Sunday, September 30, 2007

WIPIP, Panel 4

Deborah Gerhart, University of North Carolina School of Law
Consumer Investment in Trademarks: Why it Deserves More in Fair Use Returns

Abstract | Paper

She’s beginning a project on the following idea: The corporate owner of the mark is not the sole master of its meaning; it’s important that consumers get some control reflecting their investment in TM meaning. The discourse surrounding consumers in TM cases is patronizing; based on a model of consumers as superabsorbent couch potatoes. One issue: use of terms in dictionaries; dictionaries respond to TM lawyers’ threats even though there’s no cause of action – consumer interests might affect this hypervigilance by TM owners. There is a duty to police, but we should carve out exceptions honoring this consumer investment. (There’s a similar proposal by Deven Desai & Sandra Rierson, relating specifically to genericity.)

We could, for example, recognize a type of cultural dilution as nonactionable, for example the appropriation of “spam” for email. It’s not blurring because it doesn’t interfere with source identification, just creates a second parallel narrative. Another possibility: Consumers deserve to be able to use marks to search for information about those marks – informational fair use.

Eric Goldman: Likes the idea of investment, and suggests exploring more how consumers use marks to manage their lives – reducing search costs in ways that we don’t usually think about. Also asks about what happens when TM owners change their products – are they harming consumers who have invested in the brand, e.g., New Coke? And wonders about the problem of heterogeneity of meaning. Brands mean different things to different people; whose meanings should we honor?

A: This last is why the problem of consumer investment is so important. We can’t just look at the TM owner.

Goldman: So how would that play out in an infringement case? How do we account for the brand stories of people who aren’t present in the courtroom?

A: We should look at evidence of use, e.g. on the internet. We look for viral meaning, not meaning created by competitors.

Lisa Ramsey: “Spam” is a sort of dilution by blurring. Also, consider descriptive terms – consumers have a right to know about the descriptive qualities of products or services regardless of secondary meaning. The descriptive fair use defense may be too limited for consumer needs.

A: Spam may have been blurred, but it’s not something that should be actionable nor does she think it was intended to. 9/11 means something different to us now, but the Porsche 911 still has a meaningful narrative. It doesn’t harm in the same way as use by a competitor.

My comment: Gerhart made several references to the special harm done by “competitors.” This seems to be calling back to an older idea of TM infringement. Once it was abandoned and noncompetitor use became actionable, TM just kept expanding, causing us to look around for limiting principles. Those principles having failed, we return to competition; likewise, people like Mark McKenna are starting to suggest a materiality requirement for infringement. The interesting thing to me is that competition and materiality are parts of the false advertising test under the Lanham Act; and furthermore, they’ve been implied by courts. So there is a model for doing something like this already in the caselaw.

Q: Consumers can contribute to the value of a product – network effects with a computer program, for example – but may not be contributing to the value of the mark – Lotus 123. There are all sorts of investments.

A: Agrees. But surprisingly, many uses of marks are expressive – people mostly buy hybrid cars to express something about themselves, despite the price of gas.

Tom Bell: Your examples involve consumers having settled on a meaning. But when people are contending to develop meaning, we could get chaos.

Peter Jaszi: (Responding to Gerhart’s earlier invocation of “Napsterphobia,” where TM owners irrationally fear loss of control/rights.) What evidence can we use to define and address the problem?

A: She’s seen it in her own licensing work – people want to lock up their content so tight it’s useless.

William McGeveran, University of Minnesota Law School
Rethinking Trademark Fair Use
Abstract | Paper

Contours of the project: Anytime someone uses a mark to facilitate saying something, we need to at least inquire about whether that’s legitimate and not within the scope of infringement/dilution, though some will perhaps ultimately be correctly suppressed. That puts descriptive and nominative fair use together.

There are few really bad decisions when they get all the way to final judgment, but there’s a persistent chill on these types of uses despite the paucity of bad law. McGeveran argues that part of the problem is a failure to conceptualize the relevant defenses – the circuits disagree, the treatises disagree, and even copyright fair use looks better. Moreover, almost all the versions end up collapsing into likelihood of confusion inquiries – for example, this is true with “trademark use” requirements, as well as with nominative fair use. It ends up being more difficult to show nominative fair use than to show lack of confusion. And, as the previous sentence suggests, the burdens of proof get misplaced – if nominative fair use is a substitute for the confusion test, the defendant has to show it’s done nothing else to cause confusion. Relatedly, because these tests are fact-intensive and conflated with confusion, they’re not prone to resolution early on in a case, even summary judgment.

So we need to restructure the law to allow early deployment and resolution, detached from fact-intensive considerations. Hard cases will still need further factual development, but that could decrease the chill.

