By Paul Opitz
The Third Chamber of the General Court of the European Union (EGC) ruled on 5 December 2017 that the Chinese smartphone maker Xiaomi, Inc., may not register the EU word mark MI PAD for its tablet computers, since it is likely to be confused with Apple’s iPad. (Xiaomi, Inc., v. European Union Intellectual Property Office, Case T-893/16)
In April 2014, Xiaomi, Inc., (Xiaomi) filed an application for registration of an EU trade mark with the European Union Intellectual Property Office (EUIPO) to register the word sign MI PAD. Registration was sought for Classes 9 and 38 of the Nice Agreement concerning the International Classification of Goods and Services, which correspond to the descriptions of inter alia portable and handheld electronic devices and telecommunication access services.
In August 2014, Apple Inc., (Apple) filed a notice of opposition to registration of the mark in respect of all the goods and services in the applied classes. The opposition was based on Apple´s earlier EU word mark IPAD, which was filed in January 2010 and registered in April 2013, covering goods and services in the same classes. The relative grounds relied on in the opposition were those of identity with, or similarity to an earlier trademark, currently set out in Article 8 (1) (b) of Regulation 2017/1001. This opposition was upheld by the Opposition Division in December 2015, which rejected Xiaomi´s application.
Thereafter, Xiaomi filed an appeal with EUIPO against the Opposition Division´s decision, which was again dismissed in September 2016 on the grounds that the marks MI PAD and IPAD were highly visually and phonetically similar and could lead to a confusion of the relevant public. This decision by the EUIPO was now contested by Xiaomi.
Decision of the General Court
First, the Court established some background on the scope of decisions concerning the relative ground of similarity. According to settled case law, the risk that the public may believe that goods come from the same undertaking or economically-linked undertakings constitutes a likelihood of confusion. Also, this likelihood must be assessed globally and taking into account all factors relevant to the case (Laboratorios RTB v OHIM – Giorgio Beverly Hills, Case T-162/01). For the application of Article 8 (1) (b) of Regulation 2017/1001, a likelihood of confusion presupposes both that the marks are identical or similar and that the goods which they cover are identical or similar (Commercy v OHIM – easyGroup IP Licensing, Case T-316/07).
The relevant public
The Court referred to the decision of the Board of Appeal and emphasized that the goods in question are aimed at both the general public and professional consumers with specific knowledge. Regarding the relevant public´s level of attention, the Court elaborates that although the purchase price of some goods covered by the mark are relatively high, most electronics aimed at the general public are, nowadays, relatively inexpensive and have short lifespans. Therefore, they do not require any particular technical knowledge and leave the level of attention between average and high. Secondly, the Court upheld the Board of Appeal’s finding that the relevant territory is the European Union as a whole.
Comparison of the signs
At first, the Court notes that a global assessment of the likelihood of confusion must be based on the overall impression of the signs, including the visual, phonetic, and conceptual similarity. In the case at issue, the comparison of the marks must be carried out by considering each mark as a whole, since there are no dominant elements. The Court holds that the marks are visually highly similar, since the earlier trade mark IPAD is entirely reproduced in the mark MI PAD. Moreover, they coincide as to the letter sequence “ipad” and differ only as to the presence of the letter “m” at the beginning. Phonetically, the marks are also highly similar, referring to the pronunciation of their common syllable “pad” and of the vowel “I”. The latter will be likely be pronounced as the first person singular possessive pronoun “my” in English and thereby similar to the “I” in Apple’s iPad. The Court clarifies that even minor differences in pronunciation due to the letter “m” are not capable to offset the overall similarities. Conceptually, the English-speaking part of the EU understands the common element “pad” as a tablet or tablet computer, which makes it only weakly distinctive and sufficient for a finding of similarity (Xentral v OHIM – Pages jaunes, Case T-134/06).
The likelihood of confusion
For determining the likelihood of confusion, the interdependences between the similarity of the marks and that of the goods covered must be examined. The court states that the visual and phonetic differences resulting from the presence of the additional letter “m” are not able to rule out a likelihood of confusion as a result of the overall similarities. Neither are the conceptual differences resulting from the prefixes “mi” and “I” sufficient to remove this likelihood created by the common element “pad”. Taking into account that the goods in question are identical, the conceptual similarities overweigh the discrepancies.
In conclusion, the Court could not exclude the possibility that the public might believe that both tablets come from the same undertaking or economically-linked undertakings. Hence, the Court rejected and dismissed the applicant’s plea in law.
By Gabriel M. Lentner
The U.S.-based Bridgestone Licensing Services, Inc. and Bridgestone Americas, Inc. lodged a claim against Panama over trademarks at the International Centre for Settlement of Investment Disputes (ICSID).
The claim relates to a decision rendered by the Supreme Court of Panama concerning Bridgestone’s trademarks in Panama and is based on the Panama-US Trade Promotion Agreement (TPA). The arbitral tribunal is currently dealing with “Expedited Objections”.
A key issue in this dispute is whether the ownership of the FIRESTONE trademark and rights to sell, market and distribute BRIDGESTONE and FIRESTONE branded products in Panama constitute “investments” under Art 10.29 of the TPA, as argued by the claimants. Under this provision the term “investment” is defined as “means every asset that an investor owns or controls, directly or indirectly, that has the characteristics of an investment, including such characteristics as the commitment of capital or other resources, the expectation of gain or profit, or the assumption of risk. Forms that an investment may take include: … (f) intellectual property rights; (g) licenses, … and similar rights conferred pursuant to domestic law” In a footnote it is clarified that “Among the licenses, authorizations, permits, and similar instruments that do not have the characteristics of an investment are those that do not create any rights protected under domestic law.”
