Summary
concluding that asserted trade dress for cookie design was "nothing more than a nondistinctive combination of a few basic, common design elements. . . ., which naturally result from a basic and common method of making cookies."
Summary of this case from Berg v. SymonsOpinion
CIV. NO. 01-00449 SOM/LEK
May 30, 2003
Gregory Wood (argued), Fulbright Jaworski LLP, Austin, TX, Anna M. Elento-Sneed (appeared but did not argue), Carlsmith Ball LLP, Honolulu, HI, for Plaintiffs.
William G. Meyer, III (argued), Paul A. Schraff (appeared, but did not argue), Dwyer, Schraff, Meyer, Jossem Bushnell, Honolulu, HI, for Defendants.
ORDER GRANTING DEFENDANTS' MOTION FOR PARTIAL SUMMARY JUDGMENT AND DENYING PLAINTIFF'S CROSS-MOTION FOR PARTIAL SUMMARY JUDGMENT
INTRODUCTION
Plaintiff Big Island Candies, Inc. ("BIC"), has sued Defendants The Cookie Corner, James McArthur, Angus McKibbin, and the Cookie Masters of Hawaii ("Cookie Corner"), alleging trade dress infringement and dilution under the Lanham Act and unfair competition under Hawaii law. BIC seeks to enjoin Cookie Corner from marketing a chocolate-dipped shortbread cookie (the "Cookie Corner Cookie") that BIC claims uses the trade dress of BIC's chocolate-dipped shortbread cookie (the "BIC Cookie"). BIC also seeks damages.
BIC's claims relate to both the design of the BIC Cookie itself and the packaging of the BIC Cookie. Neither the BIC Cookie design nor the packaging is federally registered. The present motions for summary judgment relate only to the design of the BIC Cookie.
The court had expressly invited the parties to include in these motions any arguments that the packaging was or was not generic. Cookie Corner made no such argument, and BIC says that it consequently filed no cross-motion on that subject.
Cookie Corner has moved for partial summary judgment on BIC's claim of trade dress infringement with respect to the BIC Cookie design, arguing that the BIC Cookie design is generic and not protectable. BIC has filed a cross-motion seeking partial summary judgment on "nongenericness."
This is the second round of summary judgment motions regarding the protectability of the BIC Cookie design. In an earlier order, this court (1) granted in part and denied in part BIC's motion for partial summary judgment on the issue of functionality and certain of Cookie Corner's defenses, and (2) denied Cookie Corner's cross-motion on the issue of functionality. See Big Island Candies, Inc. v. Cookie Corner, 244 F. Supp.2d 1086 (D.Haw. 2003).
Cookie Corner's motion for partial summary judgment is GRANTED. BIC's cross-motion is DENIED.
BACKGROUND
The facts in this case are set forth in this court's earlier order and are supplemented herein only as necessary.
The BIC Cookie is a rectangular macadamia-nut shortbread cookie with "bull-nose" (i.e., rounded) corners that is diagonally dipped in chocolate. The dip extends roughly from corner to corner so that the chocolate covers about half the surface area of the cookie. The Cookie Corner Cookie is almost identical in appearance, but it is slightly larger.
The corners are rounded and smooth, rather than crumbly and squared-off, because the cookies are cut before they are baked. If the cookies were first baked in a tray and then cut, they would have squared-off, crumbly corners.
The BIC Cookie is approximately 1.25 inches wide, 2.25 inches long, and .375 inches thick, and weighs approximately 15.7 grams. The Cookie Corner Cookie measures approximately 1.5 inches wide, 2.5 inches long, and 0.5 inches thick, and weighs approximately 20.4 grams.
BIC's Second Amended Complaint ("Complaint") alleges that the BIC Cookie design elements "copied" by Cookie Corner are (1) the rectangular shape; (2) the bull-nose corners; (3) the size of the cookie; and (4) the diagonal dip. Compl. ¶ 18a.
BIC filed this action on July 3, 2001.
STANDARD OF REVIEW
Summary judgment shall be granted when:
the pleadings, depositions, answers to interrogatories and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.
Fed.R.Civ.P. 56(c); see also Addisu v. Fred Meyer, Inc., 198 F.3d 1130, 1134 (9th Cir. 2000). One of the principal purposes of summary judgment is to identify and dispose of factually unsupported claims and defenses. Celotex Corp. v. Catrett, 477 U.S. 317, 323-24 (1986).
Summary judgment must be granted against a party that fails to demonstrate facts to establish what will be an essential element at trial. See id. at 323. A moving party without the ultimate burden of persuasion at trial — usually, but not always, the defendant — has both the initial burden of production and the ultimate burden of persuasion on a motion for summary judgment. Nissan Fire Marine Ins. Co., Ltd. v. Fritz Cos., 210 F.3d 1099, 1102 (9th Cir. 2000).
All evidence and inferences must be construed in the light most favorable to the nonmoving party. T.W. Elec. Service, Inc. v. Pac. Elec. Contractors Ass'n, 809 F.2d 626, 631 (9th Cir. 1987). Inferences may be drawn from underlying facts not in dispute, as well as from disputed facts that the judge is required to resolve in favor of the nonmoving party. Id. When "direct evidence" produced by the moving party conflicts with "direct evidence" produced by the party opposing summary judgment, "the judge must assume the truth of the evidence set forth by the nonmoving party with respect to that fact." Id.
On these cross-motions, inferences are drawn against each party on its own motion and in favor of each party insofar as it opposes a motion.
ANALYSIS Generic Product Designs Are Not Protectable
Cookie Corner argues that BIC is trying to protect a product design that is generic and therefore unprotectable. BIC, by contrast, asserts that the concept of genericness does not apply to the design of any product, including a cookie. The court begins its analysis by examining whether the concept of genericness applies to product design.Courts have considered the concept of genericness in greater detail in connection with trademarks than with trade dress. Trademarks that include terms have traditionally been classified using the formulation set out in Abercrombie Fitch Co. v. Hunting World, Inc., 537 F.2d 4 (2d Cir. 1976). In Abercrombie, Judge Friendly identified the following categories of terms, in order of increasing distinctiveness: (1) generic; (2) descriptive; (3) suggestive; and (4) arbitrary or fanciful. Id. at 9. Suggestive, arbitrary, and fanciful terms are considered inherently distinctive and therefore entitled to protection, whereas descriptive terms must have acquired distinctiveness, or secondary meaning, before they may be protected as marks. Two Pesos, Inc. v. Taco Cabana, Inc., 505 U.S. 763, 768-69 (1992).
