WIPIP panel one: TM tarnishment and (c)

[title fixed because I can't keep seasons straight]

Suneal Bedi, Bad Brands: Experimental Studies in Trademark Tarnishment

What is the reputation of a mark?  Working on PhD in marketing at Penn; marketing is the best discipline to answer this Q.  Brand associations and attitudes are the reputation of a mark. TM isn’t just source indicator; it invokes nodes of meaning.  Attitudes= liking; associations= other meaning, like “Mexican” for Chipotle. Thinks that tarnishing should have to affect preferences/attitudes—should move Chipotle down in the ranked preferences/affect market share. So why not look just at sales? The statute is focused on reputation, which is distinct from decreased sales though it should predict such decreases. There can be instances of reputation harm w/o harm to sales.  WIP looks at isolating decreased sales; have to use sophisticated quantitative methods. 

What predicts tarnishment? 1. Multiple exposures.  At minimal exposure, people don’t really care. It’s an S shaped curve: middle exposure is high effectiveness; after that there’s a saturation point, which is why it doesn’t matter for Pepsi to advertise at the Super Bowl.  Tarnishment is competitive advertisement harming the senior mark. But only awareness & internalization of associations can do harm.  A few exposures or one won’t affect reputation.  Empirical strategy: multiple exposures.

2. High cognitive load.  Elaboration likelihood model predicts attitude change in 2 different ways: (1) central (rational and thoughtful); (2) peripheral (positive or negative cues, simple inferences). People use peripheral processing in times of high cognitive load.  A double-take: your mind is somewhere else, but as soon as you look close you aren’t really confused. That’s an instance of high cognitive load, and that may predict tarnishment.

Studies include pre-tested brand associations and attitudes for several brands. Chick Fil A; fictional tarnishing brand, Chicks Fill A gentleman’s club.  People prefer Chick Fil A lower after exposure and find it to be less wholesome.  It makes it a little different but not a lot; not on tastiness or liking at all. What happens about multiple exposures?  Embedded banner ads in news articles; asked people to tell them about the news article—high cognitive load.  Zero ads v. 4 ads: Then there was an effect on rank, tastiness, and liking.

[This is all an argument about the nonrational, nonfactual, non-thought based nature of tarnishment which under current 1A doctrine should put it off limits for the law, just as in the tobacco cases.]

Felix Wu: what implications for tarnishment cases?

A: courts aren’t using the right empirical evidence, or requiring the right empirical evidence. Both under and overinclusive—actual harm is underinclusive, but 6^th Circuit’s presumptive harm is overinclusive.  If it exists narrowly, Ps must introduce good empirical evidence.

Lunney: The Streisand effect—most such uses you’d never hear of except for the litigation.  Debbie Does Dallas, Pillsbury Doughboy. 

A: The distribution of awareness—how many people see it; also of the people seeing it, how many times they see it.  Both matter.  People who saw the movie saw her once in the costume, according to Lunney.  He says that it’s not therefore likely to be tarnishing.

Q: what about the ridicule factor? It doesn’t necessarily matter what the “tarnishing” is but just the challenge, mockery.

A: but how do we know it’s a mockery? Only if it’s associated with something perverse.  [Wacky Packages]

Lunney: is it ok to garrote the demand for something through criticism, a la Campbell v. Acuff Rose?

A: Wal-Qaeda: Jacoby’s survey says there was tarnishment; but overriding concerns like free speech matter.   Neutral on whether this is a good idea (thinks it’s kind of dumb).  He’s ok with tarnishment not being something the law regulates—the point of this was to discuss the standard of proving tarnishment now, which is bad.

RT: Charts show 8 ad spike/recovery in ratings. 

A: he thinks it’s a demand effect b/c respondents started to say “I think you were trying to get me to like Chick Fil A more” when he asked them what they thought the survey was trying to prove when there were 8 ads. [This seems worth further investigation]

McKenna: what makes us think that consumers’ reactions have anything to do w/the fact that it’s ads and not other things that might produce exactly the same effect. [I noted that Victor’s Secret image he used was from the Simpsons, not the actual Victor’s Secret shop.]  If there’s nothing unique about what the law targets, it’s cherrypicking from the same activities.

A: what’s unique is that the point of using it is to draw attention to the junior mark specifically people because people are aware of it.  Criticism = not using Wal-Qaeda to draw attention/coopt.