Bridgestone argues inter alia that its licenses are to be considered intellectual property rights and therefore covered investments. In addition, they contend that these licenses create rights protected under Panamanian law, since they concern trademarks registered in Panama.
Panama on the other hand challenges these arguments stating that Bridgestone does not have an “investment” within the meaning of the ICSID Convention (Art 25) and the TPA. Rather, Panama views the activities of Bridgestone as ordinary commercial transactions outside the scope of investment arbitration. More specifically responding to the Claimant’s argument, Panama disputes that the three licenses at issue do have the characteristics of an investment as they do not create any rights protected under Panamanian law.
Still pending, this case as it adds to the growing number of international investment disputes involving intellectual property rights (see cases of Philip Morris v Australia and Philip Morris v Uruguay, Eli Lilly v Canada). There is still a lot of uncertainty in this area of law and hence it will be interesting to see the final outcome and the reasoning of the tribunal dealing with the issue of investment and IP.
By Marie-Andrée Weiss
The United States Court of Appeals for the Ninth Circuit ruled on 16 May 2016 that ‘google’ is not a generic term for a search engine, and thus the famous California company did not suffer the costly indignity of having its trademarks cancelled through genericide. The case is Elliott v. Google, 2:12-cv-01072.
Plaintiffs had registered 763 domain names, each incorporating the word ‘google’ along with the name of a another brand (googledisney.com), of a person (googlebarackobama.net) or a place (googlemexicocity.com). This business plan did not fare well with the famous search engine company, which successfully asked the National Arbitration Forum to transfer all these domain names to Google.
Plaintiffs then filed a suit in the United States District of Arizona claiming that ‘Google’ “is, or has become, a generic term universally used to describe the action of internet searching with any search engines” (Complaint, p. 2), and asked the court to cancel Google’s trademarks. Indeed, the Lanham Act, 15 U.S.C. § 1064(3), provides the right to petition for the cancellation of mark if it “becomes the generic name for the goods or services, or a portion thereof, for which it is registered.”
The parties filed cross-motions for summary judgment on the issue of whether the GOOGLE marks were generic: while Plaintiffs claimed that Google is a generic term because a majority of the public use it as a verb, Google argued that use of a trademark as a verb use is not automatically generic use. On 11 September 2014, the United States District of Arizona granted summary judgment for Google. Plaintiffs appealed to the United States Court of Appeals for the Ninth Circuit, which affirmed.
The Google trademarks
As mentioned in the original complaint, ‘Google’ comes from the term ‘googol,’ meaning a 1 followed by 100 zeros. Google holds a trademark registration for GOOGLE in class 9 for “computer hardware; computer software for creating indexes of information, indexes of web sites and indexes of other information resources” and another one in class 38 for “[p]roviding electronic mail and workgroup communications services over computer networks; providing multiple user access to proprietary collections of information by means of global computer information networks.”
Generic trademark and genericide
Needless to say, if a mark becomes generic, it is quite costly for the company that invested a lot in developing goodwill towards its brand. A generic term cannot serve as a trademark because it cannot serve as identifying the source of a product or service. Several famous marks, among them aspirin, cellophane, and thermos, fell victim of their success and became generic because they were used by the general public to designate the genus of their product, not just a particular brand. This is ‘genericide’.
The primary significance test
Plaintiffs had the burden of proving the genericide since they applied for the cancellation of the GOOGLE trademarks, and a registered trademark is presumed to be valid. They argued on appeal that the district court had misapplied the primary significance test, which was coined by the Supreme Court in its 1938 Kellog Co. v. National Biscuit Co. case: a mark is not generic if “the primary significance of the term in the minds of the consuming public is not the product but the producer.” As noted by the Ninth Circuit, quoting Ty Inc. v. Sofbelly’s Inc., “a trademark only becomes generic when the “primary significance of the registered mark to the relevant public” is as the name for a particular type of good or service irrespective of its source.”
Plaintiffs argued that the district court had framed the inquiry as to whether the primary significance of ‘google’ to the consuming public is a generic name for search engines, whereas it should have inquired whether the public primarily uses ‘google’ as a verb. The Ninth Circuit disagreed with this argument for two reasons: genericide always relates to a particular good or service and using a trademark as a verb is not automatically generic use.
Genericide always relates to a particular good or service
For the Ninth Circuit, the District Court “properly recognized the necessary and inherent link between a claim of genericide and a particular good or service” (p. 9). The Court reasoned that failing to consider this would prevent some arbitrary marks to be protectable, giving as example IVORY which is arbitrary as applied to soaps, but would not be so for product made from the tusks of elephants.
The Ninth Circuit found that Plaintiffs’ “evidence was “’largely inapposite to the relevant inquiry under the primary significance test because [the Plaintiffs] ignor[e] the fact that a claim of genericide must relate to a particular type of goods or service’” (p. 13).
Using a trademark as a verb is not automatically generic use
Also, “verb use does not automatically constitute generic use” (p. 10). Plaintiffs had argued that a word can only be used as a trademark if it is used as an adjective. The Ninth Circuit disagreed, noting that it had found in Coca-Cola Co. v. Overland, Inc. that the mere fact that customers ordered “a coke” did not prove what they were thinking, a mark or a cola beverage, and more evidence was required about the customer’s inner thought process. Therefore, the use of a trademark as a noun may or may not be using it as a trademark (p. 11).
The primary significance test directed plaintiffs to provide evidence that that the primary significance of the GOOGLE trademarks is a general name for search engines, not a trademark identifying a particular search engine. The Ninth Circuit agreed with the district court which had found that, while the verb ‘google’ is indeed used to refer to searching on the internet, regardless of the search engine used, this fact alone cannot support a jury finding of genericide under the primary significance test, as it does not prove “how the public primarily understands the word itself, irrespective of its grammatical function, with regard to internet search engines” (p. 14).
How to prove that a mark has become generic
Plaintiffs also argued on appeal that the district court impermissibly weighted the evidence presented by Plaintiffs when granting summary judgment to Google. The Ninth Circuit disagreed, because, while Plaintiffs’ had presented admissible evidence that the majority of the public used ‘google’ as a verb, this was not enough to survive summary judgment, as it cannot alone prove genericide.
Plaintiffs had presented three surveys as evidence. Two were excluded by the district court because they had been conducted by Plaintiffs’ counsel, and “a valid survey design typically requires graduate training or professional experience in survey research” (p. 15). The third survey was a “Thermos survey,” that is a survey using open-ended questions, in our case, asking respondents how they would ask a friend to search something on the internet. The majority answered by using ‘google’ as a verb, and the survey was admitted as evidence that a majority of the public uses google as a verb meaning searching the internet.
Plaintiffs also gave examples of alleged generic use of ‘google’ by media and consumers, but they failed to convince both the district court and the Ninth Circuit, because Plaintiffs did not provide evidence that the use was indeed generic in the mind of the media and the consumers.
Plaintiffs had also offered expert testimony by three experts who all were of the opinion that ‘google’ is generic when used as a verb. However, this finding alone is not enough to prove genericide. Plaintiffs’ dictionary evidence did not prove either that ‘google’ is a generic name for internet search engines, only proving it is generic when used as a verb.
Plaintiffs also tried to prove that Google itself was using ‘google’ in a generic sense, presenting as evidence an email from Google cofounder Larry Page encouraging its recipients to “keep googling!” Generic use of a mark by its holder can support a finding a genericide, but the email was found by the court to be yet another example of the use of ‘google’ as a verb and did not prove that Larry Page had a particular search engine in mind (p. 19).
Finally, Plaintiff claimed that there was no efficient alternative for ‘google’ as a name for the act of searching the internet, but the Ninth Circuit drily noted that Google’s competitors do not use ‘google’ to refer to their own services (p.20).
‘Google’ may have become a verb, but this alone does not prove that GOOGLE is a generic mark. Keep googling.
By Marie-Andrée Weiss
Adidas owns multiple trademark registrations in the European Union and the U.S. for its famous three stripe design, and it fiercely protects them. It has filed, and won, several trademark infringement suits, and regularly sends cease-and-desist letters asking brands to stop selling shoes or clothes bearing stripes.
In February 2017, Adidas filed a notice of opposition with the U.S. Patent and Trademark Office Trademark Trial and Appeal Board (TTAB) to the registration of a mark that Tesla Motors was seeking to register for articles of clothing. The mark would have consisted of “three equal length horizontal stylized lines in the manner of a stylized number 3.” The trademark has since been abandoned after an inter-partes decision by the TTAB.
On 17 February 2017, Adidas also filed a trademark infringement and dilution suit against competitor Puma North America Inc. in the district court of Oregon. Adidas claimed that Puma’s new model of soccer cleats, which bear four diagonal stripes on each side, infringes on the Adidas trademark as it is likely to cause consumer confusion as to the source of the footwear. Adidas voluntarily dismissed the case on 28 February 2017, likely following successful negotiations with Puma.
On 14 February 2017, the Barcelona Football Club abandoned its application to register a mark in class 28, for sporting articles, following a notice of opposition filed by Adidas on 31 October 2016, and an inter-partes decision by the TTAB. The abandoned mark consisted of “a square containing seven vertical stripes. The 1st, 3rd, 5th and 7th stripes from the left are blue, and the remaining three stripes are garnet.”
On 17 March 2017, Adidas filed a trademark infringement and dilution suit in the Eastern District of North Carolina, against fashion company Juicy Couture, which came to fame some 15 years ago for creating a velour tracksuit. Adidas claimed that some jackets and pants, bearing stripes on their sleeves and sides, infringe several of its trademarks.
Adidas has won or settled all of the trademark infringement cases it has filed. Will the streak ever end?
The scope of the three-stripe trademark
What exactly do the Adidas trademarks protect? Are all three stripes claimed by Adidas under the trademark? Are all stripes on shoes and clothing, regardless of the number of stripes, claimed by Adidas?
Adidas owns several federal trademark registrations in the U.S. for a mark consisting “of three parallel stripes applied to footwear, the stripes are positioned on the footwear upper in the area between the laces and the sole,” (see here, here, or here). Adidas also owns trademarks for clothing bearing the three stripes (see here) and even for verbal trademarks using the term “3 stripes,” such as the trademark “THE BRAND WITH THE 3 STRIPES.” Does that mean that Adidas has a monopoly for just about every trademark featuring three stripes, every trademark featuring two or four stripes, or even for clothing featuring any number of stripes?
The February 2017 complaint against Puma stated that Adidas has been using the three-stripe trademark on shoes since 1952 and on apparel since 1967. While easily recognizable, Adidas’s three-stripe trademark is also simple: three stripes, often shown diagonally on the sides of shoes, on the sleeves of a training jacket, or the sides of training pants, shorts, or shirts. The three stripes are all of the same width when seen together, but this width varies from trademark to trademark. The distance between each stripe also varies.
In the USPTO Design Search Code Manual, category 26 is for “geometric figures and solids.” 26.17 is for “lines, bands, bars, chevrons and angles” and 26.17.01 is for “straight line(s), band(s) or bar(s).” 26.17.05 is the code for “horizontal line(s), band(s) or bar(s).”
The design search codes for the trademark which Tesla sought to register were 26.17.01 and 26.17. A recent search in the TESS database for a mark with a 26.17. 01 code yielded 89,266 records and a search for marks with the 26.17.05 code yielded 81,820 records. Amongst the 26.17.05 results, 14 were filed by Adidas.
The mark which Tesla sought to register was described in the application as consisting of “three equal length horizontal stylized lines in the manner of a stylized number 3.” Yet the stripes were not similar to Adidas stripes, which are cut in a neat angle. Tesla’s stripes were cut on the side in a soft curve, resembling a Japanese wood beam or roof. The Barcelona Football Club was trying to register as a trademark the stripes which are seen on its own logo, which is itself a registered trademark! Indeed, many sports teams around the world sport stripes on their uniforms. A stripe is a stripe is a stripe. Yet Adidas opposed these two trademark registrations.
Is Adidas going too far?
This is not the first time that Adidas sued a company over the use of stripes on shoes or clothing, even if more or less than three stripes are featured. Adidas sued several European retailers in the late nineties over the use of two stripes on the side of sports clothes, which eventually led to the European Court of Justice ruling in 2008, in Adidas AG and Others v. Marca Mode CV and Others, that Adidas’ competitors could not “be authorized to infringe the three-stripe logo registered by Adidas by placing on the sports and leisure garments marketed by them stripe motifs which are so similar to that registered by Adidas that there is a likelihood of confusion in the mind of the public” (at 32).
While there may be a need for signs which do not have a distinctive character, such as stripes, to be available for competitors, this need “cannot be taken into account in the assessment of the scope of the exclusive rights of the proprietor of a trade mark” (ruling of the Court). The European Court of Justice thus chose to protect the public against any likelihood of confusion.
U.S. fashion manufacturers also encounter legal difficulties when using stripes on garments, and their frustration is mounting. On 3 March 2017, fashion retailer and manufacturer Forever 21 filed a complaint against Adidas, asking the Central District Court of California for a declaratory judgment of non-infringement of trademark. Forever 21 claims that Adidas is now “essentially asserting that no item of clothing can have any number of stripes in any location without infringing Adidas trademarks.” Forever 21 is “[t]ired of operating with a cloud over its head with regard to its right to design and sell clothing items bearing ornamental/decorative stripes” and “has decided that enough is enough… This matter is ripe for a declaratory judgment.” However, Forever 21 voluntarily dismissed the case on 13 March 2017.
Stripes are never out of fashion, and fashion designers frequently use them on the side of pants or jackets. Is this infringement? Forever 21 had claimed that “Adidas should not be allowed to claim that Adidas, alone, has a monopoly on striped clothing.” The retailer filed the suit after receiving yet another cease and desist letter sent by Adidas, this time asking Forever 21 to stop selling clothes bearing four stripes, including a sports bra, tee shirts and pants. Forever 21 claimed that “[a]ny use of stripes on clothing sold by Forever 21 is ornamental, decorative, and aesthetically functional.”
Adidas had sent a similar letter to Forever 21 in June 2015, which claimed that a sweat shirt featuring Snoopy, with stripes on its cuffs, bottom and collar, was infringing. However, varsity jackets, or letterman jackets, traditionally sport stripes in similar places, and Forever 21 indeed described its Snoopy shirt as featuring “generic varsity-style stripe pattern.” Is Adidas too aggressive in enforcing its mark?
A need to police the mark
These cease and desist letters illustrate what trademark owner must do to avoid losing their rights through failure to control use. Section 45 of the Trademark Act states that a mark is abandoned when “any course of conduct of the owner, including acts of omission as well as commission, causes the mark to… lose its significance as a mark.” This includes failing to adequately police the mark against third-party use. Also, the three-stripe mark is famous, thus making trademark dilution another concern for Adidas. In fact, even just the appearance of dilution is a concern, since trademark owners only need to prove a likelihood of dilution, not actual dilution, after the enactment of the Trademark Dilution Revision Act of 2006. Adidas does not want its three stripes to strike out. But is it the general public which ends up losing?
By Marie-Andrée Weiss
The First Chamber of the Court of Justice of the European Union (CJEU) held on 10 November 2016, that the famous Rubik’s cube cannot be registered as a three-dimensional mark because its shape performs the technical function of the goods, a three-dimensional puzzle. The case is Simba Toys GmbH & Co. KG v. EUIPO, C-30/15 P.
The validity of the Rubik’s cube trade mark was challenged by a competitor
British company Seven Towns Ltd., acting on behalf of Rubik’s Brand Ltd., filed in April 1996 an application for registration of a Community trade mark at the Office for Harmonisation in the Internal Market (OHIM), now named the European Union Intellectual Property Office (EUIPO), for a three-dimensional sign, the famous Rubik’s cube. The mark was registered in April 1999 and renewed in November 2006.
A few days later, competitor Simba Toys applied to have the trade mark declared invalid under Council Regulation 40/94, which has been repealed and was replaced by Council Regulation 207/2009. The CJEU considered the case to be still governed by Council Regulation 40/94. Articles 1 to 36 are the same in both Regulations, and so the case is relevant under current EU trademark law.
Article 7(1)(e)(ii) of Council Regulation 40/94 prevents registration as a trademark of a sign, such as the Rubik’s cube product, “which consists exclusively of… the shape of goods which is necessary to obtain a technical result.” The CJEU had held in the 2002 Koninklijke Philips Electronics NV v. Remington Consumer Products Ltd. that a sign consisting exclusively of the shape of a product cannot be registered as a trademark if the essential functional features of the shape are attributable only to a technical result (Philips § 79 and § 80).
Simba Toys argued that the Rubik’s cube mark should be declared invalid under the grounds [absolute ground for refusal] that the mark is the shape of the goods necessary to achieve a technical result. According to Simba, the Rubik’s cube black lines are attributable to technical functions of the three-dimensional puzzle.
