Design Law Scholars Roundtable part 3

Session 3: Shaping an Overall Approach to Protection

Introductions: Jerry Reichman and Estelle Derclaye 

In what ways should different forms of protection be tailored to various types of subject matter (construed legally as forms of intellectual property, but also subject matter construed socially or commercially)? How, if at all, should we deal with concerns about overlap/cumulation of rights? Should the current procedures and institutions for granting and challenging the validity of rights be changed? Should we channel subject matter to certain forms of protection? If so, how should we draw the lines between different subject matter? Should we limit the scope of different rights so that those rights have different effects? Provide for different remedies?

Reichman: battle on concurrent protection has been won by France. Italy abandoned separability, which we adopted. If courts take conceptual separability seriously, it becomes the German test in disguise—but even the German court has now abandoned a test of superior creativity, so there’s only one test of originality in German law, which doesn’t require superior creativity.  From a legal perspective, concurrent protection is the de facto norm in EU law.

That said, he doubts that this is as important result as it seemed a decade ago. Because today, industrial designers are more inclined to seek unending TM protection. Even © protection remains iffy where form and function are fully integrated. In his view, role of sui generis design law today is to allow designs to acquire secondary meaning through use. When that demonstrably occurs, then the design can be protected indefinitely. The most ingenious pro design legislation anywhere is the EU regime of unfair competition law, prohibiting slavish imitation of unregistered designs for 3 years. Brilliant b/c in fast moving markets a successful design may acquire secondary meaning in 3 years; TM then takes over. Or it may have little/no further value on the market, in which case neither design nor © is of interest to investors.

That isn’t to say sui generis laws are useless, but they should be simple, easy to use, and inexpensive. To his mind, the best sui generis law was the PTO/© office proposal of the 1970s, attached to the draft act as Title II, but separated from the 1976 Act at the last minute on the grounds that it was anticompetitive.  We are stuck with design patent law, which expressly requires novelty, originality, though not nonobviousness and which is thus not fine under TRIPs. Courts in England and elsewhere have managed to complicate design laws so much that it no longer looks a ton better than our design patent scheme, but he still doesn’t know why we haven’t been sued.

Separability has become a farce in ©.  Imagined separability boggles the mind as framed in Star Athletica. The opinion is so meaningless as to require Denicola to make sense of it. Still a blemish on our law, b/c we exclude a bunch of designs from © and send them to design patent, but many can’t get design patents, which thus violates TRIPS.  Enacting a 3 year law as Title II might be the best option, even 50 years late.  That would stimulate our design industries further.

Estelle Derclaye: Overlaps are not bad per se. If there are different purposes, they are acceptable. Cumulating damages for the same infringement is an issue where the harm is exactly the same. © protection can be anticompetitive, where © extends (or is attempted to be used) where patent should be. E.g., you use the contract for software that you can offer b/c of © to bar the extraction of ideas from the program.  Or use of TM to extend a patent. Or overlap during same protection where the exceptions/scope of the regimes differ. TM/patent/design interface for shapes and other characteristics giving substantial value to the goods. Several ECJ cases have interpreted. Q: does the same demarcation apply to ©? She thinks it should. Not clear whether there is negative convergence such that only patents can protect these shapes solely dictated by function.  

Designer’s option to register design to cumulate: getting rid of community unregistered design right. That happened in UK, 2014 with its own design right—trying to minimize regime clashes. But if you do that at, you need a system which has a very well crafted registered design system. If you have a patent or © approach for registered designs you’ll have cycles of over and underprotection.  A design approach has to be neither © nor patent; EU is a good starting point.

Building on Ansgar Ohly’s proposal: need a rule for how you cumulate, including w/unfair competition. Imitation alone shouldn’t be banned w/o confusion as a cause of action under unfair competition.

Remedies: courts should consider the underlying nature of the harm and identify the remedy best designed to ameliorate that harm.

Sarah Burstein: Reichman is worried that designs can’t satisfy nonobviousness—but the Federal Circuit has removed constraints.

Reichman: the Fed Cir has improved it; but still, nonobviousness means that people don’t apply—the lawyers tell the designer there’s no chance.  [A lawyer who did that now might arguably be committing malpractice.] Pre Fed Cir, almost all design patents were invalid.  With the Fed Cir, the design patent system works. But even with a flexible nonobviousness standard, you’re not going to get much protection.

Burstein: the lawyers I talk to don’t consider 102 or 103 a serious bar and don’t counsel their clients that way. Not today. We know empirically that the number that are rejected is very low. The Fed Cir has made the test for hard to satisfy that rejection for obviousness is essentially impossible.

Sprigman: Examiners also don’t have the way to examine they need. Went through design patents for yoga wear. They didn’t even make a nonobviousness inquiry

Burstein: they don’t have the prior art. The examiners try, but it’s stacked against them.

Dinwoodie: state of law or practice?

Sprigman: both! The standard requires examiners to do something they’re not equipped to do.

McKenna: b/c nonobviousness is incoherent for design, the Fed Cir has made the standard impossible.

Sprigman: racerback bras: multiple variants have been granted based on narrowness/length of straps.

McKenna: agree not a meaningful bar, but trying to craft a workable standard.

Buccafusco: drop the word, but consider the concept: a judgment of similarity at some level of abstraction. No different from idea/expression.  So far, there has been virtually no abstraction at all; just a little bit of difference is enough.

Dinwoodie: every design system has a question like this: how much difference does there have to be? Reichman argued that utility patent obviousness didn’t work in a system that regarded incremental innovation as its object (design), but the framing: how different does this have to be to give you an IP right? Is a workable framing.

Sprigman: racerback bra straps—length and thickness would be normal variations. Two criss cross straps, you can patent three. That’s a problem.

Jason Du Mont: isn’t that just formalism problems?  The formalist test restricts examiners from using primary or secondary references in a way that would take that into account.

Dinwoodie: so would Reichman abolish design patents in favor of Title II design right?

Reichman: yes, replacing it w/© like regime. If Burstein is right [she is], then they’re no longer applying utility patent standard, which is what they wanted from design right. Is that enough to satisfy TRIPs?  He doubts it.  TRIPs reaffirmed the Paris Convention; a foreigner who doesn’t get a design patent because of obviousness should object.  

Sprigman: the standard probably fits now; there was convergence.

Dinwoodie: for that regime to work, separability must work robustly. 1980-2015 law was largely pushing stuff out of © is the other part of the puzzle.

Reichman: agrees. © office knew what it was doing with separability, but at that time separability was supposed to (a) be tough and (b) coexist with a design right.

Dinwoodie: Derclaye seems to want British law before we were forced to change. If you took a design that has © and apply it industrially, we then restricted your © to the scope of registered design.  Utterly compliant w/EU law but British IPO decided it wasn’t and abolished it.  Another provision of the law prohibited rights in the article via drawing of article; if you want that do you want unregistered design right abolished?

Derclaye: should abolish the Community unregistered design right b/c © does the trick. If you keep UK unregistered design right, the prohibition on getting rights in the thing from drawing of the thing is also good. Would change to election of remedies by choosing to make industrial use of a work. So you could use © without knowing you want design right, cumulation would be no problem, it would just limit the remedies/term you could get if you used © to protect what was actually an industrial design.

Buccafusco: Canada’s law is based on the old UK system. If you make more than 50 copies you have to switch to design right.

Ansgar Ohly: © and design law: Europe is a laboratory with all the possibilities that could exist—we’ve had separability, we’ve had 25 years for industrial application, we’ve had two different levels of originality (higher for ©, lower for design—need to meet a “museum test” for a design to get © protection). Now we have no idea: open Q whether French have won full cumulation or what the judgment means when it says there shouldn’t be a full overlap b/t design and ©, without providing any criteria for avoiding full overlap.

We find separability problematic—do you know what it is?

Designs and TM overlap: Article 4(e) says when a shape gives substantial value to the goods, you can’t register the TM, even w/secondary meaning. So what does that mean?

Dinwoodie: substantial value is unclear in meaning to him. British courts may think that if it is capable of being registered as a design then it has substantial value.  That would really exclude all shapes from TM.

Ohly: ECJ gave some criteria about design attractiveness: to what extent are features highlighted in advertising, and so on. But when a design becomes really well known as good, then it acquires secondary meaning and also may give substantial value to the goods.

Dinwoodie: there’s a gap between recognition and iconicness which isn’t quite recognized.

McKenna: but being recognized as a design is not the same thing as having secondary meaning.

Ohly: but they overlap.

McKenna: no, they really don’t.  Most widely recognized designs don’t have secondary meaning in the TM sense.  (Statue of Liberty, anyone?)

Dinwoodie: there is opportunistic framing by litigants, and the circumstantial evidence usually used might indicate both.

Du Mont: that’s the marketing function. We might not like that concept, but that was what they were trying to put their finger on: the difference b/t differentiation and secondary meaning.

McKenna: it’s entirely possible to be widely recognized as a design but not say anything about source as such. Eames chair: Eames designed it, but it doesn’t tell you all designs come from one source. But b/c we have such bad proof systems for secondary meaning they do overlap. 

Burstein: Dastar.

McKenna: Yes! Being recognizable as a work of design isn’t the same thing as source. 

Ed Lee: what we would do if starting from scratch for design: the system we currently have is odd and irrational to have overlapping forms of IP. Public interchangeably uses patent, copyright, and TM, and apparently so does design even though there are important distinctions we should be using. Traffix/Dastar type system to navigate the overlaps? Functionality tests are different. Remedies are also different: seems like you could elect statutory damages for © and disgorgement of all profits for design patent in the same case, even though you couldn’t otherwise get two forms of damages for the same conduct.

McKenna: remedies are very important: people who bring trade dress/design patent claims say that they want injunctions and damages, and TM makes injunction easier and design patents make damages easier, so it’s the best of both worlds for them. The Frankenstein monster system we have now is the result of many different forces. A lot of the overlaps are the result of deliberate attempts to expand certain types of subject matter b/c of perceived sense that other systems didn’t reach them—we have design patent b/c © and TM didn’t protect them, but then people tried to expand © and TM b/c perceived design patent didn’t protect them, and it worked. We have post-sale confusion b/c fashion is supposedly not protected.  That’s a different problem from the natural internal logic of a system inherently producing overlaps; the internal logic of the system is changed to make it look like a natural outgrowth but these were engineered overlaps.  That has ancillary effects on the operation of those systems in their traditional domains: manipulating TM so it can cover product configuration changes your understanding of what source identification is, what confusion is, etc. So we should ask how much damage we’re doing to the internal system in order to reach this area of overlap.

What are the practical solutions? Creating doctrines to police boundaries; that can be done different ways. Can exclude stuff from the system entirely (Dastar). We were stupid to incorporate product configuration into TM in the first place and have created all these doctrines to deal with the effects.

A sui generis system works only if you can identify its coverage in a reasonably clean way and exclude its subject matter from other systems.  The concepts have gotten blurred since Title II was on the table. Maybe this should be about “design of certain stuff” and not just “design.”

The linedrawing difficulty wasn’t inevitable outgrowth of the internal norms of each system, and we could productively rethink each core system.

Ohly: can the legislature anticipate these strategic moves? The EU unregistered design right was supposed to avoid that. 

Sprigman: the antitrust division of DOJ really didn’t like Title II. There is no evidence that these industrial designs need encouragement.

Felicia Caponigri: the leading case for fashion © in Italy was a case about moon boots; the reason court abandons the separability test is that it evaluates how iconic the design is and how it changed the evolution of the boot historically and culturally. Not saying this is correct, but when we think about overlapping rights, we may have to consider whether there are other areas of law that may be also affecting the definitions in IP law. Cultural heritage law, GIs, concepts of authentic/not authentic may be affecting ideas of copyrightability and nonobviousness. Expand horizons to other regimes.

Laura Heymann: courts are motivated by the sense that new subject matter should be in the regime.  Post-sale confusion (e.g. for Ferrari kit cars) is invented b/c courts think this should be protected, and they don’t yet have dilution. And then we get dilution formally later. There’s a sense that this is the right thing for the subject matter—how to address that.

Caponigri: the Ferrari, subject of the post-sale confusion case mentioned, was recognized as subject to © in Italy this last year.

McKenna: agrees too. If the D feels like a bad guy, the court may try to reconfigure the doctrine to get it. But that’s not the same thing as a straightforward application of the internal logic of the system.

Derclaye: the EC was trying to address the overlap b/t regimes with the unregistered design right, but naively thought that therefore © wouldn’t be used by designers any more.  In France, they are winning the most on different bases (unfair competition, TM, ©). Good intention, but left open the copyright/design interface and thus ensured cumulation.  Negative convergence for shapes solely dictated by function (if no mess is made by © reasoning in CJEU case on Brompton Bicycle) could do the trick.