McKenna: now you’re just talking about free riding.

A: yep.  That’s the actual harm of dilution/tarnishment.

Charles Duan, Implications of Oracle v. Google for Technical Standard-Setting

Is it an infringement to implement someone else’s interface? Most standardss come from private standard setting organizations that invite participants from multiple groups who propose ideas and then the SSO chooses for standardization. Ideas contributed by 3d parties.  Patentable inventions?  SSOs have very comprehensive patent policies trying to prevent patent claiming. But fewer have copyright policies.  W3C’s policy changed in 2015, but Oracle v. Google was 2014—may have been a response to that decision. Confirms his belief that programmers didn’t think APIs and standards were ©able. Should courts accommodate industry expectations?  There’s value to that, though the law is not bound to follow experts in the field.

SSOs as a practical matter may want to start adopting © licensing policies.  Standard-setting organizations like IEEE might want to be more involved in these discussions in ©, though members may have conflicting incentives.

Q: what about ASTM—software standards varying from other types of standards?

A: distinctions—gov’t mandate in ASTM case. That was copying the standards text and not implementing the standard—as if ASTM were challenging buildings made using their standards.

Carys Craig, Substantial Transformative Taking: Holistic Comparison and the Non-Infringing 'New Work'

Bakhtin’s idea of dialogue: what an utterance chain looks like from a dialogic perspective and how that should influence ©.  Utterance as work: four connected things: what has been said before, the already spoken (relational meaning); distal not yet spoken (the still to be said;) the proximal not yet spoken (anticipated response); distal already-spoken (broader cultural meaning). So how can © protect the already spoken while permitting the dialogic utterance in the chain and leaving room for the not yet said?

What is a copy?  Analysis at the moment of prima facie infringement at which you determine something is substantially similar. Canadian case: Cinar Corp. v. Robinson: pitch and pinch: someone pitches a story about a film, which is then made into another film.  Claude Robinson had spent years building materials for Adventures of Robinson Curiousité; then he saw it on TV anyway w/parallel characters, settings, scenes throughout.  On appeal, Cinar argued that court was wrong to compare the works without filtering out the unprotectable elements.  Altai abstraction filtration comparison test: Canadian SCt said it wasn’t appropriate for this type of work which has to be compared holistically.  Shouldn’t dissect work, which eviscerates it—hard to get a sense of the authorial contribution.  Comparison, abstraction, filtration, then compare again if necessary.  Criticized for undue focus on protecting author over downstream creators, influenced by very sympathetic facts.

Maybe we need to look at whether the defendant’s work is in its whole something new—new expression. Holistic comparison can be used to shift to asking about the dialogic relation b/t the texts and to see the D as a possible creator too. Could consider transformativeness of the taking into our assessment of substantial similarity, rather than waiting for fair use and fair dealing.  By the time you’re searching for an affirmative defense, you’re already on the wrong side of copyright’s moral equation—seeking a reprieve by calling yourself an author too. We could do this at the infringement stage instead.  From Canadian perspective: transformative use isn’t actually available as a defense, but we could put it in infringement. 

Example: Larrikin Music Pub’g v. EMI Songs Australia (Down Under/Kookaburra song)—line taken from common folk song and reproduced in hit; court could find no way to avoid conclusion that substantial part of the song had been taken, and there was no defense. Holistic consideration could find a new work even w/some copying of the “golden nugget.”  Substantial transformative taking should be regarded as an original work—not derivative but responsive, new utterance in the chain in a discursive struggle w/P’s already spoken work.

RT: I think Germany does this already.

A: not that surprised—old English fair use cases were about new works being created; not framed as a defense at all.

Q: what happens to the derivative work right?

A: Canada doesn’t have one and neither should you. But you should also resist the collapse b/t derivative and dialogic.  A derivative work has to be a substantial reproduction so we can use this framing to get over the derivative hurdle.

Rosenblatt: comprehensive nonliteral similarity v. fragmented literal similarity?  You maybe want to distinguish them.

A: yes.  Sampling cases—solving a problem that can’t otherwise can’t be solved.  Blurred Lines: they failed to filter out what should have been. But sampling, you can do the filtering and still look like you’re left with “golden nuggets” that are substantial, even though there’s a new work.

