveys, however, I revisit my discussion of factor one in
light of Disney’s argument that “while forward confusion is more likely when the
plaintiff’s mark is strong, the opposite is true with respect to reverse confusion, as
a strong mark is less likely to be swamped by a junior user.”99
a.
Strength of Plaintiff’s Mark (Revisited)
Evaluating the strength of plaintiff’s mark in THOIP II, I held that
THOIP’s mark was “inherently distinctive because it combines a fanciful character,
an arbitrary format, and a descriptive phrase.”100 However, district courts in this
Circuit have held that, in a reverse confusion case, the court should look to the
comparative strength of the junior user’s (Disney’s) mark when assessing the first
Polaroid factor.101
98
See id.
99
Disney Opp. Mem. at 3 n.3 (citing O’Keefe v. Ogilvy & Mather
Worldwide, Inc., 590 F. Supp. 2d 500, 521 (S.D.N.Y. 2008)).
100
101
THOIP II, 736 F. Supp. 2d at 702.
See O’Keefe, 590 F. Supp. 2d at 521 (“[A] well known mark – one
that has a high degree of acquired distinctiveness – is less susceptible to this sort of
[reverse] confusion than a mark that is not widely recognized.”); First Nat’l Bank
of Omaha, 2004 WL 1575396, at *12 (“A plaintiff with a mark that is
commercially weak . . . is more likely to succeed in establishing reverse confusion,
particularly against a defendant with a far stronger mark.”) (emphasis added)
34
Although I did not discuss the strength of Disney’s mark when
considering the first Polaroid factor in THOIP II, I did consider it in the context of
the third factor, “similarity of marks”:
The biggest difference between the parties’ shirts is
that THOIP’s shirts bear its characters and Disney’s shirts feature
its characters. Hangtags and neck labels further bore Disney’s
name. The Second Circuit has “repeatedly found that the presence
of a distinct brand name may weigh against a finding of confusing
similarity[,]” yet “has also held that, in some circumstances, ‘the
addition of a trade name does not necessarily alleviate the problem
of confusion of marks, and indeed, can aggravate it, as ‘a
purchaser could well think plaintiff had licensed defendant as a
second user.’”[102] The Second Circuit “addressed the tension
between these two lines of cases and decided that, at least where
the junior and senior marks are not identical, the presence of a
trade name will tend to militate against a finding of
confusion.”[103]
Thus, in Playtex Products, Inc. v.
Georgia-Pacific Corp., the court concluded that the presence of
the brand “Quilted Northern” on the defendants’ product
(“Moist-Ones” towelettes) served to decrease confusion with the
(citing A & H Sportswear, 237 F.3d at 231); id. at *13 (discounting the first
Polaroid factor in reverse confusion inquiry where defendant-junior user’s mark
was “no stronger [than the senior user’s mark] and [wa]s unlikely to attain
significant commercial strength”); Strange Music, Inc. v. Strange Music, Inc., 326
F. Supp. 2d 481, 488 (S.D.N.Y. 2004) (“In a reverse confusion case the court
should look to the strength of the junior user’s mark ‘because the doctrine of
reverse confusion properly should focus, not on the senior user’s mark, but that of
the intervening junior user.’”) (quoting Sunenblick v. Harrell, 895 F. Supp. 616,
627 (S.D.N.Y. 1995), aff’d, 101 F.3d 684 (2d Cir. 1996)).
102
Playtex Prods., Inc. v. Georgia-Pacific Corp., 390 F.3d 158, 164-65
(2d Cir. 2004) (quoting Arrow Fastener Co., Inc. v. Stanley Works, 59 F.3d 384,
395 (2d Cir. 1995)).
103
Id.
35
plaintiff’s product (“Wet Ones” towelettes).[104] Similarly, in
Arrow Fastener Co. v. Stanley Works, the Second Circuit
concluded that the presence of defendants’ brand
(“Stanley-Bostitch”) diminished confusion where the marks at
issue were not identical.[105]
Whether the Disney characters serve to increase or
dispel confusion is a somewhat complex question because the
characters are not, as in Playtex Products or Arrow Fastener, an
additional feature apart from the elements of Disney’s shirts that
THOIP challenges. Rather, Disney’s characters are part of the
alleged infringement, just as THOIP’s characters are part of its
mark. In any event, it is clear that, despite the strong similarities
identified above, the parties’ shirts are not identical given that
each uses its own characters. Therefore, the images of Disney’s
famous characters, along with the presence of Disney’s name on
the tags and labels, dispel forward confusion. Indeed, as
addressed below, only one out of 1,200 respondents in Disney’s
confusion survey thought that the allegedly-infringing Disney
shirts were in any way connected to THOIP or its characters.
