IPSC: TM I/(c) I

IPSC, DePaul

Trademark I

Stacey Dogan, Boston University School of Law, Hollywood's Trademark Law

Courts seem to be treating classically expressive uses, including titles of movies, as source-indicating; Kagan didn’t seem to intend this result and courts should be more cautious. But also thinking about Empire. Courts were troubled by claims of expressive use by parties whose objectives seemed brand-related and commercial. When creative industries are treating characters and franchises as brands, why should TM disagree?

First part: titles. For a long time, courts have treated title v title conflicts as potentially actionable. Deserves separate treatment.

Second: Hollywood and branding. Claim: Hollywood’s brand fetishizing has blurred line b/t expression and commodity, with adverse consequences for creators of all sorts. We should think about whether it’s possible to limit those effects and provide analytical clarity.

Film as brand isn’t new: MGM v. Pink Panther Patrol where Leval grants injunction. Movie studios have aggressively licensed characters, titles and other features on toys, clothing, and other merchandise. More recent decades: intensified to the point that brands become films (Barbie) and thus improve the brand. Self-parodying. Line b/t expression and TM use/branding has been eviscerated. Variations: title; series; character; other features (Simpsons Duff Beer, Krusty Krab).

Effects on ability of others to use in expressive works is one issue, but another is the implication of this trend for creative industries’ ability to defend their own uses of things that might be subject to others’ TMs. JDI: Rogers doesn’t apply to use as a designation of source for the infringer’s own goods. In the wake of JDI, several courts have found movie and TV titles at least potentially source-indicating. Some of these courts have emphasized Ds’ commercial and branding efforts w/r/t the mark in the course of finding potential source-indicating use. Belin v. Starz, CD Cal 2024: Ds filed an application to register “BMF” for various classes. Not entirely irrational: if the filmmaker is claiming its own level of exclusivity in a term, reasonable to ask whether it is creating confusion w/someone else’s mark. Not crazy to ask if the intention was to create a franchise instead of a more traditional title. But that casts doubt on claims of Rogers eligibility. To the extent that courts start treating titles as presumptively source-IDing, that affects other creators using titles as terms of their own work.

Options: election: parties that rely on Rogers should be estopped from asserting TM claims in the title/character/feature at issue. (Aqua specifically avoided selling Barbie related merchandise.) Cf. Empire, where studio sought (and ultimately abandoned) registration. This does not mean that use is infringing—LOC still required.

Betsy Rosenblatt: Hollywood has no interest in claiming expressiveness and 100% interest in claiming ownership in this context. Given a choice, they’d rather assert TMs. Expressiveness isn’t up to the user. Wary of election in this regard. Rather: say that when things get secondary meaning, they don’t get to use Rogers, but unless and until they do serve as source-IDs they aren’t marks. Characters aren’t inherently marks. Titles aren’t inherently marks.

A: not suggesting that election = rights, but rather would allow intention to build secondary meaning. [Discussion of fact that Hollywood overreaches by registering lots of stuff like single titles.]

RT: Consequence of failure to have coherent commerciality analysis versus commercial speech analysis. “Expressive” is a complete failure of a concept—it’s true that the Apple logo is expressive for Apple too.

Highlight more the conceptually less justified title as special: courts don’t seem willing to make the jump to Krusty Krab and Duff Beer … just yet. The Rogers victories are in cases of content (w/exception of Toy Story feelings bear litigation, where there was licensing). Non title uses as Exception v. rule? But what is a “title”? Website (Punchbowl), song title (Chainsmokers Sick Boy case is super interesting) where there is no tradition of sequels or franchises.

Note challenge of reciprocity: usual error is to treat P’s source-IDing use as D’s therefore source-IDing use.

Mark Lemley: Evaluating D’s attempt to make TM use might be workable—Krusty Krab stuffed toys might be a TM use in the category of stuffed toys w/o having a TM in Krusty Krab.

Victoria Schwartz, Pepperdine Caruso School of Law, Rethinking Use of Trademarks in Art

Note that Pepperdine is litigating against Netflix and her views are her own.

Concurrence hints in JDI: maybe Rogers isn’t good law at all. What do we do if Rogers goes away? There are alternatives. A lot of hope in nominative fair use as doing lots of the work. Need to shore it up, but most of the cases we’re worried about could easily have been decided as NFU cases. Use of a brand in all sorts of artistic work to reference that brand is at the core of NFU.

To think through: what NFU has that Rogers doesn’t is “no more than necessary.” That’s where I take some hope from © fair use analysis, where courts are pretty willing to say that you have to take enough to make the reference work. Reading that similarly in TM is you can say “Barbie” a lot if the song is about Barbie. You just can’t say “Mattel’s Barbie Girl.”

Alex Roberts: Empire isn’t about the real Empire; Honey Badger isn’t really a reference to the Honey Badger guy, but maybe descriptive fair use covers that mostly. [I was thinking differently: that there is a kind of descriptive fair use that is a descriptive use to identify the topic of the conversation, which we call nominative fair use.]

