9th Circuit shows some hostility to functionality for product design, gets federal fame right

Blumenthal Distributing, Inc. v. Herman Miller, Inc., Nos. 18-56471 & 18-56493 (9th Cir. Jun. 25, 2020) 

HM and Blumenthal/OSP engaged in litigation over “knockoffs” of HM’s Eames chairs and Aeron chairs. A jury found that HM’s registered and unregistered claimed EAMES trade dresses were protectable, and that OSP willfully infringed and diluted them, resulting in an award of $3,378,966 in infringement damages and $3,000,000 in dilution damages, and OSP was enjoined. The jury also found that HM’s registered and unregistered claimed AERON trade dresses were unprotectable because they were “functional.” 

The court of appeals affirmed the infringement judgment on the Eames chairs, reversed on dilution for want of federal fame, and reversed and remanded on the Aeron trade dress because the jury wasn’t properly instructed on functionality. The basic problem is an insight that the Ninth Circuit has had before, but has often put badly: if consumers want the product because of the way it looks, but they want the way it looks because of the reputation of the source, then the appearance is not aesthetically functional simply because consumers want it for the way it looks. The problem is causation, because if consumers want the product for the way it looks, that is often aesthetic functionality. In addition, being known for introducing a design to the market shouldn’t itself count as the relevant reputation—that’s the rule of Dastar—as opposed to being known for making quality physical instantiations of the design. But making that distinction can prove difficult.

 

Eames chairs

“HM introduced the first Thin Pad Eames chair in 1958, and has sold hundreds of thousands of them in the United States, along with a related line of Soft Pad Eames chairs. The Aeron chairs were introduced in 1994 and were even more successful; by the time of trial, HM had sold 6.5 million of them in the United States.” Versions have been “exhibited in American art museums and made repeated appearances in American pop culture.” The claimed unregistered Eames trade dress was “the overall appearances of its Thin Pad and Soft Pad Eames chairs, excluding the chairs’ colors and all components beneath the chairs’ seats.” The registered trade dress was the same, except that it excludes the chairs’ upholstery.

 

Aeron chair

The claimed unregistered Aeron trade dress was the overall appearance of the Aeron chair with an oval-shaped lumbar support, excluding the portion of the chair beneath the seat and the chair’s color. The registered Aeron trade dress was the same, except that it also included the control box under the seat. The difference between the registered and unregistered trade dresses were not material to the appeal.

The court framed the aesthetic functionality test as “whether, if one seller were given exclusive rights to use the claimed trade dress, other sellers would be forced to use alternative designs that make their products more costly to sell, or for which consumers’ willingness to pay would be lower for reasons having nothing to do with the reputation of any source (e.g., the alternative designs would not have as much intrinsic aesthetic appeal). If such competitive disadvantages would be significant, then this second requirement for aesthetic functionality is satisfied.” 

The court of appeals rejected utilitarian functionality arguments for the Eames chairs, because the fact that they had some utilitarian functionality didn’t make the overall appearance functional, nor did the functionality of various features make the overall appearance functional. “For example, the jury could have reasonably concluded that the metal trapezoidal design of the Eames chairs’ armrests was motivated by design considerations, at the expense of the comfort that a softer surface could have provided.” OSP argued that this shape enabled the “armrests to be attached to the [rest of the chair] at three points” rather than two points, but cited no evidence that the extra point of attachment has any utilitarian benefit. 

The Aeron verdict was reversed because this instruction (taken from the Ninth Circuit Model Civil Jury Instructions) was bad: 

A product feature . . . is non-functional if its shape or form makes no contribution to the product’s function or operation. If the feature is part of the actual benefit that consumers wish to purchase when they buy the product, the feature is functional. However, if the feature serves no purpose other than as an assurance that a particular entity made, sponsored or endorsed the product, it is non-functional. 

The middle sentence misstated the law, because it’s not true that being “part of the actual benefit that consumers wish to purchase when they buy the product” is sufficient proof that a feature is functional under Au-Tomotive Gold. The instruction didn’t capture the utilitarian functionality factors, such that “a feature that provides a utilitarian benefit is not functional unless the Disc Golf factors weigh in favor of finding it so,” and it didn’t capture the rule that “a feature that provides an aesthetic benefit is not functional unless that benefit is wholly independent of any source-identifying function and the feature’s protection would put competitors at a significant non-reputation-related disadvantage.” This error was presumptively harmful, requiring reversal and remand. 