The OHIM dismissed Simba’s application for a declaration of invalidity and the Second Board of Appeal of OHIM affirmed the dismissal in September 2009, reasoning that the shape of the trade mark does not result from the nature of the Rubik’s cube itself. On 25 November 2014, the General Court dismissed the action for annulment as unfounded. Simba appealed to the CJEU.
What are the essential characteristics of the Rubik’s cube trade mark?
The essential characteristics of three-dimensional signs are the most important elements of the signs, Lego Juris v. OHIM, C-48/09, § 68 and 69. They must be properly identified by the competent trademark registration authority, Lego Juris v. Ohim § 68, which must then determine whether the essential characteristics all perform the technical function of the goods (General Court § 41). The General Court identified the essential characteristics of the Rubik’s cube trademark is a “cubic grid structure,” that is the cube itself and the grid structure appearing on each of its surfaces (General Court § 45). Simba did not challenge this finding on appeal at the CJEU.
Do the essential characteristics of Rubik’s cube perform the technical function of the goods?
Under Article 4 of Regulation 40/94 and Regulation 207/2009, any sign capable of being represented graphically can be a trade mark, unless, under article 7(1)(e)(ii) of both Regulations, the sign consists exclusively of the shape of goods which is necessary to obtain a technical result.
Article 7(1) grounds for refusal to register a mark must be interpreted in light of the public interest underlying them. The public interest underlying Article 7(1)(e)(ii) is to prevent the use of trademark law to obtain a monopoly on technical solutions or the functional characteristics of a product (General Court § 32, citing Lego Juris v. OHIM, § 43). Advocate General Szpunar explained further in his Opinion that allowing such marks to be registered would give the registrant “an unfair competitive advantage” and thus trade mark law cannot be used “in order to perpetuate, indefinitely, exclusive rights relating to technical solutions” (AG Szpunar Opinion § 32 and § 34).
For the General Court, Article 7(1)(e)(ii) applies only if the essential characteristics of the mark perform the technical functions of the goods “and have been chosen to perform that function.” It does not apply if these characteristics are the result of that function (General Court § 53). Simba argued in front of the CJEU that the General Court erred in this interpretation of Article 7(1)(e)(ii).
Simba claimed that the black lines of the cube performed a technical function (General Court § 51). But the General Court found that an objective observer is not able to infer by looking at the graphic representation of the Rubik’s cube mark that the black lines are rotatable (General Court § 57). The General Court held that Simba’s “line of argument… [was] essentially based on knowledge of the rotating capability of the vertical and horizontal lattices of the Rubik’s cube. However, it [was] clear that that capability cannot result from the black lines in themselves or, more generally, from the grid structure which appears on each surface of the cube… but at most from [an invisible] mechanism internal to that cube” (General Court § 58). Therefore, the grid structure on each surface of the cube “d[id] not perform, or are not even suggestive of, any technical function” (General Court § 60). The General Court concluded that registering the Rubik’s cube shape did not create a monopoly on a technical solution and mechanical puzzles competitors could also incorporate movable or rotatable elements (General Court § 65).
But, for AG Szpunar, the General Court erred in its analysis as it should have taken into account the function of the Rubik’s cube, which is a three-dimensional puzzle consisting of movable elements. He noted that in both the Philips and the Lego Juris cases, the competent authorities had analyzed the shape of the goods using additional information other than the graphic representation (AG Szpunar Opinion § 86). While the competent authority does not have to concern itself with hidden characteristics, it must nevertheless analyze “the characteristics of the shape arising from the graphic representation from the point of view of the function of the goods concerned” (AG Szpunar Opinion § 88).
The CJEU followed its AG’s Opinion on this point and found that the General Court should have defined the technical function of the actual goods, namely, the three-dimensional puzzle, and it should have taken this into account when assessing the functionality of the essential characteristics of that sign (CJEU § 47). The General Court “interpreted the criteria for assessing Article 7(1)(e)(ii) . . . too narrowly”(CJEU § 51) and should have taken into account the technical function of the goods represented by the sign when examining the functionality of the essential characteristics of that sign (CJEU § 52). Failing to do so would have allowed the trademark owner to broaden the scope of trademark protection to cover any three dimensional puzzles with elements in the shape of a cube (CJEU § 52).
This case confirms, after Pi-Design AG v. Bodum, that the CJEU takes the view that the essential characteristics of a trade mark must not be assessed solely by the competent authority based on visually analyzing the mark as filed, but that the authority must also identify the essential characteristics of a sign, in addition to the graphic representation and any other descriptions filed at the time of the application for registration. This is necessary to protect the public interest underlying Article 7(1)(e)(ii), which is to ensure that economic operators cannot improperly appropriate for themselves a mark which incorporates a technical solution.
By Marie-Andrée Weiss
MOB, founded by Tara Martin, sells canvas bags featuring “My other bag …” on one side and a drawing of an upscale handbag, such as the Hermès Kelly bag, the quilted Chanel bag, or a Louis Vuitton bag, on the other side. The MOB bags retail for $35 to $55, depending on their size. In creating this product, Ms. Martin was inspired by the “My other car is a … [name of luxury car]” novelty bumper stickers.
French luxury accessory company Louis Vuitton (LV) did not appreciate this use of its trademark and copyright-protected designs and filed a trademark infringement, trademark dilution by blurring, and copyright infringement suit against MOB, which moved for summary judgment, claiming the bags are a parody. On 8 January 8 2016, Judge Furman of the Southern District of New York (SDNY) found the MOB bags to be parody and granted MOB’s motion for summary judgment. LV then appealed the decision to the Second Circuit.
This case is interesting, as the Judge Furman found that parody was a successful defense against trademark dilution, trademark infringement, and copyright infringement claims. An amicus curia brief even argues that applying trademark dilution law to MOB’s use of LV’s trademarks would violate the First Amendment.
Trademark dilution claim
The Federal Trademark Dilution Act of 1995 (FTDA), 15 U.S.C. § 1125(c), was passed to give owners of famous trademarks protection against uses of their marks which would lessen their ability to identify and distinguish their goods or services, even if there is no likelihood of confusion. In Tiffany (NJ) Inc. v. eBay Inc., the Second Circuit defined dilution by blurring as “the whittling away of the established trademark’s selling power and value through its unauthorized use by others.” Dilution occurs because consumers are developing new associations with the famous mark.
15 U.S.C. § 1125(c)(2)(B) enumerates six factors to be used to assess whether a particular use of a famous mark dilutes it: (1) the degree of similarity between the challenged mark and the famous mark; (2) the degree of distinctiveness of the famous mark; (3) the extent to which the owner of the famous mark is engaging in exclusive use of the mark; (4) the degree of recognition of the famous mark; (5) whether the user of the mark or trade name intended to create an association with the famous mark; and (6) any actual association between the mark or trade name.
What is a parody?
15 U.S.C. § 1125(c)(3)(C) provides a parody defense against a trademark dilution claim if the use of a famous mark is made to “identif[y] and parod[y], criticiz[e], or commen[t] upon the famous mark owner or the goods or services of the famous mark owner.” However, the FTDA does not define parody. Judge Furman quoted the Fourth Circuit Louis Vuitton Malletier S.A. v. Haute Diggity Dog, LLC case, where the Court explained that “parody must convey two simultaneous — and contradictory — messages: that it is the original, but also that it is not the original and is instead a parody. This second message must not only differentiate the alleged parody from the original but must also communicate some articulable element of satire, ridicule, joking, or amusement.”
In Haute Diggity Dog, the Fourth Circuit had found that the “Chewy Vuiton” chew toys for dogs, which were small and rather crude imitations of Louis Vuitton bags, did not dilute LV’s trademarks. The Court noted that “[t]he [LV] handbag is provided for the most elegant and well-to-do celebrity, to proudly display to the public and the press, whereas the imitation ‘Chewy Vuiton’ ‘handbag’ is designed to mock the celebrity and be used by a dog… The dog toy is a comment on the rich and famous, on the LOUIS VUITTON name and related marks, and on conspicuous consumption in general.” Judge Furman also quoted § 31:153 of the McCarthy treaty on Trademarks and Unfair Competition which explains that a particular use of a trademark is a parody if it communicates to the public that is “separate and distinct from the trademark owner” and “is poking fun at a trademark or the policies of its owner.” Judge Furman noted that “my other car” bumper stickers were “a joke — a riff, if you will, on wealth, luxury brands, and the social expectations of who would be driving luxury and non-luxury cars.” He concluded that MOB’s use of LV trademark was a parody.
Are MOB bags a parody?
On appeal, LV argues that a parody must be “directed at or ‘upon’ the trademark owner or its famous mark.” For LV, the MOB parody does not target LV but instead makes fun of MOB itself, “at best,” and is not even a parody, but, rather, a “social commentary” (LV reply brief p.8). LV cites the Harley Davidson, Inc. v. Grottanelli case, in which the Second Circuit held in 1999 that using a mark to “humorously… promote [one’s] own products and services… is not a permitted trademark parody use” (Harley Davidson at 813). The Second Circuit, however, noted in Harley Davidson that it has “accorded considerable leeway to parodists whose expressive works aim their parodic commentary at a trademark or a trademarked product . . . but ha[s] not hesitated to prevent a manufacturer from using an alleged parody of a competitor’s mark to sell a competing product” (Harley Davidson at 812). The Second Circuit found no parody in Harley Davidson because it found the use was not a comment.
MOB argues that “MOB’s parody is communicated to consumers with a combination of features of the bags themselves, including, for example, the large stylized text ‘My Other Bag . . .’ on one side, the cartoonish depictions of designer handbags on the other side, the plain canvas material, and the bag-on-bag concept itself.” MOB argued that, therefore, the parody is communicated by two simultaneous and contradictory message, within the meaning of Haute Diggity, “that it is the original, but also that it is not the original”(Appellee brief at 19) (emphasis in original).
The amicus curia brief filed by Professor Christopher Jon Sprigman and Professor Rebecca Tushnet argued that the “MOB bags are commenting humorously on society’s obsession with owning status symbols, and with showing them off” (Amicus Curia brief at 4). Indeed, LV bags are not the sole source of inspiration for MOB. Some of its canvas bags also feature bags from Hermès, Chanel, or Fendi. The Fourth Circuit had noted in Haute Diggity Dog that the Chewy Vuiton parody was “enhanced by the fact that the ‘Chewy Vuiton’ dog toys are sold with similar parodies of other famous and expensive brands” such as “Chewnel No. 5” or “Jimmy Chew.”
Judge Furman used the eight-factor Polaroid test used in the Second Circuit to assess whether an unauthorized use of the mark is infringing. He found that the trademark infringement claim “fail[ed] for much the same reasons that [the] dilution claims failed.” As noted by the Fourth Circuit in Haute Diggity Dog (at 261), this is not surprising because “[w]hile it is true that finding a mark to be strong and famous usually favors the plaintiff in a trademark infringement case, the opposite may be true when a legitimate claim of parody is involved . . . In cases of parody, a strong mark’s fame and popularity is precisely the mechanism by which likelihood of confusion is avoided.” Consumers are not likely to be confused by a parody.