Therefore, this factor favors Disney as to forward confusion.
The undisputed strength of Disney’s characters and
the Disney name, however, works against Disney on THOIP’s
reverse confusion theory. Because the face of the parties’ shirts
are very similar in all respects except the characters, the fame of
Disney’s characters and its name increases the likelihood that
consumers will mistakenly believe that Disney is the source of or
approved the Little Miss THOIP shirts. Therefore, this factor
favors THOIP as to reverse confusion.106
Indeed, Disney’s shirts – which essentially differ from THOIP’s only in their
substitution of Disney’s well-known (and well-branded) characters for THOIP’s
104
See id.
105
See Arrow Fastener, 59 F.3d at 395-96.
106
THOIP II, 736 F. Supp. 2d at 710-11 (emphasis added).
36
bubble-shaped and dot-eyed “Little Misses” – possess an “origin-indicating”107
quality that very few cartoon-bearing T-shirts could rival. And the Second Circuit
has endorsed the view that in reverse confusion cases “the use of a defendant’s
own name in conjunction with an otherwise similar mark . . . may . . . simply
increase the misappropriation by linking the defendant’s own name to the
plaintiff’s good will established by its trademark.”108
On the other hand, as this Court has recognized, “though public
awareness of THOIP’s name and that of its parent company is low, there is an
appreciable awareness of the Little Miss – and related Mr. Men – characters.”109
Indeed, THOIP has claimed that it has promoted its mark nationwide on books
107
Lang, 949 F.2d at 581 (“‘Ultimately, the strength of the mark turns on
its origin-indicating quality, in the eyes of the purchasing public’”) (quoting
McGregor-Doniger, Inc. v. Drizzle, Inc., 599 F.2d 1126, 1131 (2d Cir. 1979)).
108
Banff, Ltd. v. Federated Dept. Stores, Inc., 638 F. Supp. 652, 656
(S.D.N.Y. 1986) (“Banff I”), remanded to broaden scope of injunction, Banff II,
841 F.2d at 491. Accord A & H Sportswear, 237 F.3d at 230 (“As to the presence
of the house mark on the Victoria’s Secret product, not only is there the possibility
that consumers will fail to remember the mark when encountering A & H’s
swimwear, but there is also the possibility that the mark will aggravate, rather than
mitigate, reverse confusion, by reinforcing the association of the word ‘miracle’
exclusively with Victoria’s Secret.”); Americana Trading Inc. v. Russ Berrie &
Co., 966 F.2d 1284, 1288 (9th Cir. 1992) (“[T]he prominence of Russ’s housemark
may serve to create reverse confusion that Russ, and not [the senior user], is the
source of [the senior user’s] ‘Wedding Bears.’”).
109
THOIP I, 690 F. Supp. 2d at 240 (footnote omitted).
37
since 1981 and on the T-shirts at issue in this case since 2006.110 Nonetheless,
while the strength of THOIP’s mark may weigh against it for the purposes of
assessing the first Polaroid factor in the context of reverse confusion, the
comparatively greater strength of Disney’s mark leads to the conclusion that, on
balance, this factor still weighs in THOIP’s favor – albeit less heavily than I had
previously concluded.
However, notwithstanding the strength of its mark, Disney argues that
THOIP cannot meet its burden on its reverse confusion claim because “it does not
even attempt to articulate how Disney ‘overwhelmed’ THOIP’s Little Miss
brand.”111 In making this argument, Disney marshals an impressive number of
district court cases from this Circuit plainly supporting its position that THOIP
must show – as a “factual predicate” to its claim – that Disney “‘saturated’ the
market with its shirts, ‘overwhelmed’ the [Little Miss THOIP] brand or ‘swamped’
THOIP.”112 Indeed, it is the view of trademark law’s leading expert that a “reverse
110
See THOIP II, 736 F. Supp. 2d at 694-95, 696-97.
111
Disney Opp. Mem. at 2 (emphasis added) (quoting First Nat’l Bank of
Omaha, 2004 WL 1575396, at *12).
