Samara Bro's v. Wal-Mart Stores, Inc.

JON 0. NEWMAN, Circuit Judge,

dissenting in part:

On two key issues in this case, I respectfully dissent. First, I do not believe that the “look” of Samara’s line of dresses constituted protectable trade dress, and the Court’s upholding of such protection lowers, at least for this case, the standards we have previously established for trade dress protection of a product “line.” Second, I do not believe that Samara is entitled to a copyright on the five designs that depict, in an entirely unoriginal way, such common items as strawberries, tulips, daisies, and hearts. Indeed, if the majority’s copyright ruling were to be applied beyond the precise facts of this case, which I do not expect will ever happen, designers of Valentine’s Day cards would be at risk of copyright infringement if, having seen a Samara dress with the copyrighted designs, they place unadorned hearts on their cards, or at least show little girls in dresses with unadorned hearts on their collars.

I. Trade Dress

This Circuit has clearly stated the principles that govern a claim c? trade dress protection for product design, particularly protection claimed for the design of a line of products. First, when trade dress protection is claimed for product design or configuration, rather than product packaging, the inquiry is “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). Second, product design or configuration is far less likely to be an indicator of source than product packaging, and, indeed, we “appear[] to be moving toward a rule that packaging is usually indicative of a product’s source, while the design or con-figuration of the product is usually not so.” Id. at 379. Third, we must construe the Lanham Act “in the light of a strong federal policy in favor of vigorously competitive markets.” Id. Fourth, courts are to “exercise[ ] particular ‘caution’ when extending [trademark] protection to product designs.” Id. at 380 (quoting Jeffrey Milstein, Inc. v. Greger, Lawlor, Roth, Inc., 58 F.3d 27, 32 (2d Cir.1995)). Fifth, “ ‘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.’” Id. (quoting Jeffrey Milstein, 58 F.3d at 32). Sixth, and most pertinent to this case, a plaintiff “seeking to protect a series or line of products faces a particularly difficult challenge of showing that the appearance of its several products is sufficiently distinct and unique to merit protection as a recognizable trade dress.... Accordingly, when protection is sought for an entire line of products, our concern for protecting competition is acute." Id. (emphasis added).

This is not to say that product design or configuration can never achieve trade dress protection for a line. But until this case, we have permitted such protection only in very extreme circumstances. For example, in Harlequin Enterprises Ltd. v. Gulf & Western Corp., 644 F.2d 946, 948 (2d Cir.1981), we upheld protection for the look of a line of books where “every book in the Harlequin Presents series is uniform in its dimensions, cover design and colophon.”

On the facts of this case, it was not reasonable to find distinctiveness of Samara’s entire product line. The Court upholds the jury’s verdict by recounting five factors that Samara’s designer, Kathy Gosda, testified were present in the Samara line.1 These are (1) seersucker fabric; (2) large bold appliques; (3) large collars with the appliques generally integrated into the collar and the pockets, if any; (4) absence of printed images, black outlines, alphanumeric characters, or three-dimensional features; and (5) full-cut, one-piece conservative bodies.

For several reasons, these five factors cannot be permitted to constitute a protectable *134trade dress “look.” First, the fifth factor— full-cut, one piece conservative bodies — is not a fabric design feature at all, but is a dress design. We do not extend copyright protection to dress designs, see Whimsicality, Inc. v. Rubie’s Costume Co., 891 F.2d 452, 455 (2d Cir.1989); 1 Nimmer on Copyright § 2.08[H][3] (1998) [hereinafter “Nimmer”], and we should not let dress designers secure protection for their dress styles by claiming a “look” protectable as trade dress under trademark law. Though a particularly distinctive dress design feature, combined with distinctive fabric design elements, might suffice for trade dress protection, Samara’s dress design of full-cut, one-piece conservative bodies is far too common to serve as an element of a “look” that warrants trade dress protection.

Second, the fourth factor — the absence of various fabric design elements — is totally unavailing. There are hundreds, if not thousands, of fabric design elements that Samara does not use. A “look” warranting trade dress protection must consist of the design features that are present in the line of clothing claimed to have a look, not in the myriad features that the designer elected not to use.

Third, with factors 4 and 5 eliminated, or, at least counted for little, we are left with three factors that are far too general to constitute a protectable “look.” Seersucker fabric is not at all distinctive. Large bold appliques are a common design feature, used by numerous dress designers. Moreover, Samara uses such different items as appliques that a unified “look” cannot be reasonably discerned. With four products, the applique consists of flowers; with six, it consists of fruits; with one, it consists of hearts; with two, it consists of vehicles; and with one, it consists of sailboats. The use of large collars is also extremely common, and the fact that the appliques are placed on the collar, or pockets if any, is hardly noteworthy. I cannot tell what Ms. Gosda meant by saying that the appliques are “integrated” into the collar. From the photographs in the record, the appliques appear simply to be sewn onto the collars, sometimes extending beyond the collar to the body of the dress. None of the factors identified by Ms. Gosda, nor the combination of them, can reasonably be found to constitute protecta-ble trade dress of a product line. Indeed, since the appliques depict such different objects from one garment to another, there is no one combination of design elements to assess. Samara’s dresses constitute a “line” of clothing, just like those of many manufacturers who sell a “line” of somewhat similarly designed or decorated clothing, but they cannot reasonably be found to have such an identifiable “look” that the entire product line warrants trade dress protection.

Fourth, a more fundamental flaw in the Court’s protection of the Samara “look” is that many of the products in the alleged product line do not have all of even the few features that are claimed to constitute the “look.” Of the sixteen products claimed to have been copied by Wal-Mart, the Court acknowledges that two do not merit trade dress protection. Of the remaining fourteen, four do not have the applique integrated on the collar or pockets; in two products, the applique is placed on the top front of the main fabric of the dress, and in two products, on a mid-riff running across the main fabric of the dress. How can there be a product line with a protectable trade dress “look” when nearly one-third of the products claimed to constitute the infringed line lack a key feature alleged to be an ingredient of the “look”?

Moreover, Samara’s complaint alleged that its look included “specially designed stylized appliques selected for consistency across the line,” Complaint ¶ 14 (emphasis added), yet the evidence discloses that the appliques depicted a wide variety of objects including flowers, fruits, hearts, trucks, and boats.2 Surely there was no “consistency across the line” as to the objects depicted in the appliques. The complaint also alleged that the “look” included “carefully coordinated combinations of silhouettes, applique designs, fabric patterns and colors.” Id. (emphasis add*135ed). The variety of objects depicted totally refutes any claim of “consistency” as to “silhouettes,” “designs,” or “patterns,” and there is no consistency at all as to “colors,” as the differing combinations set out in the margin indicate.3 Where courts have upheld a trade dress that included color as a component, they have been careful to emphasize a “combination of particular hues of [plaintiffs red and yellow] colors,” Chevron Chemical Co. v. Voluntary Purchasing Groups, Inc., 659 F.2d 695, 703 (5th Cir. Unit A Oct.1981), and in Chevron Chemical the “particular hues” of the same two colors were “arranged in certain geometric designs, presented in conjunction with a particular style of printing,” id. No case has upheld trade dress protection for an alleged “line” of products that depicts entirely different decorative objects, rendered in a wide variety of combinations of different colors.

Despite the very strict standards this Circuit has articulated for trade dress protection of the design of a product “line,” and despite the very weak nature of the plaintiffs evidence, the Court upholds trade dress protection in this case essentially because the jury found in plaintiffs favor. That reliance, unjustified in my view, requires consideration of the proper approach to a jury’s verdict on an issue such as trade dress protection.

I agree with the Court when it rejects Wal-Mart’s contention that reviewing courts should apply a “lower level of deference,” 165 F.3d at 124, to the jury’s finding that Samara’s trade dress was sufficiently distinctive to be protectable. In my view, however, “level of deference” is the wrong dimension to assess. Every jury verdict is entitled to a fairly high degree of deference regardless of the nature of the issue decided. To state that proposition, however, does not preclude recognition of the principle that the boundaries within which an issue is reasonably a fact issue for a jury are narrower in some contexts than in others. When juries are called upon to resolve conflicting versions of historical facts, the boundaries are fairly wide. Reviewing courts take a generous view of what a reasonable jury could conclude with respect to historical facts, such as what happened at the scene of an accident. But the broad range of reasonable fact-finding as to historical facts and familiar legal issues like negligence is not appropriate for issues infused with technical legal meaning, issues with which juries are almost completely unfamiliar.