Interestingly, several celebrities who probably indeed own one or more LV bags, or, at least, could certainty afford to own one, have been seen carrying a MOB bag, some even using it as a grocery bag as envisioned by Ms. Martin when she first imagined the MOB bag. The Instagram account of MOB shows the picture of a woman carrying what appears to be an authentic quilted Chanel bag on one hand, and the MOB featuring the same Chanel bag on the other hand. These consumers are clearly not confused. MOB notes in its brief that LV “has admitted repeatedly… it knows that consumers are not confused” (Appellee brief at 46).
LV argues that there is post-sale confusion, which is recognized by courts, including the Second Circuit (Lois Sportswear, USA, Inc. v. Levi Strauss & Co.). For LV, “potential consumers or passersby could come to believe that MOB . . . is associated with [LV] or it at least consented to the use of its trademark” (Appellant reply brief at 21). However, as noted by the SDNY in Gucci America, Inc. v. Guess?, Inc., “a post-sale confusion plaintiff must still establish a likelihood of confusion among an appreciable number of post-sale observers, taking into account all the vagaries involved with post-sale observation,” further noting that “the fact that post-sale observers are removed from purchasing decisions makes post-sale trademark cases inherently difficult to prove, speculative, and subject to increased scrutiny” (Gucci America at 239). It remains to be seen if LV will meet this high standard. Professor Sprigman and Professor Tushnet noted in their brief that “the risks of post-sale confusion are lower with parody,” adding that the MOB is obviously a parody, which appears to the consumer either pre-sale or post-sale (Amicus Curia brief at 18).
Judge Furman had also found that MOB’s use of the LV trademark was fair use. As LV does not find that MOB bags are a parody, it argues that there is no fair use. Professor Sprigman and Professor Tushnet posit in their amicus curia brief that “LV did not file this case to protect its incentive to create new expressive works [but] to protect its brand from mockery,” adding that “any effect on the value of [LV]’s trademarks… is not the kind of harm copyright aims to avoid” (Amicus Curiae brief at 23).
Is the TDRA unconstitutional?
Professor Sprigman and Professor Tushnet argued in their brief that the TDRA “creates a content-based right that applies to non-misleading commercial speech” (Amicus Curiae brief at 20). They note that the Second Circuit held in United States v. Caronia, applying the 2011 Supreme Court Sorrell v. IMS Health Inc. case, that “content-based suppression of non-misleading speech, including commercial non-misleading speech… must be shown to be narrowly tailored to serve compelling state interests” (Amicus Curiae brief at 19). They argue that Congress has not shown there was a compelling interest to enact the TDRA and that it did not used the least restrictive way available in order to prevent dilution of famous marks. Therefore, the TDRA violates the First Amendment.
We saw in another post that the Federal Circuit recently found the anti-disparaging provision of the Trademark Act violated the First Amendment. Now, the TDRA may be found unconstitutional. These leaves the lingering question: will the tension between the First Amendment and trademark law continue to increase or begin to diminish?
By Marie-Andrée Weiss
We saw in the last issue of the TTLF newsletter that the Federal Circuit held en banc that the disparagement provision of § 2(a) of the Trademark Act, 15 U.S.C. 1052(a), which forbids registration of disparaging trademarks, violates the First Amendment.
The case is about the mark THE SLANTS, which Simon Tam is seeking to register in connection for live performances of his dance-rock music group. The United States Patent and Trademark Office (PTO) refused to register it, claiming it was an ethnic slur disparaging to persons of Asian ancestry. The Trademark Trial and Appeal Board (TTAB) affirmed, but the Federal Circuit ruled in favor of Mr. Tam and remanded the case for further proceeding.
Now the PTO has filed a petition for a writ of certiorari asking the Supreme Court to answer “[w]hether the disparagement provision in 15 U.S.C. 1052(a) is facially invalid under the Free Speech Clause of the First Amendment.”
The Federal Circuit held that §2(a) “penaliz[es] private speech merely because [the government] disapproves of the message it conveys.” The PTO argues in its petition that §2(a) does not prohibit any speech, but only “directs the PTO not to provide the benefit of federal registration to disparaging marks” (petition p. 8).
Would refusing to register THE SLANTS as a trademark merely deny Mr. Tam the benefits of federal trademark registration, or would his freedom of speech be violated?
PTO argument: Mr. Tam is merely denied the benefits of federal registration
The petition notes that federal registration does not create trademarks, but is merely “a supplement to common-law protection” and that a person who first uses a distinct mark in commerce acquires rights to this mark, citing the 1879 In re Trade-Mark Cases Supreme Court case (petition p. 3 and p. 11). The PTO further argues that “[t]he holder of a trademark may use and enforce his mark without federal registration” (petition p. 3). Mr. Tam would still have federal remedies available to him to protect his mark, even if THE SLANTS is not federally registered. For example, the Anti-Cybersquatting Consumer Protection Act, 15 U.S.C. § 1125(D), does not require the mark seeking protection to be registered (petition p. 12).
For the PTO, Section 1052(a) is not unconstitutional, as it does not prohibit speech, nor does it proscribe any conduct or restrict the use of any trademark. Instead, it merely “directs the PTO not to provide the benefits of federal registration to disparaging marks.” Since a mark can function as a mark without the benefit of federal registration, even if a mark is speech, it does not need the benefit of federal registration to be expressed, and therefore, it is not a violation of the First Amendment to refuse to register it.
The PTO further argues that the purpose of Section 1052(a) is to avoid the federal government “affirmatively promot[ing] the use of racial slurs and other disparaging terms by granting them the benefits of registration” (petition p. 10) and that “Congress legitimately determined that a federal agency should not use government funds to issue certificates in the name of the United States of America conferring statutory benefits for use of racial slurs and other disparaging terms” (p. 15-16).
However, in In re Old Glory Condom Corp.(at FN3), the Trademark Trial and Appeal Board noted that “the issuance of a trademark registration for applicant’s mark [does not amount] to the awarding of the U.S. Government’s “imprimatur” to the mark.”
PTO argument: no violation of free speech, as Mr. Tam can still use his mark to convey his message
The PTO also argues that Section 1052(a) is not an affirmative restriction on speech because the federal law does not prevent Mr. Tam “from promoting his band using any racial slur or image he wishes,” does not limit Mr. Tam’s choice of songs played, or the messages he wishes to convey (petition p. 12).
The PTO gives Sons of Confederate Veterans, Inc. as an example of a case where the Supreme Court recognized the government’s right to “take into account the content of speech in deciding whether to assist would-be private speakers.” However, this case can be distinguished from our case. The Supreme Court held in Sons of Confederate Veterans that a state can refuse to issue a specialty license plate if it carries a symbol which the general public finds offensive, in that case a confederate flag. But the owner of a car can still reap the government benefits of car registration, which is mandatory to operate a motor vehicle, even though his speech has been suppressed by the government as disparaging, while, in our case Mr. Tam cannot reap the benefits provided to the holder of a federally registered mark. The fact that he still has some benefits as the owner of a common law mark is irrelevant.
The petition also gives National Endowment for the Arts v. Finley as an example of a case where the Supreme Court upheld the government’s right to take moral issues into consideration when denying a federal benefit. In this case, a court of appeals had held that §954(d) (1) of the National Foundation of the Art and the Humanities Act violated the First Amendment. This federal law requires the Chairperson of the National Endowment for the Arts (NEA) to make sure “that artistic excellence and artistic merit are the criteria by which applications are judged, taking into consideration general standards of decency and respect for the diverse beliefs and values of the American public.” The Supreme Court found in Finley that §954(d) (1) was constitutional.
However, Finley can also be distinguished from our case. The Supreme Court noted there that respondents had not “allege[d] discrimination in any particular funding decision” and that, therefore, the Supreme Court could not assess whether a particular refusal for the NEA grant was “the product of invidious viewpoint discrimination.” In our case, we do know that the only reason the PTO refused to register THE SLANTS is because it assessed the mark to be disparaging, and so the Supreme Court could very well find this decision to be an “invidious viewpoint discrimination.” Also, while not receiving a grant from the NEA may make it more difficult for an artist to create art, it does not entirely prevent it, even a particular piece of art which would shock standards of decency.
If a mark is a racial slur, should the intent of applicant for registering the mark matter?
The TTAB affirmed the refusal to register THE SLANTS because it was disparaging to persons of Asian ancestry and because the mark was disparaging to a substantial composite of this group. The PTO noted in its petition that the TTAB had determined “that Section 1052(a) prohibits registration of respondent mark despite the fact that respondent’s stated purpose for using the mark is to “reclaim” the slur as a sign or ethnic pride” (emphasis in original text, p. 13 of the petition). The PTO seems to argue that Section 1052(a) views disparaging content neutrally, without questioning the intent behind the choice of disparaging speech as trademark.
Judge Dyk from the Federal Circuit wrote in his concurring/dissenting opinion that he would have held that Section 1052(a) is facially constitutional because “the statute is designed to preclude the use of government resources not when the government disagrees with a trademark’s message, but rather when its meaning “may be disparaging to a substantial composite of the referenced group,” citing In re Lebanese Arak Corp.” In this case, the USPTO had refused to register KHORAN as a trademark for alcoholic beverages because it was disparaging to the beliefs of Muslims.
For Judge Dyk, the purpose of Section1052(a) is “to protect underrepresented groups in our society from being bombarded with demeaning messages in commercial advertising” and Section 1052(a) “is constitutional as applied to purely commercial trademarks, but not as to core political speech, of which Mr. Tam’s mark is one example.” Judge Dyk argued further that, while the First Amendment protects speech which is offensive to some in order to preserve a robust marketplace of ideas, “this principle simply does not apply in the commercial context,” giving as example racial or sexual harassment in the workplace.
But this argument seems to make a difference between registrants: Mr. Tam could register a racial slur to make a point, but could not do so if his purpose for registering the same mark would be to insult people of Asian descent. This interpretation of Section 1052(a) is troubling, as courts would have to determine if a particular mark is indeed political speech, then decide if it is “good” political speech or “bad” political speech. This is noted by the PTO in a footnote to the petition as being viewpoint discrimination which violates the First Amendment (p. 13).
The PTO argues that if Section 1052(a) is unconstitutional, then the PTO can no longer refuse to register as a trademark “even the most vile racial epithet” (p.10). Mr. Tam does not deny that “slant” is an ethnic slur. Indeed, he choose to name his band “The Slants” because it is a slur, in order to “take on stereotypes” about Asians (petition p. 5). Therefore, the mark may be an ethnic slur, but it is not disparaging. It all depends on the eyes and ears of the beholder. This was also the idea behind the attempted registration of HEEB or DIKES ON BIKE as trademarks.
Therefore, the question of who are the members of the group of reference is important. But it should not be.
Is it possible to protect minorities and the First Amendment?
If Mr. Tam would be authorized to register his trademark, it would be a victory for freedom of speech. The Slants would be able to promote further their anti-xenophobic message, and this would benefit the nation as a whole. But what if a person or an entity wishes to trademark a racial slur in order to advocate xenophobia? Owning the trademark could then serve as a tool to censor speech opposing racism. It would not be the first time that trademarks are used to suppress speech.
One can also argue that allowing disparaging trademark to be registered could confuse consumers about the origin of the product. Some consumers would not understand that a particular term is a racial slur. Others may understand it, but not know that it was meant to be used to fight prejudice. Since the function of a trademark is also to reduce consumer search costs, federal law could create a sign informing consumers that the trademark is used in an ironic way. I propose adding in these cases an irony punctuation (¿) after®. Is it a good idea?¿