Sharon Sandeen: What about a Markman hearing type procedure? In her experience, it did help settle cases.

A: That’s the kind of thing he’s thinking. There’s no need to change the FRCP, just doctrine and possibly judicial management practices. The substantive problem with our current defenses is that they depend so much on the facts.

My Q: How do you cash this out?

A: Use news reporting, use in political campaigns, etc. as explicit carveouts, as with dilution law. It may leave us with keywords as an issue that has to be thought out, but at least we can get rid of suits against artists and Ralph Nader.

Farley: You have to answer a question that puzzled the Court in KP Permanent: What does “fair” mean? How can it be fair if there’s confusion?

(I think the answer can be Scalia’s in Wal-mart: given the costs of sorting out the few confusing uses from the many nonconfusing ones in a defined subset of cases, the game isn’t worth the candle. This of course makes you rely on generalizations like “most uses in headlines aren’t confusing,” but if it’s good enough for Scalia, why isn’t it good enough for us?)

A: This is a problem that shows up in the 9th Circuit’s rule on remand in KP Permanent, where the court denies summary judgment – the remand makes the problem even worse by complicating the test for descriptive fair use. We must consider how our beautiful tests work structurally and procedurally, or they’ll suppress good uses.

Ramsey: consider anti-SLAPP procedures as a model. And consider saying “we just don’t care about certain types of sponsorship confusion,” maybe on constitutional grounds.

A: That could be good!

Mike Carroll: rather than needing a coherent theory of fair use, we need a more robust theory of TM’s limiting principles outside of confusion. Then we can figure out how to clean up our standards and implement our normative limiting values. Lists of specific safe harbors have pernicious unintended consequences, as we’ve learned in copyright – floors become ceilings; courts draw negative inferences. At a minimum, safe harbors should be defined functionally. And we should also loosen the standards for declaratory judgment so people can get answers quickly.

Mark McKenna, Saint Louis University School of Law
Trademark Use and the Problem of Source in Trademark Law
Abstract | Paper

The search cost theory of TM law is not sufficiently rich to give us those limiting principles Carroll talked about. Our failure to find limiting principles once the competition requirement was abandoned has led to the development of the new contender, trademark use – defined roughly to mean a use that indicates source. (You could require affixation as in the statute, but that wreaks havoc on a lot of doctrine.)

He’s emotionally on the side of Lemley & Dogan, but ultimately trademark use reduces to confusion – do consumers understand something as an indication of source? – so it doesn’t get us out of the problem.

So far the only things that make irrelevant the basic question of source identification are from outside TM: Dastar and Traffix considerations of interference with copyright and patent regimes. This is a much bigger problem than TM use: everything in modern TM law is tied to a concept of consumer understanding of “source,” and the judicial interpretations and practical meanings of source have kept expanding. Even the start time of a baseball game – 7:11 – can indicate sponsorship, so consumers can rationally conclude that anything can. The only logical endpoint is complete control over a mark. But that’s so obviously bad that we need principles to deal with it.

And this is search cost theory’s biggest weakness: once we decide consumer confusion is the harm we care about, as opposed to an indicator of some other harm, it’s hard to identify what the stopping point is and it’s very easy to characterize plaintiff’s claims in consumer protection terms. So let’s detach some defenses from confusion, increase the quantum of confusion necessary to support a claim, add a materiality requirement, and maybe other things.

Farley: We really need to bring “source” back within some reasonable bounds. McKenna says TMs are functioning in the world more broadly than they used to, so of course the law needs to follow, but she’s not convinced.

A: He thinks we can’t go all the way back to trade diversion, but he wouldn’t say we have to follow “source” as far as the market does in practice. We could define source differently for infringement purposes. It might actually matter to some consumer decisions if there is an affiliation between the plaintiff and defendant, and we should be willing to recognize it but perhaps presume irrelevance in certain circumstances. If people start to believe start times signify source, though, then maybe a team shouldn’t be allowed to pick a 7:11 start time to advertise itself.

My Q: Nominative fair use, trademark use, and inherent distinctiveness can be seen as a bunch of presumptions, some rebuttable and some not. If we raise the evidentiary standards and don’t allow owner-favorable presumptions (such as occurred in Boston Hockey), and we regulate surveys carefully, we might find there’s very little evidence of interference with trademark rights in many of these cases. Consumer perceptions, in other words, might not be as trademark-driven as we think.

A: He agrees, but it doesn’t solve the problem of kicking cases out early.

Calboli: Maybe you’re going back to technical TMs v. unfair competition.

A: If he had confidence in our ability to limit the concept of unfair competition (requiring, say, competition) he might endorse that. But he’s not sure that TM can be distinguished from unfair competition in that way. We need to define the harm agains which we’re protecting people.

No comments: