Houston IPIL conference: Mark McKenna

2011 IPIL/Houston Symposium
Trademark: Today and Tomorrow

Mark McKenna, Dysfunctionality

Very difficult to teach functionality: why? In mechanical functionality, courts have at least two very different ideas about what functionality is supposed to do, making it difficult for them to agree in any particular case. Worse in aesthetic functionality: some courts reject the concept outright, others are uncomfortable with the consequences.

Window into bigger problem with TM: despite all TM’s talk about how its goal is to promote competition, it doesn’t have a theory of fair competition. Thus when you’re trying to judge unfairness, there’s no backstop. What is competition supposed to look like? This is why courts are attracted to free riding and market preemption ideas, because they have no idea where the concepts are supposed to stop.

Competitive baseline: mechanical/utilitarian functionality. Traffix is an attempt to resolve a tension between 2 different views: (1) Right to copy: Sears/Compco; basic idea is that the baseline is that everything is available for free copying, and deviations from that are to be secured from copyright and patent. Labeling is the remedy. (2) Need to copy: puts burden on defendant to establish need for a particular feature, and if not TM operates unproblematically. Traffix is (1) because many patented features don’t create market power and if the patent is expired there is often equally good technology given the advance of tech over time. Expired patent is not strong evidence of (2), but it’s good evidence of (1).

But a lot of courts refuse to accept what the Court said in Traffix. Federal Circuit says that Traffix confirms what it was doing all along, which isn’t so. Courts have oscillated between views (1) and (2) for a hundred years. Sears/Compco is (1), Bonito Boats is (2)—recharacterizes Sears/Compco to say that as long as there’s a functionality defense and a secondary meaning requirement, preventing access to the feature is fine.

These are both competitive baselines; they just conflict. (2) says that if TM would regard this as a source identifier there’s no problem as treating it as such without a special need. (1): the defendant always wins. (2): the plaintiff always wins.

What about aesthetic functionality? No similar concerns about design patents. Very clear trend: design patents are not to be worried about. Courts don’t think aesthetic/design features are competitively important. Access to mechanical features is important, but design features are not and can be designed around. Implicit assumption about what kind of competition we should have: competition to develop aesthetically pleasing designs, but not non-patent competition about mechanical features.

One reason courts really struggle with aesthetic functionality is that they’re uncomfortable with its consequences. They think of it as a doctrine that knocks out protection for a feature full-stop—not protectable against anyone, like genericness. But that’s often not really what the defendant is saying—the deft is saying that “the way I’m using this is not as an indicator of source.” Au-tomotive Gold is like this—they’re not saying that VW isn’t a mark for cars, just saying that when we use it on keychains we’re using it in a way functional with respect to keychains—people don’t care whether VW made it, they just want it to match their cars. Courts are uncomfortable with thinking of aesthetic functionality as a “functional use” defense—it’s functional for me.

Market definition: Dippin’ Dots pellet ice cream case. P says they don’t need to make pellets to compete in the ice cream market, look at Ben & Jerry’s. Court says: they don’t want to compete in the ice cream market: they want to compete in the flash frozen ice cream market. They aren’t substitutes for McKenna’s kids. How you define the relevant market in which there’s supposed to be competition is dispositive, but courts do it without a methodology or without thinking of substitution. Not advocating antitrust test, but courts aren’t even aware of the fact that figuring out whether there’s a competitive disadvantage you need to know what the market is. Example: orange flavored soda doesn’t need to be orange; brown/purple would still taste like orange soda. [This is not quite true.] The problem is that if it wasn’t orange people wouldn’t buy it. So functionality requires that we define the market is “orange soda,” not soda, and TM doesn’t have a way to talk about it.

Extreme doctrines that people have been critical of like post-sale confusion: is competition just price and quality competition, or does it include competition on the dimension of brand image? Many cases think: even if access to a feature is a relevant quality consideration, but we should encourage people to compete on brand image, we shouldn’t allow them access to those features—make them create new images. Coca-Cola isn’t a market, it’s a brand image. That helps explain what’s going on in the more extreme cases.

My thoughts: [research on how color influences flavor judgments, re the orange soda example] Functional use: like descriptive fair use—it’s descriptive for me. Interesting test cases: iPodmybaby where, assuming the click wheel is functional, it isn’t functional for this use; TerraCycling bottling/fake safes made out of actual soda cans or Cheetos cans. Functionality of the Hooters trade dress: what kind of competition is ok? Safecut/baby bottles case: how do configuration claims come into this, when the claim is that this particular configuration offers special advantages and disadvantages compared to others?

Ann Bartow: disclosure doesn’t work, so labeling might not be enough in Au-tomotive Gold.

McKenna: but that’s what functionality is: some confusion is tolerated because the feature is functionality.

Bartow: but that’s an upward slog.

