The Ninth Circuit don’t care: successful Rogers defense reversed because plaintiff’s trademark is “artistic”

Gordon v. Drape Creative, Inc., No. 16-56715 (9th Cir. Jul. 30, 2018)

The Ninth Circuit routinely invents some new epicycle for trademark defenses; here it unfortunately mushes together Rogers and transformativeness (absent the word itself, replaced with “artistic”).  There are a number of good reasons why a trademark right is not the same thing as copyright in an artistic work.  [See also: there’s no copyright in short phrases, and an artistic work can’t be a trademark for itself.]  This opinion unfortunately does not appreciate those distinctions, creating a sort of tradecopymarkright, the kind of thing Dastar cautioned against.

Gordon is the creator of a popular YouTube video known for its catchphrases “Honey Badger Don’t Care” and “Honey Badger Don’t Give a S---.” He registered the former phrase as a trademark for various classes of goods, including greeting cards. Defendants made greeting cards using both phrases. The district court granted summary judgment on Gordon’s trademark claims for defendants, using Rogers v. Grimaldi, 875 F.2d 994 (2d Cir. 1989).  The court of appeals reversed.

The court of appeals found a triable issue of fact on artistic relevance: “Defendants have not used another’s mark in the creation of a song, photograph, video game, or television show, but have largely just pasted Gordon’s mark into their greeting cards. A jury could determine that defendants did not add any value protected by the First Amendment but merely appropriated the goodwill associated with Gordon’s mark.”  What has happened here is that the word “relevance” has been ignored, and “artistic” has been reinterpreted as a thing that the defendant must add to the original, rather than having a meaning-based relationship between the original and the accused work.  I don’t understand why “Barbie” wasn’t just “pasted” multiple times into Barbie Girl, etc.—the court has some definition of “pasting” that includes “forming a joke around” but not “forming a song around.” 

What’s really going on here is a denigration of the greeting cards as works—protected by the First Amendment, yes, but not real works like songs, photos, video games, etc.  This is not without irony given the ways these classes of works have been denigrated in the past, but it also speaks to the confidence courts have in making judgments about how much art is in a work, which they seem to think differs from judgments about merit.  And therein lies a deeper irony, which is that trademarks are one prominent example of how lots of meaning can be concentrated in small semantic/symbolic units.  Hamlet these cards ain’t, but they shouldn’t need to be, and it seems to me that, as with cockfighting in Bali, one could perform a pretty serious reading of current American culture using these greeting cards.  The court of appeals basically doesn’t think that defendants did anything more than tell the same joke that Gordon did.  Ok, but it’s still a joke that has nontrademark meaning, and preserving the freedom to engage in that kind of expression is the point of Rogers.

The court of appeals found that the defendant had met its initial burden of showing that the allegedly infringing work was part of an expressive work protected by the First Amendment [pause for the usual sigh: the court means noncommercial speech, since commercial speech is also usually expressive].  The burden shifted to the plaintiff to show not only likelihood of confusion but also at least one of Rogers’ two prongs: explicit misleadingness or lack of artistic relevance.

The greeting cards were, of course, expressive works: “Each of defendants’ cards relies on graphics and text to convey a humorous message through the juxtaposition of an event of some significance—a birthday, Halloween, an election—with the honey badger’s aggressive assertion of apathy.”  The level of necessary artistic relevance must merely be “above zero.”  If this weren’t the Ninth Circuit, it would be startling for the next sentence to be in a case that finds a triable issue of fact: “The honey-badger catchphrase is certainly relevant to defendants’ cards; the phrase is the punchline on which the cards’ humor turns.”