Some courts, including the Supreme Court in Two Pesos, 505 U.S. at 768, have used the phrase "generic mark" in their discussion of genericness and distinctiveness. That phrase, however, is an oxymoron: "Either a designation is protectable as a `mark' or it is a `generic name' of a thing or service, in which case it can never be a protectable mark. A given designation cannot be both at once." 2 J. McCarthy, McCarthy on Trademarks and Unfair Competition § 12:1 (4th ed. 2003) [hereinafter "McCarthy"]. This court is aware of no case in which a court has held that a "generic mark" was protectable.
While the Abercrombie classification clearly applies when a plaintiff seeks trademark protection for a term, it applies in only certain types of trade dress cases. For instance, restaurant decor may be inherently distinctive, but color alone cannot be. Id. at 776 (holding that inherently distinctive restaurant decor was protectable without a showing that it has acquired secondary meaning); Qualitex Co. v. Jacobson Prods. Co., Inc., 514 U.S. 159, 162-63 (1995) (noting that "a product's color is unlike `fanciful,' `arbitrary,' or `suggestive' words or designs, which almost automatically tell a customer that they refer to a brand").
The circuit courts of appeal were divided as to whether the traditional categories set forth in Abercrombie applied to cases of product design or product configuration. Leatherman Tool Group, Inc. v. Cooper Indus., Inc., 199 F.3d 1009, 1013 n. 6 (9th Cir. 1999), vacated in part on other grounds, 532 U.S. 424 (2001) (noting a circuit split regarding "whether the traditional categories of generic, descriptive, suggestive, arbitrary, and fanciful can be applied meaningfully in product configuration cases").
The Fourth Circuit had held that the Abercrombie categories "provide the appropriate basic framework for deciding inherent distinctiveness in product configuration cases." Ashley Furniture Indus., Inc. v. Sangiacomo N.A. Ltd., 187 F.3d 363, 372 (4th Cir. 1999). The Second and Third Circuits had held that it did not make sense to apply the Abercrombie categories to product design. Knitwaves, Inc. v. Lollytogs, Ltd., 71 F.3d 996 (2d Cir. 1995); Duraco Prods., Inc. v. Joy Plastic Enterprises, Ltd., 40 F.3d 1431 (3d Cir. 1994). The Second Circuit, however, had "declined to follow the Third Circuit's application of a wholly-new, multi-pronged test to product design . . . cases under § 43(a)," and "instead simply asked whether the design was likely to be understood as an indicator of the product's source." Landscape Forms, Inc. v. Columbia Cascade Co., 113 F.3d 373, 378 (2d Cir. 1997) (discussing Knitwaves and Duraco). Landscape Forms suggested that the Second Circuit was "moving toward a rule that packaging is usually indicative of a product's source, while the design or configuration of the product is usually not so." Id. at 379.
The Supreme Court resolved the circuit split in Wal-Mart Stores, Inc. v. Samara Bros., Inc., 529 U.S. 205 (2000). Noting that "even the most unusual of product designs . . . is intended not to identify the source, but to render the product itself more useful or more appealing," the Supreme Court held in Wal-Mart that product design, like color, cannot be inherently distinctive. 529 U.S. at 211-12, 213. As a result, product design is "distinctive, and therefore protectible, only upon a showing of secondary meaning." Id. at 216. Wal-Mart did not directly address, however, whether every product design that acquires secondary meaning is protectable. The Court did not discuss whether a product design may be "generic," i.e., whether a design may be of a type that can never be distinctive, and therefore is not protectable.
In Abercrombie Fitch Stores, Inc. v. American Eagle Outfitters, Inc., 280 F.3d 619 (6th Cir. 2002) ("American Eagle"), the Sixth Circuit recognized that Wal-Mart "[left] open the conceptual possibility that generic designs bearing secondary meaning are protectable," but concluded that "generic product configurations," or "designs regarded by the public as the basic form of a particular item," should not be protectable even upon a showing of secondary meaning. American Eagle, 280 F.3d at 638. BIC argues that the Sixth Circuit misread Wal-Mart. Opp. at 9-10. Emphasizing that the Second Circuit is "one of the most experienced circuits in dealing with trademark matters," BIC argues that Wal-Mart adopted the Second and Third Circuits' approach to product configuration cases and that this court should do the same. Id. at 7-8.
What BIC does not mention in its papers, however, is that the Second Circuit, even after Wal-Mart, has held that "generic" product designs are not protectable. See Yurman Design, Inc. v. PAJ, Inc., 262 F.3d 101, 115 (2d Cir. 2001) (noting that "even a showing of secondary meaning is insufficient to protect product designs that are overbroad or `generic'"). In Yurman, the Second Circuit, examining the rule that genericness bars protectability, said:
This check on monopoly rights limits all marks, but for two reasons it is particularly important in cases of product design. First, overextension of trade dress protection can undermine restrictions in copyright and patent law that are designed to avoid monopolization of products and ideas. Patent and copyright law bestow limited periods of protection, but trademark rights can be forever. Second, just as copyright law does not protect ideas but only their concrete expression, neither does trade dress law protect an idea, a concept, or a generalized type of appearance. Product design is driven primarily by the usefulness or aesthetic appeal of the object; trade dress protection for product design therefore entails a greater risk of impinging on ideas as compared with protection of packaging or labeling.
Id. at 115-16 (internal citations and quotations omitted).