Felix Wu: Is it necessary to use the mark at all? Once you’ve decided to have a survey about the New Kids, you need to use their name. But do you need to use LV in particular to talk about luxury? That’s what the concept of artistic freedom gets you: you get to pick whether you want to talk about LV when you talk about luxury. And courts don’t agree on whether you need to justify your choice.

A: sort of like the satire/parody distinction—and that’s bad.

Mark Lemley: What are the consequences of fitting these cases into other boxes? One consequence is that, in the 2d Circuit, you have to do a full LOC analysis and you add a Third Circuit defense on top of that. NFU law is screwed up! A broader challenge: the thing Rogers did that was helpful was that it said “we don’t care about actual confusion/your survey.”

Felicia Caponigri: what do you do when Mattel says that it’s an artist too and therefore there’s a LOC as to sponsorship/affiliation?

[My opening speculation is about trying to restore false advertising concepts to TM, not just materiality but the relevance of explicit v. implicit representations—this would fit nicely with focusing on source confusion/speaker confusion versus other kinds of confusion as less likely to cause harm.]

Mike Grynberg: if people can undermine the test with surveys, we can look to JDI’s skepticism of surveys. How much better/worse is LOC once you start disbelieving surveys in this context?

McKenna: worth remembering that in many circuits there’s no such thing as NFU; every single circuit to consider Rogers has, w/exception of 7^th, adopted it. So you’d be starting brand new and have to figure out which version from the other circuits you should adopt.

The range of Rogers cases include obvious references to nonentertainment brands; title v. title disputes; and increasingly a third category in which my nonentertainment brand name is similar to something you’re using but it’s not nominative (Punchbowl). [Also, “entertainment” is now a huge category—is a restaurant entertainment? Is it entertainment if they have live music? Florabama case.] Rogers made those differences unimportant, but they’re now potentially three different categories. Which can be dismissed using a motion to dismiss? Picking a lane has structural/doctrinal consequences, so we shouldn’t erase a rule that keeps us from having to do a lot of that.

A: to be clear, not endorsing getting rid of Rogers!

Jake Linford: reminds him of PTO’s adoption of failure to function to deal with losing scandalousness and disparagement. Do you try to hide that move or are you explicit about it? [SCt said you couldn’t do scandalousness as a category, but one defense of the category even at the time was that such marks were especially unlikely to function; moving that assessment from blanket rule to case-by-case is pretty clearly the mission of current SCOTUS TM jurisprudence (Booking, etc.). Maybe the project of the paper could be to show how there could be case by case analyses in different ways?]

Rosenblatt: Rogers was a reaction to the Debbie Does Dallas case; maybe that’s a good comparator. There’s a value to having a 1A doctrine that does work separate from other doctrines; one value is to say that sometimes TM and the 1A conflict, and we should privilege the 1A under appropriate circumstances. What are those circumstances? Under those circumstances, we should be able to use 12(b)(6) across all circuits. NFU doesn’t get us there in every circuit.  

Copyright I

Edward Lee, Santa Clara University School of Law, Is the Lay Audience Real or Fake?

About 75 years from Arnstein v. Porter, written by Judge Frank, a prominent legal realist. Adopts a lay audience test. Likewise, de minimis test looks at the “average audience” and its recognition or lack thereof of the appropriation.

Experiment: looked at a rug case and VMB Salsoul v. Ciccone. Varied prompts—lines to highlight rug similarities; textual description of similarities; lines + text description. For music: short clips to highlight similarities focusing on horn hits, plus two minutes of both songs; text description of similarities plus 2 minute clips; and short clips plus textual description.

Results: when presented as a hypothetical legal dispute and asking subjects to imagine they were jurors deciding substantial similarity, with no prompt, 63.4% found substantial similarity; with lines, 70.3%; with text, 74%; with both lines and text, 82.2%. Significant results. Consistent with hypothesis.

Song: 23.7% said SS with no prompt; short clips, 40.3%; text 29.9% and combination 29.2%, which is a little weird.

Also did a version where they didn’t say it was a legal dispute. Rugs got weird: most said they were SS with no prompt (92% whereas lines + text got 80.8%); songs with no prompt were 44% and short clips 61.4%, text + description was 48.6%).

We think courts should modify SS test. Modest intervention: would the ordinary lay audience notice the similarities b/t the works. Cf. doctrine of foreign equivalents where the court asks whether consumers would stop and translate the foreign term into its English meaning. If the lay audience wouldn’t notice the similarities, that would be the end of the road.

Option 2: bifurcated trial: would ordinary lay audience find these 2 works similar? If no, stop. If yes, trial on copying and substantial similarity. [How would that work?]

RT: how would that work? Maybe two things are both paintings. What if they’re separate types of media but both about superheroes?

Bruce Boyden: did you define substantial similarity?

A: no for the no context groups; for the “legal case” groups, we provided an infringement standard but didn’t define “substantial” except to say that it didn’t have to be identical in every respect.

Boyden: Arnstein seems to incorporate wrongfulness of the taking; other standards say “recognition” of the taking.

Tim McFarlin: difference v. similarity—if you’re prompted for similarity, then it’s a test of whether you can spot what’s the same. What about prompting for difference too?