112
Id. at 1 (citing O’Keefe, 590 F. Supp. 2d at 520-21 (reverse confusion
“rests on a theory that the defendant has saturated the marketplace with a mark
similar to the plaintiff’s less well known mark to such a degree that consumers will
be misled into believing that the plaintiff is ‘an unauthorized infringer’ of the
defendant’s mark”); Sly Magazine, LLC v. Weider Publ’ns L.L.C., 529 F. Supp. 2d
425, 438 (S.D.N.Y. 2007) (“the essence of plaintiff’s reverse confusion claim is
38
confusion case is proven only if the evidence shows that the junior user was able to
swamp the reputation of the senior user with a relatively larger advertising
campaign.”113 And most other Circuits have endorsed that view.114
that the junior user . . . overpowers the senior user’s mark”); First Nat’l Bank of
Omaha, 2004 WL 1575396, at *12 (“Reverse confusion occurs when a bigger and
better-known ‘junior user saturates the market with a similar trademark and
overwhelms the senior user.’”) (emphasis added) (citation omitted); M & G Elecs.
Sales Corp. v. Sony Kabushiki Kaisha, 250 F. Supp. 2d 91, 106 (E.D.N.Y. 2003)
(“[T]he evidence shows that Sony and its retailers do not aggressively advertise the
Magic Gate technology thereby defeating the reverse confusion theory.”)).
113
McCarthy on Trademarks § 23:10 at 23-66 (citing “[t]he seminal
precedent” on reverse confusion, id. at 23-68, Big O Tire Dealers, Inc. v. Goodyear
Tire & Rubber Co., 408 F. Supp. 1219 (D. Colo. 1976), aff’d and award modified,
561 F.2d 1365 (10th Cir. 1977) (reverse confusion found where Goodyear
promoted its “BIGFOOT” tire in a six million dollar nationwide marketing
campaign, swamping a small, regional tire company’s “BIGFOOT” tire mark)).
114
See Visible Sys. Corp. v. Unisys Corp., 551 F.3d 65, 72 (1st Cir. 2008)
(“‘A reverse confusion case is proven only if the evidence shows that the junior
user was able to swamp the reputation of the senior user with a relatively much
larger advertising campaign.’”) (quoting McCarthy on Trademarks § 23:10). See
also Therma-Scan, Inc. v. Thermoscan, Inc., 295 F.3d 623, 630 (6th Cir. 2002)
(“Reverse confusion occurs where ‘the junior user saturates the market with a
similar trademark and overwhelms the senior user. The public comes to assume
the senior user’s products are really the junior user’s or that the former has become
somehow connected to the latter. The result is that the senior user loses the value
of the trademark – its product identity, corporate identity, control over its goodwill
and reputation, and ability to move into new markets.’”) (quoting Ameritech, Inc.
v. American Info. Techs., Corp., 811 F.2d 960, 964 (6th Cir. 1987)); MicroStrategy
Inc. v. Motorola, Inc., 245 F.3d 335, 346 -347 (4th Cir. 2001) (“‘[R]everse
confusion occurs when the junior user saturates the market with a similar
trademark and overwhelms the senior user’”) (quoting A & H Sportswear, 237
F.3d at 228); Johnny Blastoff, Inc. v. Los Angeles Rams Football Co., 188 F.3d
427, 436 (7th Cir. 1999) (same); Minnesota Pet Breeders, Inc. v. Schell &
Kampeter, Inc., 41 F.3d 1242, 1246 (8th Cir. 1994) (“‘Reverse confusion occurs
39
Meanwhile, THOIP maintains that, at least in the Second Circuit,
“[t]he test for reverse confusion is likelihood of confusion as determined under the
Polaroid factors, analyzed in the reverse confusion context; there are no special
requirements to [sic] reverse confusion,” citing a Second Circuit case from 1983.115
Relying on the Second Circuit’s 1988 Banff II decision, THOIP urges that “it is
simply not a requirement that, for a finding of reverse confusion, the junior user
must have already swamped the senior user’s mark; the test for reverse confusion
is likelihood of confusion.”116 Indeed, in Banff – where reverse confusion was
“perhaps the primary type of confusion involved”117 – the Second Circuit found a
sufficient likelihood of confusion between Bloomingdale’s junior “B-Wear” mark
and Banff’s senior “Bee Wear” mark to affirm the issuance of a preliminary
injunction against Bloomingdales, where Banff
demonstrated the likelihood of confusion due to Bloomingdale’s
use of a similar mark on women’s apparel. In this infringement
case, threat of irreparable harm due to consumer confusion is
particularly acute due to the identity of the products on which the
when a large junior user saturates the market with a trademark similar or identical
to that of a smaller, senior user.’”) (quoting Sands, Taylor & Wood Co. v. Quaker
Oats Co., 978 F.2d 947, 957 (7th Cir. 1992)).