When reviewing courts consider jury verdicts on such issues as the reasonableness of a challenged restraint of trade or the protect-ability of a trade dress, I believe reviewing courts should recognize that the boundaries in which jury fact-finding may permissibly occur are somewhat narrow. It is not. a matter of giving less deference to a jury’s fact-finding, reached within an allowable scope. Rather, it is a matter of recognizing that the scope of allowable fact-finding is narrower. When the issue is negligence, many combinations of fact patterns can reasonably be found in the typical contested case, and juries have wide latitude to determine what they think the facts were and thereby to find either that there was negligence or there was not.4 But on more esoteric issues, courts should recognize that there are fewer combinations of fact patterns that could reasonably be found in a given case, and should therefore somewhat narrow the range in which jury fact-finding is permissible. Otherwise, we are ceding to juries *136broad authority to determine the substantive scope of the law on topics such as antitrust, copyright, trademark, and, in this case, trade dress.5

Our Circuit has recognized this important point. As we said with respect to the similar issue of the scope of fact issues in copyright cases, “Courts have an important responsibility in copyright cases to monitor the outer limits within which juries may determine reasonably disputed issues of fact. If a ease lies beyond those limits, the contrary view ... of a particular jury[] cannot be permitted to enlarge ... the scope of statutory protection enjoyed by a copyright proprietor.” Warner Bros., Inc. v. American Broadcasting Companies, Inc., 720 F.2d 231, 245 (2d Cir.1983).

Juries cannot realistically be expected to appreciate, much less to apply rigorously, the several precepts that our Circuit has said are applicable to resolution of a claim of distinctiveness as to product configuration. Indeed, any jury that finds, as this one did, that copying, sufficient to constitute infringement of a valid copyright, has occurred, will be likely also to find a Lanham Act violation, yet such a jury, in all likelihood, will have no appreciation of the subtle point that copyright accords a rather narrow monopoly that protects only against copying of particular items, whei’eas trademark grants a much broader monopoly that protects against marketing of any product with an appearance likely to cause confusion as to source.

The high bar to trade dress protection for a product line that this Court so emphatically raised in Landscape Forms, Inc. v. Columbia Cascade Co. 113 F.3d 373 (2d Cir.1997), and Knitwaves, Inc. v. Lollytogs, Ltd., 71 F.3d 996 (2d Cir.1995), has been unjustifiably lowered in this case. Though I suspect that this is an aberrational result, not likely to be repeated, I cannot assent to it.

II. Copyright

Wal-Mart challenges copyright validity as to six of Samara’s thirteen copyrights. Validity is challenged on the ground that these copyrights display common items in a manner that fails to satisfy even the “minimal degree of creativity” necessary to constitute an original work of authorship. See Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340, 345, 111 S.Ct. 1282, 113 L.Ed.2d 358 (1991) (citation omitted); 1 Nimmer § 2.01[B], at 2-14. The Court rejects Wal-Mart’s challenges by stating that the prima facie evidence of validity arising from Samara’s certificates of copyright registration, see 17 U.S.C. § 410(c), has not been overcome by any evidence. The Court points out that Wal-Mart “failed to adduce evidence to show that the works were not ‘independently created by its author, and not copied from someone else’s work.’ ” 165 F.3d at 132 (quoting Folio Impressions, Inc. v. Byer California, 937 F.2d 759, 764 (2d Cir.1991)). But Wal-Mart is not attacking Samara’s copyrights on the six challenged designs on the ground that Samara did not independently create the designs. The claim is that the designs lack minimal originality. No matter how “independently” a designer works in the confines of some remote atelier, if the resulting design is just the unadorned outline of three hearts, no valid copyright may be obtained.