Q: so how do you distinguish b/t dialogue and “pinching”? [I would frame this perhaps as: was Bakhtin’s formulation even set up to contemplate “infringement” or other acts that might be considered illegitimate moves in a dialogic chain? What if we care about things other than dialogue—can incentives fit in?]

A: good Q. Wouldn’t necessarily get Cinar off the hook but would limit attempt to control downstream uses?

Rob Walker, The Great Disappearing Doctrine: Making Sense of Scenès à Faire

Doctrinal oddities: has a descriptive take but no solutions yet.  Scenes a faire is apparently a term from 19^th c French dramaturgy, but he hasn’t been able to find  much/any use there, so if someone can figure out what the judge who first used the concept meant that would be great.  [Paging Francesca Coppa.]

Hard to distinguish what courts are doing/what is or isn’t a scene a faire, though we have some definitions, e.g. for software it’s expression dictated by peractical realities; incidents, characters or settings which are as a practical matter indispensable, or at least standard, in the treatment of a given topic—in some ways this is just an application of originality. 

What do we mean by indispensable? May be clear in literary work with traditional structure, like Chekov’s gun.  But what about avant-garde works or genre-busting works? Isn’t this idea of indispensability just a judgment about what a work needs to be a “good” version of itself, contra Bleistein?  Is Star Wars a sf film, a fantasy film, or a western?  What are the conventions that might be standard to it?  Is an earnest treatment of a topic using conventions in a traditional realist way unprotected, whereas a parody of the conventions could be protected, e.g., Blazing Saddles and Young Frankenstein and Spaceballs?  What’s the time at which we set a convention—at the time the P’s work was created, or the D’s?  Using the conventions of Elizabethan drama to write a new play would be strikingly unusual today, but not in Elizabethan times.

As courts get into the analysis, scenes a faire tends to disappear into the ether and is replaced by idea/expression and merger. Example: Fulks v. Knowles-Carter, in which avant garde filmmaker sued for infringement in Beyonce’s Lemonade video.  Court’s attempt to pinpoint scenes a faire collapses into idea/expression and merger. The exception: software cases where functional elements, programming conventions etc. matter.  But no guidance in literary/artistic context. 

Biggest problem: shifts focus away from proper analysis of similarities b/t the works and their substantiality; talks about the overall genre or milieu instead.  Underprotects realist works and over protects genre defying or defining works, creating a Bleistein discrimination problem.  Sherlock Holmes: when written, new/creative (maybe; he hasn’t done the research), but at a certain point this character solidifies into a set of tropes copied in many different versions over time—eviscerated?  [Public domain?]

Fix: return to focus on originality at the time the accusing work was created. In software, focus on functional/nonfunctional.

Q: relational context from Rosenblatt & Craig—Fulks submitted his video to Beyonce’s producers.  You lose the relational element.  Taylor Swift: the court says that players gonna play is too banal.  Making old lyrics generic.  Timing: can Swift make anything generic through success? Then we lose something about the contribution of the earlier work.

Madison: Judicial biography: the judge who used the term first was committed to balancing access & rights—a pragmatic intervention along the fair use continuum; another way to get at similar scope problems.

Q: why when we think about genre works it would result in underprotection? Why wouldn’t recognizing contribution of individual work result in right-sizing?  Originality might not give us right-sized results; shifts us to thinking about what the author contributed, not what the author took.

A: Will look into relational interactions.  May change his under/overprotection analysis.

Rosenblatt: every scene a faire was invented by someone but we often don’t know by whom.  Functionally an expiration doctrine?

A: likes that.  Certain things shouldn’t be protectable. If we start thinking about originality and take Feist seriously—the bar is low, but it exists—that’s the right conversation.

RT: Doesn’t originality assessment require an assessment of prior art? The reason we try to figure out what other works have done is to figure out what shouldn’t be attributed to the original work of the accusing author.  And one function of scenes a faire is that it deals with what the patent folks might call “Obvious to try”: someone was the first to use the missed phone call instead of the missed appointment at the top of the skyscraper but that doesn’t mean they should get the rights to that plot point; see also the Jurassic Park case—once you come up with the idea of a dinosaur island, certain tropes follow for anyone with minimal exposure to narrative structures.  That’s the function of scenes a faire that allows it to parse originality.