115
THOIP Opp. Mem. at 1 (citing Plus Prods., 722 F.2d at 1003-04).
116
Id. at 2.
117
841 F.2d at 491.
40
marks are being employed . . . and the great ability of
Bloomingdale’s to dilute the plaintiff’s mark through its strong
market position. . . . Banff competes with Bloomingdale’s in each
of its locations by selling women’s apparel to other department
stores.118
Moreover, Banff (the senior user) had been marketing a successful line of apparel
under the unregistered trademark “Bee Wear” to retail stores nationwide for over
fifteen years, with aggregate net sales exceeding $75 million.119
Having carefully reviewed the parties’ arguments and the case law on
which they are based, I conclude that a correct statement of the law lies somewhere
between the legal positions they advance. Although aggressive advertising efforts
by junior users or their ability to “saturate” the market with their mark are
undeniably important factors for courts to consider in weighing the relative
strength of the junior and senior marks (for the purposes of analyzing the first
Polaroid factor), they are not “factual predicates” to a reverse confusion claim.120
118
638 F. Supp. at 658 (emphasis added).
119
See id. at 653.
120
Accord A & H Sportswear, 237 F.3d at 236 (“[A]lthough economic
disparity between the companies and the marks is an important consideration in the
evaluation of the marks’ commercial strength, the District Court legally erred in
fashioning a threshold ‘economic disparity’ requirement before a reverse confusion
claim will even be considered.”) (upholding the viability of reverse confusion
where the senior user had ten percent of market and had spent over one million
dollars in advertising); id. at 235-36 (“[A] company need not be all that weak
within its market in order to bring a viable reverse confusion claim.”); Fisons
Horticulture, Inc. v. Vigoro Indus., Inc., 30 F.3d 466, 479 (3d Cir. 1994) (finding
41
Thus, I will assess the parties’ arguments and evidence showing why Disney is or
is not likely to “overwhelm” THOIP’s Little Miss shirts if its Miss Disney and
Little Miss Disney shirts return to the market; but my conclusion will not be
determinative as to a finding of reverse confusion. Rather, it will influence the
weight I place on the first factor – the comparative strength of Disney’s mark.121
THOIP argues that the mere “strength of the Disney brand and
characters” is sufficient to support a finding that Disney will overwhelm THOIP’s
Little Miss brand:
In the absence of injunctive relief, the overwhelming strength of
the Disney brand will overwhelm and undercut the value of the
THOIP mark. THOIP faces one of the strongest brands in the
world, in circumstances where Disney’s shirts are “strikingly
similar in design” to those of THOIP and where Disney “almost
viable reverse confusion claim where senior user held twenty-five percent of
market and had been using mark for thirty years); Fuji Photo Film v. Shinohara
Shoji Kabushiki, 754 F.2d 591(5th Cir. 1985) (finding a viable reverse confusion
claim where the plaintiff had spent millions of dollars). But see First Nat’l Bank of
Omaha, 2004 WL 1575396, at *13 (“As an initial matter, [plaintiff] has failed to
establish the relevance of an analysis of the likelihood of reverse confusion
between the SMARTONE and ONESMART marks among consumers” because
“[plaintiff] has presented no evidence that [defendant] MasterCard is likely to
saturate the consumer market with the ONESMART mark.”); id. (“There is no
basis to find that the ONESMART mark is likely to overwhelm FNBO’s existing
SMARTONE mark, as is required to form the factual predicate on which a claim
for reverse confusion is based.”).
121
See M2 Software, Inc. v. Madacy Entm’t, 421 F.3d 1073, 1089 (9th
Cir. 2005) (“[I]n reverse confusion cases, the inquiry focuses on the strength of a
junior mark. The strength of the junior mark is closely related to how much the
market is saturated by the junior mark.”).
42
undoubtedly intended to copy THOIP’s shirts.” . . . [T]he strength
of the Disney brand and characters overwhelms even strong
brands, not the advertising budget of an individual product. It is
the strength of the Disney characters that makes the consumer
think that the “Miss” and “Little Miss” formats are Disney
formats, thereby capturing THOIP’s mark.122
Moreover, as I have already noted repeatedly, “the Little Miss THOIP, Little Miss
Disney, and Miss Disney lines are within the same narrow category of goods
directed at the same set of consumers.”123 Thus, although reverse confusion may
not be likely if “the goods or services of the parties are not competitive, the
respective markets are separated and the advertising of the parties is directed at
different types of purchasers,”124 that is not this case.