At the hearing, when asked about Yurman's statement that "generic" product designs are not protectable, BIC argued that that portion of Yurman was dicta. Alternatively, BIC argued that that portion of Yurman was simply incorrect and does not represent the law in the Second Circuit. It is unclear whether or not BIC intends to stick to its position that the Second Circuit is a leader in the development of trademark law. Regardless, this court does not agree that Yurman's discussion of the genericness requirement is dicta.
Yurman held that "a plaintiff seeking to protect its trade dress in a line of products must articulate the design elements that compose the trade dress." 262 F.3d at 116. Among the reasons for that holding were (1) that "no juror can evaluate secondary meaning, overbreadth, or nonfunctionality without knowing precisely what the plaintiff is trying to protect"; and (2) a plaintiff's inability to articulate the design elements of the alleged trade dress "may indicate that its claim is pitched at an improper level of generality, i.e., the claimant seeks protection for an unprotectible style, theme or idea." Id. at 117.
Yurman then cites the Second Circuit's Milstein decision, generally seen as the first case in which the Second Circuit defined "generic trade dress" and held that it was unprotectable. Jeffrey Milstein, Inc. v. Greger, Lawlor, Roth, Inc., 58 F.3d 27, 32 (2d Cir. 1995); 1 McCarthy § 8:6.1; 2 McCarthy 12:36.1. As discussed in a later section of this order, Milstein defined "generic trade dress" as trade dress that was so generalized, or overbroad, that protecting it would amount to protecting an idea or concept. Accordingly, it is clear that the holding in Yurman is driven in large part by the concern that the plaintiff might be seeking protection of an "overbroad" or "generic" product design. In expressing this concern, the Second Circuit was clearly aware of Wal-Mart's impact on the law regarding product design. See Yurman, 262 F.3d at 116 (recognizing that the articulation reqirement had been announced "in the context of whether a plaintiff had satisfied the test for `inherent distinctiveness' — which is no longer at issue in product design cases").
This court agrees with the Second and Sixth Circuits that generic product designs are not protectable even upon a showing of secondary meaning. Id.; American Eagle, 280 F.3d at 638. The court sees no indication in Wal-Mart that the Court intended to except product design from the nongenericness requirement that limits protection of other types of marks. Making such an exception would make it easier for the party asserting protectable product design to prevail. Wal-Mart, however, expresses concern about deterring competition. Wal-Mart makes it more difficult to obtain trademark protection for product design features than for other types of trade dress, such as packaging. See 529 U.S. at 213-14 (stating that "[c]onsumers should not be deprived of the benefits of competition with regard to the utilitarian and esthetic purposes that product design ordinarily serves," and rejecting one proposed test for inherently distinctive product design because "[s]uch a test would rarely provide the basis for summary disposition of an anticompetitive strike suit").
The policy concerns articulated in Wal-Mart are not addressed by BIC in its papers. Rather, BIC argues that, in contrast to cases involving alleged word marks, there is no concern in product design cases that competitors must be free to use generic names to describe their goods. BIC also argues that there are no First Amendment policy considerations in the area of product design. However, as Wal-Mart illustrates, there are other important concerns in the product design context. See also Landscape Forms, 113 F.3d at 380 ("While trademarking a generic term would create a monopoly in a necessary word or phrase, granting trade protection to an ordinary product design would create a monopoly in the goods themselves.").
Wal-Mart's failure to discuss or even mention the Abercrombie category of "generic" terms (or trade dress) should not be read as a rejection of the idea that trade dress can be generic and hence unprotectable. Because Wal-Mart's holding limits how trade dress can be shown to be distinctive, Wal-Mart's discussion of the Abercrombie categories is understandably limited to those categories that either are inherently distinctive or may be shown to be distinctive. 529 U.S. at 210. Wal-Mart first explains that, although the text of section 43(a) of the Lanham Act, 15 U.S.C. § 1125(a), does not require distinctiveness, courts have universally required that a producer seeking to protect its trade dress show that its trade dress is distinctive. 529 U.S. at 210. The Court then explains that "courts have held that a mark can be distinctive in two ways," either (1) because it is "fanciful," "arbitrary," or "suggestive" and thus inherently distinctive, or (2) because it has developed secondary meaning. Id. The Court's failure to mention a category that can never be shown to be distinctive, in a discussion regarding how to show distinctiveness, does not indicate the nonexistence of such a category.
BIC has not identified, and this court is not aware of, any post-Wal-Mart authority for the proposition that there is no such thing as a "generic" product design — i.e., a product design that is unprotectable even upon a showing of secondary meaning, assuming that all other requirements for protection are met. Rather, BIC relies primarily on its Wal-Mart arguments, which this court rejects for the reasons explained above. Accordingly, the court concludes that, if the design of the BIC Cookie is "generic," it is not protectable even upon a showing of secondary meaning.
The question of what it means for a product design to be "generic" is discussed in a later section of this order.
BIC Has the Burden of Proving Nongenericness.
Because Cookie Corner has raised the defense of genericness, BIC has the burden of showing that the BIC Cookie design is not generic. Filipino Yellow Pages, Inc. v. Asian Journal Publ'ns, Inc., 198 F.3d 1143, 1146 (9th Cir. 1999) ("If a supposedly valid mark is not federally registered, however, the plaintiff has the burden of proving nongenericness once the defendant asserts genericness as a defense."); 2 McCarthy § 12:12.BIC, relying on Murphy Door Bed Co., Inc. v. Interior Sleep Systems, Inc., 874 F.2d 95, 100-101 (2d Cir. 1989), argues that Cookie Corner has the burden of proof. In Murphy Door, the Second Circuit held that, "where the public is said to have expropriated a term established by a product developer, the burden is on the defendant to prove genericness." 874 F.2d at 101. Even assuming that the holding of Murphy Door is applicable in the Ninth Circuit, however, it does not change the allocation of proof burdens here because the determinative issue is what Cookie Corner claims, not what Cookie Corner is able to prove. Id. ("Thus, critical to a trial court's allocation of proof burdens is a determination of whether the term at issue is claimed to be generic by reason of common usage or by reason of expropriation."). Cookie Corner is not claiming that BIC established the BIC Cookie design and that the design was subsequently expropriated by the public. Rather, Cookie Corner alleges that the design features that BIC seeks to protect were widely used before BIC began manufacturing the BIC Cookie in 1985. Accordingly, BIC has the burden of proving nongenericness.
Of course, if Cookie Corner made, for instance, a bad-faith claim of "common usage" (before 1985), then it is possible that the burden would fall on Cookie Corner to show genericness under the Murphy Door standard. In this case, however, the court has no reason to conclude that Cookie Corner is making a bad-faith assertion.
The BIC Cookie Design Is Generic and Not Protectable.
The Test for Genericness in the Product Design Context May Be Discerned from Existing Ninth Circuit Law.This court recognizes the difficulty of applying traditional definitions of "genericness" to product configuration or product design features. Most cases regarding "genericness" relate to trademarks and forms of trade dress other than product design, and the tests articulated in those cases do not translate easily into the product design context. See, e.g., AM General Corp. v. DaimlerChrysler Corp., 311 F.3d 796, 828 (7th Cir. 2002) ("The traditional method of measuring a mark's strength with a yardstick that runs from generic at one end to arbitrary or fanciful at the other is a clumsy method when measuring the strength of trade dress or product design."); Landscape Forms, 113 F.3d at 378 (noting that "one cannot meaningfully ask whether a product's design features are generic or otherwise descriptive of the product itself"). The circuit split regarding the applicability of the Abercrombie classification to product design, which was resolved (at least in part) by Wal-Mart, stemmed from the difficulty of translating traditional concepts of trademark law to product design cases.
Cases addressing product design suggest that the term "genericness" in the product design context covers different concepts. First, a product design may be generic if it is overbroad or too generalized. "[T]rade dress law [does not] protect an idea, a concept, or a generalized type of appearance," but "the concrete expression of an idea in a trade dress" may be protectable, assuming that the other requirements for trademark protection are met. Jeffrey Milstein, Inc. v. Greger, Lawlor, Roth, Inc., 58 F.3d 27, 32 (2d Cir. 1995). "Drawing the line between `ideas' or `concepts' on the one hand and `concrete expressions' on the other may sometimes present close questions." Id. at 33. However, Milstein points out:
Often a helpful consideration will be the purpose of trade dress law: to protect an owner of a dress in informing the public of the source of its products, without permitting the owner to exclude competition from functionally similar products. . . . The level of generality at which a trade dress is described . . . may indicate that that dress is no more than a concept or idea to be applied to particular products.
Id. It is irrelevant whether or not the party asserting trade dress is the first to use the idea or concept for which it seeks protection. "[T]he first company to depict a heart and an arrow on Valentine's cards or to produce cards depicting tabby cats could not seek protection for those designs because they are concepts, defined abstractly. . . ." Id. at 33. Therefore, trade dress consisting of die-cut photographic greeting cards with plain white interiors is not protectable. Id. at 34 (holding that the plaintiff was seeking trade dress protection for a "generalized idea"). Nor will courts extend protection to product lines if the producer fails to identify a set of specific design elements that constitute its trade dress, in part because its failure to do so may indicate that its claim is overbroad. Yurman, 262 F.3d at 117.
The policy behind the Milstein definition of genericness is similar to the policy behind the functionality doctrine. 1 McCarthy § 7:69 (noting that trade dress category held unprotectable in Milstein "was dubbed `generic' — but its policy is similar to that of functionality"). Both policies seek to prevent trade dress protection from "undermin[ing] restrictions in copyright and patent law that are designed to avoid monopolization of products and ideas." Milstein, 58 F.3d at 32.
Second, a product design may be generic if it is the basic or a basic form of a type of product. The classic definition of a generic term is one that "refers to the genus of which the particular product is a species." Park `N Fly, Inc. v. Dollar Park Fly, Inc., 469 U.S. 189, 194 (1985). A "generic" product design may also be defined as one that "refer[s] to the genus of which the particular product is a species." Yurman, 262 F.3d at 115 (quoting Two Pesos, 505 U.S. at 768). Courts will not protect "basic" design features because of the concern that producers will be able to use trademark law to monopolize products. See Landscape Forms, 113 F.3d at 380. See also Indonesian Imports, Inc. v. Smith, 1999 WL 183629 at *6 (N.D.Cal. Mar. 30, 1999) (refusing trade dress protection to a "basic crocheted handbag").
Finally, a product design may be "generic" because it is so common in the industry or in the marketplace that it cannot be said to identify any particular source. E.g., Mana Prods., Inc. v. Columbia Cosmetics Mfg., Inc., 65 F.3d 1063 (2d Cir. 1995) (holding that a producer's makeup compacts' "size and shape . . . are common characteristics of the entire genre of makeup compacts" and consequently generic, and holding that black was a generic color for makeup compacts because it was so widely used in the industry).
Because a makeup compact is not discarded after purchase, it "could be considered to be either the product or part of its packaging." Landscape Forms, 113 F.3d at 379.
The above-described concepts of genericness are not necessarily separate. For instance, basic product shapes are usually common in a given industry, and a court may deny protection to a product design based on both its commonness and its basicness. See Indonesian Imports, 1999 WL 183629 at *7 ("Indonesian's claimed trade dress is . . . a composite of common and basic handbag features," and "Indonesian has not produced any evidence . . . that these generic and otherwise unprotectable features are combined together in a unique and source identifying way"). A court may also rely on the commonness of a product design to conclude that that design is "basic." Courts have recognized that there may be more than one applicable formulation of "genericness" in a product design case:
In the context of product design, a "generic" design can be understood as a design dictated by the product itself, or a design that is one of a small number of available alternatives. Or it may mean that the design is so common or unoriginal that it is utterly lacking in distinctiveness. If the first manufacturer to make a desktop electronic calculator used a rectangular case, we do not think it could claim trade dress protection and forbid others to make rectangular calculators, or even rectangular calculators having specified proportions, even if other shapes would not be more expensive to make.
Hanig Co., Inc. v. Fisher Co., Inc., 1994 WL 97758 at *4 (N.D.Ill. Mar. 24, 1994).
