I’m not terribly surprised the court of appeals dodged the functionality issue by holding that Louboutin’s trademark rights were limited to contrasting colors, but it’s notable just how many contradictions the court has to swallow in doing so. In particular, the decision apparently resurrects the concept of “use as a mark,” but not in any explicitly defined or justified way.
The sleight of hand starts with the question presented, as the court frames it: “whether a single color may serve as a legally protected trademark in the fashion industry and, in particular, as the mark for a particular style of high fashion women’s footwear.” Formally, the court’s answer is yes, it can—except not here! Because, we will discover, Louboutin doesn’t have a valid registration for a single color; it has a valid registration for a color contrast. Qualitex, on which the court of appeals relies to hold that single-color trademarks must be allowed, is thus somewhat beside the point.
Louboutin’s evidence only showed that the red sole had limited secondary meaning as a contrasting color. Thus, under 15 U.S.C. § 1119, the court of appeals limited the registration (and thus the mark) to uses in which the red outsole contrasts with the remainder of the shoe. (The court of appeals held that the district court had properly accorded the registration its presumption of validity, but then found that YSL had proved functionality. One might ask whether the court of appeals did the same—since that court supposedly refused to resolve functionality, on what ground did it override the presumption of validity of the unmodified registration?)
A further step was required: YSL’s monochrome red shoe wasn’t a “use” of this modified trademark, and therefore the court didn’t need to consider whether YSL’s red monochrome shoes were likely to cause confusion or whether the remaining Louboutin mark was functional. Thus, the denial of the preliminary injunction was affirmed, but the apparent invalidation of Louboutin’s entire mark was reversed, and the case was remanded for further proceedings on YSL’s counterclaims.
There’s a summary of the history of color marks (approved!) and of functionality doctrine (for competition! Citing Mark McKenna!), neither of which officially matter because of the outcome, though they’re clearly driving the analysis.
How the Second Circuit sees aesthetic functionality: if there’s Traffix functionality, there’s no need to inquire further. But product features can also be functional if giving the claimant an exclusive right would put competitors at a significant non-reputation-related disadvantage. The first step is to ask the more specific questions about whether a feature is either “essential to the use or purpose” or “affects the cost or quality” of the product at issue. If necessary, the court then asks the more general competition question: “if the design feature is not ‘functional’ from a traditional perspective, it must still pass the fact-intensive Qualitex test and be shown not to have a significant effect on competition in order to receive trademark protection” (emphasis added). Note the potential effect of this formulation on the burden of a plaintiff claiming rights in unregistered trade dress, who must show nonfunctionality.
The Second Circuit fully accepts aesthetic functionality, but doesn’t ask whether the feature was an “important ingredient” in a product’s success, since that test “inevitably penalized markholders for their success in promoting their product.” Instead, the question is whether protection would “significantly limit the range of competitive designs available.” Aesthetic functionality bars protection of a mark that is “necessary to compete in the [relevant] market.” Traffix likewise requires courts to “inquire” as to whether recognizing the trademark “would put competitors at a significant non-reputation-related disadvantage.” (The court got in a dig at the Fifth Circuit’s rejection of the doctrine; the Second Circuit believes the Supreme Court has validated its approach.)
Courts must carefully weigh “the competitive benefits of protecting the source-identifying aspects” of a mark against the “competitive costs of precluding competitors from using the feature,” especially since it can sometimes be difficult to distinguish source indication from aesthetic function. This is a highly fact-specific analysis that must take into account both a claimaint’s “right to enjoy the benefits of its effort to distinguish its product” and the public’s “right to the ‘vigorously competitive market’ protected by the Lanham Act, which an overly broad trademark might hinder.” Denoting a desirable source doesn’t make a feature functional.
With that background, the district court erred in adopting a per se rule that a single color can’t be a mark in the fashion industry. “Qualitex requires an individualized, fact-based inquiry into the nature of the trademark, and cannot be read to sanction an industry-based per se rule.” Nonetheless, it's true that fashion has special trademark concerns, and that color can serve as a fashion design tool rather than as mere ornamentation. Still, a competitor isn’t guaranteed the greatest possible range, only the ability to fairly compete in a given market. (By contrast, a footnote stated, copyright does allow creativity and originality to interfere with the rights of an existing owner, in cases of independent creation. Trademark aims to prevent consumer confusion “even at the expense of a manufacturer’s creativity.”)
The court also dropped a footnote about efforts to increase US protection for fashion design, citing various academics (hi, Chris Sprigman!). Unfortunately, the footnote didn’t stop there:
It is arguable that, in the particular circumstances of this case, the more appropriate vehicle for the protection of the Red Sole Mark would have been copyright rather than trademark. See generally Kieselstein-Cord v. Accessories by Pearl, Inc., 632 F.2d 989, 993–94 (2d Cir. 1980) (addressing the broad issue of aesthetically functional copyrights and holding that decorative belt buckles that were used principally for ornamentation could be copyrighted because the primary ornamental aspect of the buckles was conceptually separate from their subsidiary utilitarian function); Laura A. Heymann, The Trademark/Copyright Divide, 60 SMU L. Rev. 55 (2007).
Ed. rant: Oh, for pete’s sake. This is a ridiculous footnote inserted without any consideration of the actual effect on the namechecked doctrine or any opportunity for the parties to point out what’s dumb about this statement. The idea that a color or even a color contrast could make a utilitarian article copyrightable (because that’s what has to be contemplated in order to make this aside relevant—not the shape of the shoe or the shape of the sole, though claims to those would be wrong too) is nuttier than anything the district court or either of the parties claimed. I trust that Louboutin won’t burden the Copyright Office with rejecting an attempt to register a work that consists of the red sole of a shoe, contrasting or not.