Unlike every Rogers inquiry up until now (including Parks), the court continues: “the ‘artistic relevance’ inquiry does not ask only whether a mark is relevant to the rest of the work; it also asks whether the mark is relevant to the defendant’s own artistry.” [Citation needed.] How do we define relevance?  “The use of a mark is artistically relevant if the defendant uses it for artistic reasons. Conversely, the use of a mark is not artistically relevant if the defendant uses it merely to appropriate the goodwill inhering in the mark or for no reason at all.” [A key problem here is that “to appropriate the goodwill” isn’t the opposite of “for artistic reasons”—what the court is really doing is importing transformativeness from the right of publicity defense to Rogers, which I can’t see as a benefit, especially given the mess it further makes of the video game cases, which I would’ve thought couldn’t get worse. Silly me.  Also, “to appropriate the goodwill” sounds worryingly intent-based, though the court isn’t clear on this, and Rogers itself was supposed to draw a line between “to invoke the cultural significance of” and “to infringe” which is now apparently infinitely regressed into the test.  Suppose the defendants credibly testify that the purpose of the use was to create a joke that felt contemporary and casual.  Is that “artistic”?]

The court continues: “[f]or artistic relevance to ‘be above zero,’ the mark must both relate to the defendant’s work and the defendant must add his own artistic expression beyond that represented by the mark.”  [Apparently original artwork and minimal jokery does not qualify as artistic expression as a matter of law; I would have thought it should.]  For example, “[n]o one seeing Warhol’s work would think he was merely trying to appropriate the goodwill inhering in Campbell’s mark; no one thought Warhol was selling soup, just art.” [And what do they think defendants are selling?  Trademarks?  Unsurprisingly, here we also have evidence-free judicial factfinding about consumer perceptions of Warhol—a hallmark, no pun intended, of transformativeness cases. Also, if you can plausibly explain to me why Andy Warhol is not “simply” copying the soup cans but defendants here might be, in a way that doesn’t rely on Andy Warhol’s own (partially soup-generated) fame, I will knit you a sweater from very nice wool. Warhol famously (like honey badger itself) didn’t care, and much critical reaction at the time thought he was making nonsense; the canvases sold badly.  Query what would happen if Richard Prince invested in defendants today.]

Next, the court treats the trademark owner as an “artist” even though it is beyond peradventure that the trademark laws are not about creativity: “[I]t cannot be that defendants can simply copy a trademark into their greeting cards without adding their own artistic expression or elements and claim the same First Amendment protection as the original artist. That would turn trademark law on its head.” [This would only be true if you assume that all trademarks are “artwork” and ignore “source-indicating use” as a meaningful limitation on trademark protection—the latter of which, in fairness, is the modern US trend, even if it shouldn’t be because it creates problems like this one.  The court’s counterfactual reminds me of Justice Breyer’s example in the Wal-Mart oral argument of a hairbrush shaped like a grape: although I can imagine a card that just said “Coca-Cola” on the front and nothing on the inside, I can’t really imagine who would bother to make it or who would buy it, so I don’t know why assuming that such a card would infringe helps us answer the question of whether a greeting card that does more than that could infringe.  Would that even be a greeting card?]

For purposes of summary judgment, “defendants may have merely appropriated the goodwill inhering in Gordon’s mark without adding any creativity of their own.”  [Query: if there is testimony that the greeting cards as a whole are copyrightable despite using a stock phrase, should that satisfy the inquiry?  If not, why not?  What other standard for “creativity” do we have?]  It was true that the uses weren’t non sequiturs and made sense in context [presumably contra Parks], but there was a triable issue of fact “whether defendants added their own artistic expression, as opposed to just copying Gordon’s artistic expression.”

And then the court notes that Gordon presented evidence that he sold products bearing the mark, including greeting cards (and that defendants declined a licensing deal with him and defendants’ president testified [unconvincingly] that he couldn’t recall what inspired the cards).  With the placement of this discussion, the court seems to be grasping towards some sort of creativity-substitution standard—did they fill the same market niche, satisfying the same demand for expression?—but this isn’t a copyright infringement case.  A short phrase like HBDC isn’t copyrightable.  In fact, as framed, the court has just put copyright preclusion squarely on the table: if the question is whether there was copying of expression, that sure sounds like a right equivalent to copyright protection.

Anyway, “a jury could possibly conclude that defendants used the phrases for artistic reasons on one or more cards but not on others.”  This wasn’t like other prior cases, the court said.  In Rogers, “[t]he film did not appropriate Ginger’s mark; it came in praise of her craft, celebrating Rogers and Astaire’s ‘elegance and class’ and contrasting it with the ‘gaudiness and banality of contemporary television.’” [Which is to say it appropriated more of her image than just her name, which is apparently ok.] The film’s title was “not a disguised advertisement for the sale of goods or services or a collateral commercial product.” [Also true of these cards, which are themselves the expressive product being sold.]