Although it does not address product design specifically, the Ninth Circuit's decision in Kendall-Jackson Winery, Ltd. v. E J Gallo Winery, 150 F.3d 1042 (9th Cir. 1998), is instructive because its definition of genericness is applicable in the product design context. Kendall-Jackson defines genericness in terms of both commonness and generality. Kendall-Jackson held that grape-leaf designs had become generic emblems for wine because grape leaves are so commonly used to decorate labels on wine bottles that the grape leaf alone "has lost the power to differentiate brands." Id. at 1049. In so holding, the Ninth Circuit quoted a First Circuit case defining "generic terms" as those that have "passed into common usage to identify a product," and a treatise stating that certain pictorial representations may be so "commonplace" that they cannot "creat[e] a separate commercial impression in a buyer's mind." Id. (quoting Boston Beer Co. v. Slesar Bros. Brewing Co., Inc., 9 F.3d 175, 180 (1st Cir. 1993), and 2 McCarthy § 7:36, at 7-62 (4th ed. 1997)). See also Leatherman, 199 F.3d at 1013 (noting that product designs "which are widely used and therefore in the public domain" may be "copied faithfully"); 2 McCarthy § 12:36.1 (explaining that Kendall-Jackson adopts Milstein's "view of the genericness of common business methods or styles and applie[s] it to commonplace product and packaging shapes").
Kendall-Jackson further stated that a "producer's depiction of a grape leaf may, however, be so distinctive as to warrant protection from copying." Id. By so stating, Kendall-Jackson made clear that the level of generality is relevant to the genericness determination. While the concept of a grape leaf on a wine label might be generic, a specific expression might not be.
Kendall-Jackson's definition of genericness also suggests that the traditional "what are you/who are you" test for genericness is applicable to product design. The Ninth Circuit has used the "what are you/who are you" test to determine whether a term is a generic name for a product or whether that term is used to identify the source of the product. See U.S. Jaycees v. San Francisco Junior Chamber of Commerce, 513 F.2d 1226 (9th Cir. 1975) (per curiam); 2 McCarthy § 12:1 ("A mark answers the buyer's questions `Who are you? Where do you come from?' . . . But the name of the product answers the question `What are you?'"). Under the "what are you/who are you" test, product design features are more likely to be generic than other forms of alleged trade dress. In Wal-Mart, the Supreme Court observed:
In the case of product design, as in the case of color, we think consumer predisposition to equate the feature with the source does not exist. Consumers are aware of the reality that, almost invariably, even the most unusual of product designs — such as a cocktail shaker shaped like a penguin — is intended not to identify the source, but to render the product itself more useful or more appealing.
Wal-Mart, 529 U.S. at 213 (emphasis added). Essentially, Wal-Mart explains that product design features, compared to other forms of trade dress, rarely answer the question "Who are you?".
Finally, when a product design consists entirely of a few design features that individually would be generic and unprotectable, the party asserting protectable trade dress must show that its combination of such elements is somehow "distinctive." See Kendall-Jackson, 150 F.3d at 1049 (noting that a "particular rendering of a grape leaf" might be source-identifying and therefore protectable, but that no "distinctive" feature of the plaintiff's rendering had been duplicated by the defendant); Indonesian Imports, 1999 WL 183629 at *7 (denying protection to a claimed trade dress consisting only of "common and basic features" because the producer "has not produced any evidence . . . that these generic and otherwise unprotectable features are combined together in a unique and source identifying way"). Requiring such a showing is consistent with the policy of limiting trademark protection for product designs. "Consumers should not be deprived of the benefits of competition with regard to the utilitarian and esthetic purposes that product design ordinarily serves." Wal-Mart, 529 U.S. at 213.
Cf. Leatherman, 199 F.3d at 1013 ("[T]rade dress must be viewed as a whole, but where the whole is nothing more than the assemblage of functional parts, . . . it is semantic trickery to say that there is still some sort of separate `overall appearance' which is non-functional.").
The Elements of the BIC Cookie Design Are Limited to Those Listed in Its Complaint.
The BIC Cookie design must be evaluated as a whole. See Clicks Billiards Inc. v. Sixshooters Inc., 251 F.2d 1252, 1259 (9th Cir. 2001); Rachel v. Banana Republic, Inc., 831 F.2d 1503, 1506 (9th Cir. 1987). However, a plaintiff seeking protection for a product design must articulate which elements make up its claimed trade dress. Without some articulation of those elements, it would be impossible for the court to evaluate secondary meaning, genericness, or nonfunctionality, and it would be impossible for the court to shape narrowly tailored relief. Yurman, 262 F.3d at 117. Moreover, without knowing what the protected design consists of, a successful plaintiff's competitors would not know what new designs would be subject to challenge. Id. See also Tie Tech, Inc. v. Kinedyne Corp., 296 F.3d 778, 785 (9th Cir. 2002) (listing those design elements that are disputedly functional); Clicks Billiards, Inc. v. Sixshooters Inc., 251 F.3d 1252, 1256-57 (9th Cir. 2001) (listing elements of asserted trade dress); Leatherman, 199 F.3d at 1011 (listing elements of asserted trade dress).At various points in this litigation, BIC has offered different versions of the elements that it claims constitute the protectable BIC Cookie design. The Complaint lists the following as elements of the asserted BIC Cookie trade dress: (1) the rectangular shape; (2) bull-nose corners; (3) the approximate size of the cookie, measured lengthwise and widthwise; and (4) the chocolate diagonal dip. BIC has also suggested in its papers and at hearings before this court that its trade dress includes (5) the fact that the BIC Cookie is a shortbread cookie (with or without macadamia nuts), (6) the "two curved longitudinal sides"; and (7) the fact that it is a "two-bite" cookie (a matter related to its size). See Hr'g Tr. (5/19/03). BIC at one point contended that its protectable trade dress included square cookies as well as rectangular cookies, but BIC has since abandoned that claim. See Hr'g Tr. (5/19/03); BIC's Supp. Resps. and Objections to Defs.' First Req. for Answers to Interrogs. Dated July 3, 2002, No. 17 (Ex. O to Meyer Dec., Defs.' CS (filed Dec. 24, 2002)).
Although this court's ultimate holding would be the same regardless of whether the latter three elements were considered as part of BIC's asserted trade dress, the court holds that none of those latter three elements is part of BIC's claim. First, allowing BIC to add elements to its listing of trade dress elements at this late stage would prejudice Cookie Corner. The Complaint clearly lists those design elements that Cookie Corner allegedly "copied" in making the Cookie Corner Cookie, and BIC did not give adequate notice that it intended to assert that elements (5), (6), and (7) were part of its claimed trade dress. That is particularly true of elements (6) and (7), as the "two curved longitudinal sides" of the BIC Cookie and the purported "two-bite" nature of the cookie were not even mentioned, much less asserted, as elements of the BIC Cookie design until oral argument on the present motions.
As noted above, there has already been one round of summary judgment motions on the merits of BIC's trade dress claim regarding the design of the BIC Cookie.
Second, it appears that BIC itself does not have a clear idea about which elements make up its trade dress. When deposed, Allen Ikawa ("Ikawa"), BIC's chief executive officer and president and the person designated to testify on BIC's behalf under Rule 30(b)(6) of the Federal Rules of Civil Procedure, stated that he "[hadn't] thought about" the elements that BIC claims as part of the BIC Cookie design. Ikawa Dep. at 225 (Ex. B to Pl.'s CS). Ikawa also testified that he did not know and was not sure which elements BIC was asserting as its trade dress. Id. at 308-10.
Third, with respect to element (5), counsel for BIC could not positively state whether or not the "shortbread" nature of the BIC Cookie was an element of its alleged trade dress. When questioned by the court, BIC's counsel ultimately appeared to claim that "shortbread" was one of the design elements of the BIC Cookie because "shortbread" allegedly has a certain appearance. BIC's counsel would not or could not state whether BIC was claiming that a non-shortbread cookie that appeared virtually identical to the BIC Cookie would be an infringement of BIC's alleged trade dress.
Counsel could only state that BIC's claim must depend on the fact that the BIC Cookie is a shortbread cookie because "that's what we make." This response did not clarify the elements of the asserted trade dress. It left open the possibility that, if BIC added a cookie to its repertoire that looked like the present shortbread cookie but was not a shortbread cookie, BIC could include the new type of cookie in its assertion of trade dress rights.
Finally, with respect to element (7), BIC has not defined what a "two-bite" cookie is. It apparently is a "relatively small cookie." BIC did not offer any evidence that the term "two-bite cookie" is an accepted or generally used term. When asked by the court whether a "three-bite cookie" might be challenged as using the BIC Cookie trade dress, counsel for BIC answered that it would not. However, when asked by the court whether a "one-and-a-half-bite cookie" would be challenged, counsel admitted that he could not answer because he did not know how large a "bite" was. Accordingly, element (7) does not offer a meaningful description of a product design feature. The court notes, however, that the approximate size of the cookie is included in the elements asserted in the Complaint, and the court does consider size as part of BIC's claimed trade dress in that respect.
Of course, the inclusion of the size element presents a question that is unanswered by BIC: How much larger or smaller would a competitor have to make a cookie to avoid a challenge? The Cookie Corner Cookie is longer and wider than the BIC Cookie by about one-quarter inch. Would BIC challenge a cookie that was one-half inch longer? BIC's failure to clarify the size element of its trade dress claim would lead to uncertainty for competitors if this court were to grant BIC judgment on this claim. "Competition is deterred . . . not merely by successful suit but by the plausible threat of successful suit. . . ." Wal-Mart, 529 U.S. at 214. For instance, at the hearing, BIC contended that the rectangular diagonally dipped shortbread cookie manufactured by Kelly's Kookies was "substantially larger" than the BIC Cookie and therefore would not infringe the BIC Cookie.
The BIC Cookie Design Is Generic and Therefore Unprotectable.
Based on the evidence submitted by the parties in support of the present motions, the court concludes that the BIC Cookie trade dress is generic. The BIC Cookie design is nothing more than a nondistinctive combination of a few basic, common design elements. Cookies are commonly rectangular and often have bull-nose corners, which naturally result from a basic and common method of making cookies. In particular, the rectangular shape is a classic shape for shortbread. Dipping cookies halfway in chocolate at various angles, including diagonally, is also common. In fact, Cookie Corner has provided evidence that shortbread that is half-dipped in chocolate is generically known as "royal shortbread," and BIC does not dispute that evidence.BIC has not provided any evidence that the generic elements of the asserted trade dress are combined in a unique, distinctive, or source-identifying way. By contrast, Cookie Corner has provided evidence that other rectangular diagonally dipped shortbread cookies are sold in the same markets as is the BIC Cookie. If the BIC Cookie's particular combination of these generic elements is not unique, and BIC offers no additional evidence that there is something distinctive about that combination or that consumers use those features as an indicator of source, then the court cannot conclude that that combination deserves protection.
As noted earlier in this court's discussion of the articulation of the BIC Cookie design elements, this court's finding of genericness is not affected by the inclusion or exclusion of "shortbread" as an element of the asserted trade dress. The record shows, for instance, that shortbread cookies dipped in chocolate are so common as to have a generic name ("royal shortbread") and that rectangular shapes for shortbread are traditional. Consequently, the addition of a "shortbread" design element would not make the BIC Cookie's combination of generic elements any more likely or able to serve primarily as an identifier of source.
Addressing the genericness issue has required a review of voluminous submissions. First, Cookie Corner has submitted the testimony of numerous purported experts with extensive training and experience in the baking and cookie industries. They state that the trade dress elements asserted by BIC are common and widespread. BIC contends that the testimony offered by Cookie Corner is "either inadmissible or entitled to very little probative weight." Pl.'s Opp. at 14. However, none of BIC's objections to the evidence offered by Cookie Corner is compelling.