On the disputed issue of whether the challenged copyrights lack minimal originality, Wal-Mart produced evidence in the form of exhibits depicting hearts, flowers, and strawberries in the ordinary shape and design *137displayed in Samara’s copyrights. Though I would have thought a court could take judicial notice of the fact that the outline of an unadorned heart is sufficiently common to lack minimal originality, Wal-Mart’s exhibits are ample evidence of this fact.

Of the six challenged copyrights, I would uphold the challenges to five. Registration No. 751-722 is just three hearts, grouped in a slight arc; the only arguable “arrangement” is that the tips of the hearts extend over the collar. Registration No. 751-718 and 751-720 are just three strawberries, connected by a slightly twisted vine. There is nothing minimally original about the strawberries, and strawberries with similarly slightly twisted vines are shown in an exhibit of McCall’s sewing patterns. Registration No. 751-721 is just three “daisies” separated by two leaves. The “daisy” design, which does not look like a daisy at all, is an utterly common five-sided pattern with rounded edges and a circular center, depicting simply a generic flower. Registration No. 751-725 is three tulips, each with two leaves. Both the tulips and the leaves are totally lacking in originality. I agree that Registration 751-719 reaches the minimal level of originality, since the grouping of the strawberries overflowing the pocket suffices as a minimally original arrangement.

The Copyright Office says that a fleur-de-lys design is not original enough. See Compendium II: Compendium of Copyright Office Practices § 503.02(a) (1984) (“[Rjegistration cannot be based upon the simplicity of standard ornamentation such as ... the attractiveness of a conventional fleur-de-lys de-sign_”). Basic depictions of hearts, daisies, strawberries, and tulips should fare no better. As this Court observed in invalidating the copyright on an Uncle Sam bank, “Here as elsewhere in the copyright law there are lines that must be drawn even though reasonable [persons] may differ where.” L. Batlin & Son, Inc. v. Snyder, 536 F.2d 486, 492 (2d Cir.1976) (in banc).

Though I suspect that the Court’s trade dress and copyright rulings will prove to be limited to the precise facts of this case, I dissent to the extent indicated in this opinion, and concur on all other issues.

. Though the Court excludes from the protected "line" of products those depicting a circus truck, an airplane, and a sail boat, it leaves within the protectcd "line” products that display three sailboats, three cars, and three trucks.

. With one, the colors are pink, green, and blue; with three, they are red, white, and blue; with one, they are red, green, and blue; with one, they are red, white, green, and blue; with three, they are pink, white, and blue; with one, they are pink and blue; with one, they are pink, yellow, and blue; with one, they are red, yellow, and blue; with one they are red, yellow, blue, and green; with one they are red, yellow, blue, green, and white; with one they are red, white, green, and black; and with one they are red, green, blue, yellow, and white.

. As I have pointed out elsewhere, this application of a negligence standard to a fact pattern, once determined, is really an issue of law, but because there are so many possible fact patterns in a typical negligence dispute, courts do not ask juries to find just the facts, but instead instruct them on applicable principles, of negligence and then ask them to apply the law to the facts as they find them. See Antilles Steamship Co., Ltd. v. Members of American Hull Insurance Syndicate, 733 F.2d 195, 206 & nn. 4, 5 (2d Cir.1984) (Newman, J., concurring). This is a matter more of convenience than doctrinal purity.

. Unlike the Court, 165 F.3d at 124 n. 1, I derive no reassurance at all from the fact that in a particular case a jury found in favor of a plaintiff on a copyright claim and against the plaintiff on a trade dress claim. See KeyStone Retaining Wall Systems, Inc. v. Westrock, Inc., 997 F.2d 1444, 1447 & n. 2 (Fed.Cir.1993). The Court thinks this is some evidence that juries can be counted on to understand the differences between the Lanham Act and the Copyright Act. If that were so, then the many cases in which juries found for plaintiffs on both claims or found against plaintiffs on both claims would be some evidence that juries cannot distinguish the relevant concepts. Moreover, a jury's differing outcomes on different claims in particular cases are as likely to reflect a compromise as a sophisticated understanding of the relevant principles. My view that juries have a difficult time understanding the principles of such unfamiliar fields of law as copyright and trademark is based on the fact that judges, including myself, have difficulty fully grasping the subtleties of these doctrines, even though we deal with them far more often than do juries.