McKenna: it’s almost always a source indicator and functional. His view is that you can decrease confusion through things short of an injunction, but you can’t do anything to preserve competition once you’ve enjoined. (He also thinks that confusion over authorization shouldn’t generally matter; see his & Lemley and my work on materiality.)

Greg Lastowka: what kind of competition do we think is socially useful? Counterpart to aesthetic functionality is copyright, not patent.

McKenna: in a lot of cases it’s design patent—design of bags or clothes.

Lastowka: Betty Boop case—an appeal to things that are aesthetically pleasing. We’re not depriving the public of something useful.

McKenna: but that’s the judgment that he questions. Why do courts make that normative choice without fully explaining it.

Lastowka: echoed in patent and copyright—you can’t have Mickey Mouse absent fair use.

McKenna: but we’re comfortable with that in patent too. The analogy works fine (except that copyright no longer has a functionally limited term).

Lastowka: copyright won’t protect some of the broader trade dress claims.

McKenna: courts are descriptively comfortable with using TM, but from a right to copy perspective the fact that a design is outside of copyright is a reason not to protect it.

Lastowka: roots of doctrine are in utility not aesthetics; always shadowy/questionable.

McKenna: some courts have accepted competitive need, but parties have shown up saying “I need this aesthetic feature too.”

Lastowka: but the competitive need is [often] derived from the brand not from the product. So that could cut deeply into brand value.

McKenna: this kind of brand value exists only because of the law. No one would’ve thought 100 years ago that a Betty Boop T-shirt implicated brand value. TM has been a one-way ratchet.

A couple of cases use “contextual functionality”—Rosetta Stone: Google is using the mark for the functional purposes of locating things.

Aesthetic functionality gets a lot of pressure because our other tools are bad—we accept almost anything as secondary meaning. Aesthetic functionality as functionality of d’s use may make more sense with noncompetitors as in Au-tomotive Gold, because when competitors are using the feature their claim is likely to be “everyone in this market should be free to use this feature.” (Though he notes that VW is likely to claim that it’s a competitor in the market for keychains.)

Laura Heymann: the patent might be able to tell us what the market is in some sense. On labeling: the court in Traffix says you don’t have to put a box about the springs—you can compete without engaging in extra effort to hide the feature/decrease confusion.

McKenna: though we now see labeling requirements in genericness cases. Putting a box around the springs requires making changes to the design; there’s nothing problematic about requiring labeling and packaging in the sale, but you can’t require a box that would stay on during use.

Heymann: aesthetically, courts are concerned with the defendant creating the market through its own activities. The question isn’t so much “is there a market for black outdoor motors” but “are consumers likely to see source indication.” But that gets back to circularity/how law shapes preferences.

McKenna: it’s often really hard to tell whether the courts are saying there’s aesthetic functionality or no secondary meaning in the classic cases like Wallace. But it’s striking that courts in those cases aren’t saying any of the things courts ordinarily say about secondary meaning: empirical evidence. Courts are saying “there’s no secondary meaning here and don’t try to tell us there is,” which is the same as saying “don’t bother to prove secondary meaning; we don’t care if there is.”

Heymann: does calling that functionality get you something, as opposed to saying that we’re going to call it nonconfusing by fiat? [This is the nominative fair use debate.]

McKenna: he’s taking the terms as they find them. McCarthy says aesthetic functionality is bunk, but that there are some things functional for marketing purposes—he just doesn’t like the word. Not trying to change the terminology. Focusing on secondary meaning makes him nervous because courts will inevitably find it in bad cases.

Michael Grynberg: secondary meaning intrudes into right to copy v. competitive need. Right to copy = anything not copyrighed or patented. But definition of right to copy almost excludes product design.

McKenna: trade dress can often be patented.

Grynberg: but trade dress is not a null set.

McKenna: part of this is terminological. Packaging/design. To the extent there is a category of protectable configuration, it would be something outside the legitimate subject matter of patent or copyright. Lots of things are allowed—packaging, maybe restaurant design. [This really does limit TM in most cases to words and symbols. I’m surprisingly okay with that.]

Sapna Kumar: are courts capable of coming to consensus on what it takes to promote free competition? We don’t usually think of them as all that competent on the matter.

McKenna: courts have done all of TM—created all the relevant doctrines, so there’s greater comfort. Not sanguine about turning it over to legislature; usually a disaster in TM. This is the kind of policy judgment Congress could make, and one could say that Congress did make it by creating patent and copyright law.

Me: (1) Discussion with Heymann shows structural similarities between emergence of nominative fair use from descriptive fair use and emergence of aesthetic functionality from utilitarian functionality: as it turns out, and corresponding to the expansion of the subject matter of TM and the scope of what we call actionable confusion, the thing we want to do with fair use/functionality needs to be done in 2 different ways depending on how the defendant is using the mark at issue. (2) ok, talk about symbols. Many will be within subject matter of copyright, so there are strong implications for them too.