BIC argues that some of the declarations offered by Cookie Corner are entitled to little weight because the declarants are not qualified to offer their opinions. For instance, BIC urges this court to reject the testimony of Wally Amos ("Amos") because he did not graduate from high school and because he does not have a degree in a culinary field. Contrary to BIC's suggestions, Rule 703 does not require that expert witnesses have formal degrees. Amos founded the "Famous Amos Cookie Company" in 1975, as well as a number of other companies. He states that he has read hundreds of cookie recipes and that, during his travels, he has met hundreds of bakers and cookie makers. Amos's work experience is extensive enough to qualify him as an expert in the cookie industry, whether or not he holds an academic degree. For purposes of Rule 56 and these motions, BIC's objections to the declarations of other experts offered by Cookie Corner are similarly without merit.
Cookie Corner has also submitted evidence that very similar cookies are commonly made and sold, both in Hawaii and elsewhere in the United States. Some of the cookies made by other producers are also rectangular diagonally dipped shortbread cookies. BIC has two primary arguments in response to this evidence: (1) only a few of the pictured cookies are rectangular and diagonally dipped, indicating that there are many alternate configurations; and (2) there is no evidence that any dipped shortbread cookie was made or sold before 1985, the date that BIC contends it first began manufacturing and selling the BIC Cookie. Neither argument is persuasive.
With respect to argument (1) (the multitude of alternative designs), the court recognizes that alternate configurations for shortbread cookies abound. Nevertheless, the BIC Cookie design consists entirely of design elements that are so basic that the primary purpose of their combination cannot be source identification. The most basic cookie shapes, e.g., circles, rectangles, and squares, do not become protectable just because cookies may be baked in the shape of pineapples, giraffes, or people. One of the reasons that genericness precludes protectability is that producers cannot monopolize basic product designs. The central question asked by the Ninth Circuit in Kendall-Jackson was whether the plaintiff was seeking protection for a feature or features that had the ability to identify source. If the BIC Cookie design's primary significance is not to identify a single source, no matter what that source might be, then it does not matter whether there are many other available cookie designs.
Even in a traditional functionality analysis, in which the availability of alternative designs is consistently considered a relevant factor by courts, the relevance of alternative designs is always affected by the cost of producing those designs. As Cookie Corner points out, it is true that competitors might be able to use various machines and techniques to produce an infinite variety of cookie shapes and dip patterns. However, BIC, which has the burden of proof with respect to genericness, does not even allege that the methods of producing the alternative designs it lists (such as pineapple shapes, heart shapes, and butterfly shapes) are as simple or basic as the methods used to produce traditional rectangular or bar cookies.
With respect to argument (2) (BIC's allegation that it was the first to make a cookie like the BIC Cookie), the court restricts its consideration to the period since BIC filed suit. Although BIC claims that it has marketed and sold the BIC Cookie since 1985, see Ikawa Decl. ¶ 3 (Ex. 3a to Pl.'s CS), BIC does not show that its cookie has remained the same ever since. Indeed, Allen Ikawa, BIC's president and founder, testified that the basic dip pattern has varied over time. Ikawa Dep. at 199 (Ex. B to Pl.'s CS). Ikawa was not able to state with certainty that the diagonal dip pattern now used on the BIC Cookie was the same as that used when BIC first began making the BIC Cookie. Id. at 200. Ikawa also testified that the dimensions of the BIC Cookie changed over time, although he did not know how the dimensions had changed. Id. at 196. Therefore, BIC may have been producing some version of the BIC Cookie since 1985, but there is no evidence that, since 1985, BIC has continuously sold the version of the BIC Cookie for which it now seeks trade dress protection.
In opposition to Cookie Corner experts' declarations as to the commonness of cookies like the BIC Cookie, BIC offers the opinion of John Olejko ("Olejko"). Olejko sold cookie-making machines to both BIC and Cookie Corner and has extensive experience in the cookie industry. Olejko testified that he could not "positively remember or recall" ever having seen "a rectangular-shaped shortbread cookie diagonally dipped in chocolate." Pl.'s Opp. at 28-29. This statement provides little support for BIC's argument that rectangular shortbread cookies diagonally dipped in chocolate are not commonplace. If various types of chocolate-dipped shortbread cookies are extremely common, it is also possible that one would not "positively remember" a rectangular shortbread cookie dipped diagonally in chocolate in light of its unremarkability as a combination of entirely generic elements. Moreover, Olejko's lack of recollection does not negate the affirmative recollections of others. BIC does not even suggest that Cookie Corner's witnesses are fabricating evidence. This is therefore not a credibility issue that must be left for resolution at trial.
In addition to the testimony of Olejko, BIC offers the results of a consumer survey that it commissioned (the "BIC Survey"). BIC argues that the survey shows that the BIC Cookie design answers the question "who are you" rather than the question "what are you." BIC argues: "16.9-29.4% of respondents who viewed [the BIC Cookie] answered the `who are you?' question by identifying BIC. This level is . . . more than sufficient to show that the dress is not generic." Pl.'s Opp. at 19. BIC's Opposition indicates that it believes that the survey data is the strongest evidence in the record for a conclusion of nongenericness.
In the BIC Survey, conducted on Oahu, respondents were first shown a picture of a BIC Cookie wrapped in cellophane printed with the word "Big Island Candies." They were then asked if they had seen the BIC Cookie before. If they stated that they had, they were asked, "Who makes this product?" In response to that question, 16.9% to 29.4% of respondents answered "BIC."
More precisely, 6.5% of respondents answered "Big Island Candies," while 10.4% gave some variation of "Big Island Candies," such as "Big Island Cookies." Others could not name the company but identified the maker of the BIC Cookie as a company located on the Big Island (9.0%) or in Hilo (3.5%).
It is true that relevant survey data or statistics of some sort might assist this court. See 2 McCarthy § 12:14 ("Consumer surveys have become almost de rigeur in litigation over genericness. Judges . . . often expect to receive evidentiary assistance by surveys in resolving generic disputes."). As BIC contends, "the most important type of evidence used in evaluating genericness is consumer surveys." Pl.'s Opp. at 19.
Unfortunately for BIC, the BIC Survey is of no help to it on the present motion. First, the BIC Survey does not test for genericness. The survey asked the "Who are you?" question. That method cannot possibly test whether respondents view the BIC Cookie as answering the question "Who are you?" (rather than answering the question "What are you?"). The question is not which source consumers most often associate with the BIC Cookie, but whether the BIC Cookie design's primary significance to the consumer is that of source identification. In fact, James Dannemiller ("Dannemiller"), the president of the firm that conducted the BIC Survey and the apparent author of the BIC Survey questions, testified that the BIC Survey did not test for and was not intended to test for genericness.
Dannemiller also testified that the BIC Survey did not test for secondary meaning or likelihood of confusion.
Second, no attempt was made to test consumers' perceptions of the BIC Cookie design rather than the BIC Cookie packaging (or the combination of the BIC Cookie design and packaging). Survey respondents were never shown a plain BIC Cookie. Rather, the only way in which respondents saw the BIC Cookie was when it was wrapped in cellophane printed with the words "Big Island Candies." The words "Big Island Candies" were clearly legible on the wrapper. In fact, more than one respondent stated that they "recognized" the BIC Cookie because they read the name of the producer on the wrapper.
Moreover, the picture of the BIC Cookie was only one of a series of pictures shown to each respondent in random order. The other pictures included pictures of the outside of the Big Island Candies box, which has "Big Island Candies" prominently written on it. Some respondents may have been shown the BIC box before they were shown the BIC Cookie.
Finally, the court notes that BIC misunderstands what percentage of respondents must respond in a certain way to establish nongenericness. See Pl.'s Opp. at 19. BIC suggests that, if more than 15% of survey respondents identified BIC after viewing the BIC Cookie, that percentage was "sufficient to support a summary judgment finding of distinctiveness, so it is more than sufficient to show that the dress is not generic." Id. That argument is incorrect because (1) the type of survey used to test for genericness differs from the type of survey used to test for distinctiveness, and (2) for a genericness survey, "majority usage controls." 2 McCarthy § 12:6. For a term to be held nongeneric, the seller must prove that the "primary significance of the term in the minds of the consuming public is not the product but the producer." Kellogg Co. v. Nat'l Biscuit Co., 305 U.S. 111, 119 (1938) (emphasis added). In contrast, there is no such requirement of "primary significance" with respect to secondary meaning. Therefore, BIC's reliance on Grupo Gigante S.A. de C.V. v. Dallo Co., Inc., 199 F. Supp.2d 1083 (C.D.Cal. 2000), which involves a secondary meaning survey, is misguided. There was no allegation in Grupo Gigante that the mark at issue was generic.
Before concluding, the court addresses BIC's argument that evidence of secondary meaning supports a finding that the BIC Cookie design is not generic. Opp. at 32-43. The court does not find that section of BIC's Opposition persuasive. First, BIC cites no legal authority in support of its claim that evidence of secondary meaning is relevant to the determination of genericness. Opp. at 31-32. In cases involving generic terms, it is clear that "no amount of evidence of purported secondary meaning can give legal protection to that generic name." 2 McCarthy § 15:24.
For that reason, any buyer association between a generic name and a single producer is called "de facto secondary meaning."
Second, although certain types of evidence might be relevant to both the issues of secondary meaning and genericness, the evidence that BIC does cite is not relevant. The only kind of evidence that is relevant to the determination of genericness is evidence showing that "a majority of buyers recognize the term as a trademark, not as the name of the product." Id. Here, as discussed above, BIC has not presented any evidence that a majority of buyers recognize the alleged BIC Cookie design as identifying BIC or any other single source. See Opp. at 33-34 (arguing that survey is relevant to determination of genericness). The money that BIC has spent in marketing and advertising and the publicity that BIC has received, see Opp. at 35-37, are clearly irrelevant to the present motions, as is evidence of Cookie Corner's alleged intentional copying of the BIC Cookie design and packaging.
As discussed above, survey respondents were never shown the BIC Cookie without its cellophane wrapping, which has "Big Island Candies" printed over the entire surface area. Moreover, the BIC Survey did not attempt to determine whether or not the BIC Cookie (even in its wrapping) served a source-identifying function. "Who makes this cookie?" was the survey question. This question is obviously problematic because it limits the respondent to identifying one maker.
Even if secondary meaning were relevant to the determination of genericness, as BIC argues, see Opp. at 31-32, the BIC Survey is still of no help to BIC because it tests for neither genericness nor secondary meaning. See, e.g., Dannemiller Dep. at 28, 30-31.
BIC also contends that it has substantially exclusively and continuously used the BIC Cookie design for eighteen years, see Opp. at 32-33, but, as discussed above, BIC has admitted that the design has changed during that time.
Moreover, with respect to the final portion of BIC's Opposition regarding Cookie Corner's alleged intentional copying, see Opp. at 37-43, evidence of intentional copying offers no support for BIC's argument that the BIC Cookie design is not generic. For instance, Cookie Corner could have copied the BIC Cookie for multiple reasons, including merely that it knew that this particular generic cookie design was a common one and popular with customers.
To sum up, BIC, which has the burden of showing nongenericness, has not met its burden. BIC offers only the BIC Survey, which is irrelevant to the issue of genericness, and the testimony of Olejko, which does not negate the testimony by persons who commonly saw cookies similar to the BIC Cookie. Meanwhile, Cookie Corner has offered evidence from various experts in the cookie industry that the BIC Cookie design is common and basic, as well as evidence that there are numerous similar cookies being sold in the same market. Even construing all the evidence in the light most favorable to BIC, the court concludes that BIC fails to meet its burden of proving the nongenericness of the BIC Cookie design.
CONCLUSION.
For the foregoing reasons, Cookie Corner's motion for partial summary judgment is granted, and judgment is granted to Cookie Corner on all claims with respect to the design of the BIC Cookie. BIC's cross-motion for partial summary judgment is denied. The claims remaining for adjudication are those relating to the packaging of the BIC Cookie.
IT IS SO ORDERED.