However, as Disney notes, THOIP has proffered no evidence that
Disney ever engaged in – or plans to engage in – any advertising or promotion with
respect to the Miss Disney and Little Miss Disney shirts.125 Meanwhile, as noted
122
THOIP Opp. Mem. at 4-5 (citations omitted) (quoting THOIP II, 736
F. Supp. 2d at 39, 48).
123
THOIP I, 690 F. Supp. 2d at 233.
124
McCarthy on Trademarks § 23:10 at 23-74.
125
See Disney Opp. Mem. at 1. Of course, if Disney ever decided to reissue the shirts and advertised, promoted, and sold them in a different manner in
the future than it had in the past, a court could then consider those changed
circumstances at that time. Finding reverse confusion now based on what THOIP
posits Disney might do, rather than what the admissible evidence shows Disney
has done or intends to do, would constitute an improper advisory opinion.
43
above, THOIP has claimed that it has promoted its mark nationwide on books since
1981 and on the T-shirts at issue in this case since 2006.126 While THOIP has
stated that licensees and retailers bear the bulk of the expenses relating to the
advertising of the Little Miss brand, THOIP contends that within approximately
four years of acquiring the brand, it had spent “over $1.7 million marketing and
promoting the MR. MEN and LITTLE MISS brands to U.S. licensees, consumers
and retailers.”127 THOIP claims that it also promoted the Little Miss brand through
THOIP’s television show, “The Mr. Men Show,” which launched in 2008.128
Indeed, from 2006 through 2008, clothing featuring the Little Miss family of
characters alone generated over ten million dollars in wholesale sales.129
Moreover, although the Disney and THOIP shirts clearly target the
same set of customers, it is undisputed that the Little Miss Disney shirts –
undoubtedly the most susceptible to reverse confusion – were only sold in the
Disney theme parks and the World of Disney store in New York,130 which has
126
See THOIP II, 736 F. Supp. 2d at 694-95, 696-97.
127
THOIP’s Statement of Undisputed Facts Pursuant to Local Rule 56.1
dated March 22, 2010 [Docket No. 103] ¶ 156.
128
See id. ¶¶ 138, 139.
129
See id. ¶¶ 199, 201, 203.
130
See THOIP I, 690 F. Supp. 2d at 238.
44
since closed.131 And there is nothing to indicate that the Miss Disney shirts and the
Little Miss THOIP shirts ever appeared in the same stores at the same time.132 In
the absence of any advertising or promotional efforts on Disney’s part, then, the
Disney shirts’ ability to overwhelm THOIP’s mark would derive entirely from
customers’ encountering Disney’s shirts in stores, and then separately encountering
the Little Miss THOIP shirts in proximate stores. Yet THOIP has not shown that it
is “more likely that consumers will first encounter the junior user’s – rather than
the senior user’s – goods.”133
Finally, because “the market for T-shirts is ‘fragmented’ with
thousands of T-shirts put out by numerous manufacturers, retailers and other
organizations,” it is “extremely difficult for any particular T-shirt to overwhelm
another brand.”134 Along similar lines, the “low involvement” nature of T-shirt
buying further reduces “the likelihood that a T-shirt could have a swamping
131
See Disney Opp. Mem. at 6.
132
See THOIP I, 690 F. Supp. 2d at 237.
133
First Nat’l Bank of Omaha, 2004 WL 1575396, at *12.
134
Disney Mem. at 3 (citing and purportedly quoting Transcript of
Deposition of Dr. Yoram Jerry Wind (“Wind Dep.”), Ex. 3 to Simonson Decl., at
57:2-58:10). Disney in fact quoted Dr. Simonson characterizing Dr. Wind. See
Deposition of Dr. Itmar Simonson (“Simonson Dep.”), Ex. 3 to Simonson Decl., at
76:3-7.
45
effect.”135 Balancing all of these sub-factors, I conclude that – at least on the
record before this Court – Disney’s ability “to amass such trademark strength in its
imitative mark that [THOIP’s] products become associated with [Disney] in the
minds of consumers” is relatively low.136
b.
Actual Confusion
As I explained in THOIP II,
This factor examines whether the parties’ products have actually
confused consumers in the marketplace.[137] While “actual
confusion need not be shown to prevail,”[138] evidence of actual
confusion is “highly probative” of the likelihood of confusion.[139]
Proof of actual confusion is generally shown through consumer
surveys or anecdotal evidence of confusion.
Though THOIP’s shirts were available on the market
with the Little Miss Disney and Miss Disney shirts for eight
months and ten months respectively, THOIP has not presented
evidence of actual confusion.140
As noted above, the Ostberg Survey – which returned reverse confusion rates of 8
percent for the Miss Disney shirts and 19 percent for the Little Miss Disney shirts –
135
Id. (citing Wind Dep. at 57:2-12).
136
See Star Indus., 412 F.3d at 388 n. 3.
137
See Centaur Commc’ns, Ltd. v. A/S/M Commc’ns, Inc., 830 F.2d
1217, 1227 (2d Cir. 1987).
138
Lois Sportswear, U.S.A., Inc. v. Levi Strauss & Co., 799 F.2d 867, 895
(2d Cir. 1986).
139
Sunenblick, 895 F. Supp. at 630.
140
THOIP II, 736 F. Supp. 2d at 712 (one footnote omitted).
46
is not a reliable indicator of consumer confusion, and is therefore excluded. Thus,
THOIP presents no (admissible) evidence of reverse confusion.141 In contrast,
Disney’s Helfgott Survey showed a reverse confusion rate of zero percent. THOIP
does not ask the Court to exclude the Helfgott Survey, but argues it has no
probative value because “[w]hether Disney’s adoption or re-adoption of THOIP’s
‘trend’ using its own famous characters did or would cause the consuming public
to associate THOIP’s T-shirts with Disney, . . . this could not be tested without
participants knowing of Disney’s shirts,”142 which had been off the market for over
two years. I agree. While an Eveready survey may have been appropriate to
measure forward confusion – at a time when the Little Miss THOIP shirts were still
on the market – the Helfgott Survey, too, fails to replicate actual (past or future)
marketplace conditions, rendering its results largely irrelevant.143
141
I find it telling, however, that despite the fact that the Ostberg
Survey’s flaws likely had a severely inflating effect on the confusion rates
reported, it still returned a reverse confusion rate of only 8 percent with respect to
the Miss Disney shirts.
142
THOIP’s Memorandum of Law in Support of Motion for Partial
Summary Judgment (“THOIP Mem.”) at 11.
143
Disney argues that “[i]f swamping had occurred so as to give rise to
reverse confusion [in 2008], the Disney shirts would have had such an impact that
consumers would be able to recall the shirts two years later.” Disney Mem. at 8. I
do not agree.
47
Thus, the court stands in essentially the same position it did a year
ago, with neither party proffering (probative, reliable) evidence of reverse
confusion. Because the burden in this case lies with THOIP, I therefore conclude
that the Polaroid “actual confusion” factor weighs in Disney’s favor.
2.
Balancing the Factors
The (1) competitive proximity and (2) similar quality of the shirts at
issue in this case and the (3) strength of Disney’s mark weigh slightly in THOIP’s
favor on the issue of reverse confusion. Had THOIP submitted strong (probative
and reliable) survey evidence of reverse confusion, I would conclude that a
reasonable juror could find reverse confusion. However, in light of (4) the lack of
(probative and reliable) survey evidence suggestive of actual reverse confusion, (5)
Disney’s lack of bad faith, and (6) the lack of (non-hypothetical) evidence that the
Miss and Little Miss Disney shirts are likely to overwhelm THOIP’s mark, I
conclude that no reasonable juror could find reverse confusion. The clerk of the
court is directed to close these motions [Docket Nos. 138 and 141] and this case.
48
Dated:
New York, New York
May 10,2011
49
- Appearances For Plaintiff:
Abraham Y. Skoff, Esq.
David Rabinowitz, Esq.
Martin Schwimmer, Esq.
Jordan Daniel Greenberger, Esq.
Moses & Singer LLP
405 Lexington Avenue
New York, New York 10174
(212) 554-7800
For Defendants:
Dale M. Cendali, Esq.
Melanie Bradley, Esq.
Courtney Schneider, Esq.
Kirkland & Ellis LLP
601 Lexington Avenue
New York, New York 10022
(212) 446-4800
50
Disclaimer: Justia Dockets & Filings provides public litigation records from the federal appellate and district courts. These filings and docket sheets should not be considered findings of fact or liability, nor do they necessarily reflect the view of Justia.
Why Is My Information Online?