Hey, look, there’s actual circuit precedent! Boisson v. Banian, Ltd., 273 F.3d 262, 270, 271 (2d Cir. 2001) (“color by itself is not subject to copyright protection”; “a particular color is not copyrightable”); Streetwise Maps, Inc. v. VanDam, Inc., 159 F.3d 739 (2d Cir. 1998) (individual colors shouldn’t be viewed as protectable elements, but must be combined with other expressive choices). Also, the Copyright Office agrees, as do courts to whom the issue was actually presented. See 37 C.F.R. § 202.1 (“[E]xamples of works not subject to copyright [include] mere variations of … coloring” ); U.S. Copyright Office, Compendium II, Copyright Office Practices § 503.02(a) (1984) (“[M]ere coloration cannot support a copyright even though it may enhance the aesthetic appeal or commercial value of a work. For example, it is not possible to copyright a new version of a textile design merely because the colors of red and blue appearing in the design have been replaced with green and yellow, respectively.”); see also Century Tile, Inc. v. Hirsch Glass Co., 467 Fed.Appx. 651 (9th Cir. 2012) (coloring and color blends are uncopyrightable elements); Baby Buddies, Inc. v. Toys R Us, Inc., 611 F.3d 1308 (11th Cir. 2010) (color of useful article not protectable); Amsterdam v. Triangle Publications, 189 F.2d 104 (3d Cir. 1951) (“Nor can the plaintiff copyright the arbitrary color schemes, symbols or numbers that he uses on his map.”); Banzai, Inc. v. Broder Bros. Co., 2009 WL 1285518 (E.D. Pa. 2009) (simple color combinations in tie-dye pattern not copyrightable).
Note that “the red and the black” are already a well-known color combination; the decision to put it on a shoe, even if that choice had never been made before, would not constitute copyrightable creativity any more than the choice to use red, white, and blue on a tie-dye pattern was copyrightable even if it had never been done before. And "red plus non-red contrast color" is even worse as a copyright claim. Frankly, I expect better from the Second Circuit.
Now the court turned to the red sole mark. The first question was whether the mark was protectable; functionality was an affirmative defense for the second stage of the inquiry (infringement). This sequence is odd in the ordinary case—whether a claimed feature is functional surely affects whether it’s protectable. There may be cases where the defendant’s use is functional, or the features that provide the similarity between the parties’ designs are functional even if the overall appearance of the plaintiff’s product is nonfunctional, and in those cases the court’s sequence makes sense. I think the case at bar is an example of defendant-side functionality, but I wouldn’t therefore say that functionality is always a stage-two issue.
Anyway, protectability: registration is prima facie evidence of validity. The district court erred in finding that YSL rebutted that presumption by showing that a single color can never be a valid trademark in the fashion industry. (And that’s the last you’ll hear of the presumption!) The court of appeals found that the red outsole contrasting with the remainder of the upper had acquired secondary meaning.
The record here contained sufficient undisputed facts to resolve the question of distinctiveness. The lacquered red outsole applied to a contrasting upper was a mark, but “the record fails to demonstrate that the secondary meaning of the Red Sole Mark extends to uses in which the sole does not contrast with the upper―in other words, when a red sole is used on a monochromatic red shoe. It is the contrast that “causes the sole to ‘pop,’ and to distinguish its creator.” The record showed a close association with contrast in particular. Of hundreds of pictures of Louboutin shoes in the record, only four were monochrome red. “And Louboutin’s own consumer surveys show that when consumers were shown the YSL monochrome red shoe, of those consumers who misidentified the pictured shoes as Louboutin-made, nearly every one cited the red sole of the shoe, rather than its general red color.” Thus, Louboutin hadn’t established secondary meaning in a red sole for a red shoe. So, the court directed the PTO to limit the registration appropriately; as modified, the mark was valid and enforceable.
Now the magic happens: “This holding disposes of the Lanham Act claims brought by both Louboutin and YSL because the red sole on YSL’s monochrome shoes is neither a use of, nor confusingly similar to, the Red Sole Mark…. Louboutin could not have shown a likelihood of success on the merits in the absence of an infringing use of the Red Sole Mark by YSL” (emphasis added).
Comment: use as a mark, we just can’t quit you. How does the court of appeals know that the YSL shoes aren’t confusingly similar to the mark? In the ordinary case, trademark rights extend to marks that, while not identical to the mark owner’s use, are confusingly similar thereto: you show infringement by showing validity plus confusion. Regardless of the merits of its survey, Louboutin purported to be ready to do part two of that, and the court just held it won on part one. So why doesn’t it get to go back and try again? Mark McKenna has justly criticized the circularity of trademark use, which does move in both directions (here, and in Rogers v. Grimaldi and New Kids as well, where the courts want to have their cake and eat it too by simply declaring that consumer protection doesn’t conflict with free speech).
In the end, the court held, because YSL’s red sole wasn’t a “use” of Louboutin’s mark, “it is axiomatic that we need not―and should not―address either the likelihood of consumer confusion or whether the modified Mark is functional.” But of course the court did address likely confusion, albeit in the most cursory fashion; and the functionality of monochrome use has to be the reason why YSL’s use isn’t “use as a mark”/“use of the mark.” This opinion isn’t worse than many important appellate cases about the proper scope of trademark protection in punting on these issues and shoving important normative and empirical questions into ipse dixits like “it’s not a use,” but it’s disappointing to see nonetheless.