Comment: As far as I can recall, no court has ever opined on how to determine “significance” for these purposes. Is it qualitative? Quantitative? What increment of non-reputation-related market share is a trademark claimant entitled to by virtue of having some reputation-related market share, especially since functionality ordinarily trumps secondary meaning? With respect to phrases conveying a particularized message such as "Lettuce Turnip the Beet," our recent amicus brief indicates that the First Amendment interests in conveying such a message are inherently significant. Again, I think there's a real insight here but the word "significant" isn't super helpful in implementing it: the mere fact that a design is, aesthetically speaking, decent doesn't mean that it is aesthetically functional. I have occasionally suggested framing the question as whether, in a counterfactual world in which the plaintiff has no source-related reputation, the design would still have particular attractions beyond "eh, it's a fine design"; Justin Hughes has suggested looking for a larger tradition or principles, which would often accomplish much the same thing as my test, to protect things that have preexisting associations or meanings, e.g., a champagne container in the shape of a slipper or a heart-shaped chocolate box or a black (slimming) engine for a boat. The Betty Boop case is a good one here because it clarifies the normative nature of the inquiry: given that the copyright in Betty Boop expired, is the ability to compete in the Betty Boop market significant? I think Dastar means that it is, even if other mugs and T-shirts exist at the same price point.

Dilution: requires federal fame, which means “household name” status. This HM did not prove. The Ninth Circuit’s earlier Trek case held that Trek failed to show such fame even though the company spent “between $3 million and $5 million per year” on advertising, including in mainstream publications; had around 4.5 million visitors to its website per year; made products sold by over 1,600 independent dealers in 2,000 locations across the nation; and sponsored superstar Lance Armstrong, who prominently used Trek bicycles. The court reasoned that “incidental media coverage,” such as that connected to Lance Armstrong, did not “by itself constitute evidence” that the mark was famous, because “[m]any products receive broad incidental media coverage.” Also, “[a]dvertising to a mass audience is not the same as achieving fame with a mass audience and, by themselves, such advertisements prove only that Trek desires widespread fame, not that it has achieved it.” By contrast, “surveys showing that a large percentage of the general public recognizes the brand, press accounts about the popularity of the brand, or pop-culture references involving the brand would provide evidence of fame.” Though Trek was a FTDA case, the TDRA didn’t lower the standard for fame among the general consuming public. 

HM’s evidence, viewed in the most favorable light, showed: 

HM spent, on average, $550,000 per year on advertising the Eames chairs from 2004 through 2015 (and under $400,000 per year from 2004 through 2009); the Eames chairs appeared in obscure publications such as Contract, Metropolis, and an “industry publication” called Monday Morning Quarterback; at the time of trial in 2016, HM had, at the very most, around 875,000 unique followers on Facebook, Twitter, and Instagram combined; most of the Eames chairs are sold through a distribution channel consisting of only around 45 independently owned dealers with 130 locations across the country; and the Eames chairs were “very heavily” featured in the TV show Mad Men, have appeared in other TV shows and movies, and have been exhibited at several American museums, including the Museum of Modern Art and the Henry Ford Museum.

This was plainly weaker than the Trek evidence, which was legally insufficient. Even if the jury instructions accurately stated the law, HM didn’t offer evidence sufficient to allow a reasonable jury to conclude that the Eames trade dress met the standard for fame. The dissent wanted more leeway for trade dresses than trade names, but it’s still required that the mark be famous among the general consuming public as a mark, not just recognizable (e.g., the Statue of Liberty is famous, but it’s not famous as a mark). “Even assuming that the shape of the Eames chair is more recognizable than the name ‘Trek,’ there nonetheless was no basis from which the jury could reasonably infer that the general consuming public would link all Eames-shaped chairs to a single source of goods.” 

Judge Friedland, as mentioned, dissented, relying on the idea that a mark can be a mark even if the consuming public doesn’t know the identity of the (singular) producer; that HM was not well known did not mean that the Eames chair was not well known. The dissent would have held that the jury could “deem HM’s experts credible and … infer from their testimony that the general consuming public had become familiar with the Eames chairs through encounters in business environments, pop culture, and museums.” Because the design was distinctive, the jury could find that it served as “a signature of chairs made by a leading furniture manufacturer, even if they could not specifically name HM as that manufacturer.” 

I think the majority clearly has the better of this because of the difference between being well known and well known as a mark. 

The dissent rejected the side-by-side comparison of the facts here with the facts in the Trek case because it thought there was more evidence here of “actual consumer recognition” because the Eames chairs are “ubiquitous” in office environments and depicted in “countless” TV shows and movies. [One could do a very interesting class analysis of the imagined “general consuming public” here: although Mad Men was surely culturally significant for us New Yorker types, what portion of the public was actually exposed to these TV shows and office environments, or went to the museums that focused on the chair design?] Also, the design here is “iconic” and museum-worthy while Trek is a “non-distinctive four-letter term with multiple meanings.” The media coverage wasn’t “incidental” as it was in the Trek case because the chair was a distinctive product design “that the jury could have inferred was memorable to many consumers who saw the chairs.” 

Excellent example of casual empiricism here: “While members of the public can consume products or encounter advertisements for products without focusing on the marks they feature, it would be difficult for consumers to interact with a product without forming an impression of its overall appearance (its dress)—particularly when that appearance is distinctive.” [I think this is just made up: at least, I don’t recall any literature suggesting that overall appearance does any better than any other kind of feature at sticking in a consumer’s mind; like most things we encounter, most design probably doesn’t go into long-term memory at all, certainly not as source-identifying. And as a legal matter it’s inconsistent with the Wal-Mart rule that all product design requires secondary meaning for protection.]