The junior users in the Barbie cases “also viewed Barbie as a cultural icon, even if they did not treat her with the same adulation that Fellini did Rogers.” [And honey badger isn’t a mini cultural icon?] In the Empire case, “Empire” has a bunch of common, relevant meanings. And in the video game cases, “the borrowed mark was part of a much larger context,” [which notably didn’t translate into a finding of transformativeness in the portrayal of the players themselves as applied to the right of publicity claim,] which was also true of the GTA/Pig Pen case.  Thus, the court contrasted “the junior user’s own artistic expression” with “use in the same way that Gordon was using it—to make humorous greeting cards in which the bottom line is ‘Honey Badger don’t care.’”  [Which again conflates the expressive content of a work with a trademark for a work—assuming Gordon had a valid mark, it was valid as an identifier of the source of goods or services, not as the content of a work. “Honey Badger don’t care” might be capable of being used as a trademark for some things, though it’s a pretty long one.  Gordon has no trademark rights over the “bottom line” of an expressive work—that’s what copyright is for, except copyright doesn’t cover short phrases.]

“A jury could find that defendants’ cards are only intelligible to readers familiar with Gordon’s video and deliberately trade on the goodwill associated with his brand.”  [Here we have another way of making the fundamental mistake: the “brand” if any, and the catchphrase are fully distinguishable, just as the Statue of Liberty is famous but not famous as a mark even though there are entities using it as a trademark such as Liberty Mutual.  I would be stunned if there were noticeable secondary meaning here in the sense of consumers being familiar with the honey badger phrases as an indicator of source for stuff in the way that Hallmark is an indicator of source for greeting cards.  The defendants might thus easily have been relying on the cultural meaning of the phrase without free riding on Gordon’s “goodwill” in any trademarks.  This can be true even when there is substantial secondary meaning, as when I say “this is a Mickey Mouse operation”—unintelligible to those who don’t know the icon, but also protected by the First Amendment when it’s, say, in a restaurant review.

Another way to get at this issue: I doubt anyone would contend that Barbie Girl is “intelligible” to listeners unfamiliar with Mattel’s Barbie, or Fred and Ginger to viewers unfamiliar with Astaire and Rogers; at the very least, just as with the greeting cards, those works would lose a lot of their resonance without their cultural contexts.  Rogers, until now, allowed speakers to invoke a well-known, or not well-known, entity without equating that invocation to use of goodwill in a trademark sense. Indeed, the court’s blithe equation here of familiarity with trademark meaning is exactly what Rogers rejected.]

Anyhow, a jury must decide whether defendants used the phrases “in any way that distinguishes their use from Gordon’s and thus … ‘imbued’ their product with any ‘expressive value’ apart from that contained in Gordon’s trademarked phrase.”  [Note that this formulation is not the same as what the court apparently thinks is the real test, the repeated question of whether defendants added “their own artistic expression or elements.”  Suppose the jury finds that artistry is in the beautiful portrayal of the honey badger. Does that count?  This is basically the same question as I asked above about copyrightability.  (1) If artistry in honey badger rendition is insufficient because the question is instead whether the parties’ ideas were the same (they were using both using the phrases for greeting card punchlines), then we have Dastar problems times a million.  (Imagine what happens in a few years when somebody makes a Steamboat Willie cartoon—not a parody, just another standard Steamboat Willie cartoon—how will we determine what “distinguishes” that from an authorized Disney Steamboat Willie cartoon for purposes of this new, improved Rogers test?)  (2) If instead the question is whether the parties’ expression was the same, then we only have copyright redux (and summary judgment for defendants should be warranted, unless there’s something the court isn’t saying). (3) More likely the court is just thinking “there’s not enough” without giving any useful guidance on what “enough” might be, or why more than “more than zero” ought to be required.]

But hey, the defendants might win on artistic relevance, or on likely confusion or on abandonment.