Dinwoodie: the problem isn’t the Commission, the problem is the Court. The Commission left it open for Germany and the UK to do what they did.

Derclaye: well, they left it open for cumulation assuming that © would not be used. Naïve.

Buccafusco: we could do simple things to make design patents better: they could cost more; they could have renewals. But we might therefore push people into trying to use TM/© in scarier ways.  Trade dress cases over expired design patents: if we want Traffix for design patents, that’s great, but may also further push people into using other areas of the law.  Maybe we should treat design patents as not that scary, even though they are silly, compared to the other fields.

Burstein: will the line hold? Design law as safety valve. But it’s not clear that works—Europe seems to suggest not.  In litigation: if people think there are multiple claims, they are more likely to find liability.

Lee: it also matters that you get to go to the Fed Cir if you stick a design patent claim in.

McKenna: presence of multiple claims may let the courts approach the case in broader terms—may let you get away with weaker claims on each individual cause of action. Intent evidence may contaminate similarity analysis for design patent.

If you accept that courts in many of these cases that courts have a protective instinct, the most useful intervention may be to remind them that some other right is available. You don’t have to screw up TM law to get this; they can get a design patent. That’s one reason that destroying unfair competition as a separate doctrine has been bad: the existence of a lesser remedy was a safety valve to prevent TM from expanding. Booking.com: generic terms can be generic and there can still be a very limited remedy for passing off (Blinded Veterans). Reminding courts that just b/c the structure of a doctrine doesn’t cover conduct doesn’t mean that other doctrines don’t exist. Copyright expires for a reason.

Sprigman: but the grass is always greener for doctrines you don’t know as much about. Depends a lot on the structure of unfair competition law.  Antitrust law is one thing, but letting plaintiffs’ lawyers have free rein over unfair competition is also a problem.

Stacey Dogan: there’s nothing self contained about unfair competition law. To say it’s limited to passing off is descriptive, not normative.  The impulse to have unfair competition is itself normative, but the constraints on it (remedy is labeling only) are descriptive.

Mid-point discussants: Rebecca Tushnet

Balloon problem/grass is always greener problem: there is a rule about defamation that is a channeling device and it works a ton better than the IP channeling doctrines.  If you relabel your defamation cause of action—if the gravamen of your claim is for damage to your reputation, whether you call it intentional infliction of emotional distress or tortious interference with contractual relations or even trespass by deception—you have to meet defamation standards.  Smithfield Foods, Inc. v. United Food and Commercial Workers Int’l Union, 585 F. Supp. 2d 815, 820-21 (E.D. Va. 2008) (“[I]f a plaintiff seeks damages which are ‘reputational’ in nature, constitutional libel standards (i.e., falsity and actual malice) apply to the plaintiff’s damage claims. To allow otherwise would be to countenance ‘an end-run around First Amendment strictures.’ . . . [T]he label of the claim is not dispositive ….”). (TM is an exception to this b/c no one knows what goodwill means, but it’s only a partial exception given the limits TM law has placed on suing over criticism.)

We have a better idea of what defamation is for than what © is for, and US courts are strongly committed to a distinctive First Amendment tradition, so there’s definitely no slam dunk in making this difference stick. (But … if courts start to distrust causes of action, they can be pretty rigorous. Antitrust, RICO, even some TM claims.)  If you are really committed to what these doctrines are for, you can channel. But without consensus you can’t.

Carol Rose: Crystals and mud as a useful framing device?  Bad guys and sad sacks drive the change in focus of the law.  But we are at different points in the “cleaning up the mud” phases in different fields—©, TM, design patent.  This ensures that no crystalline rule can ever get established successfully.

The centrality of functionality: Star Athletica and functionality as the remaining limit. The potential dangers of saying it allows protection of functional elements (e.g. Buccafusco Lemley Masur piece): compare to the dangers of saying fair use is unpredictable. Courts may be able to do a lot with de facto and de jure functionality as in TM.

Relation b/t costly screens and biased screens in assigning the right in the first place: if there are thresholds, who will pass them (even assuming they have the money, which is also biased)? We have to understand that screens will disproportionately burden disadvantaged groups, although it’s also true that courts won’t be better.

Jason Du Mont: Might be ok w/a different functionality rule for design patent. Thinks of design patent as a baseline: if an innovation is going to go anywhere, it ought to be there unless it makes it into utility patent (or is in the public domain). Overlaps aren’t a problem in themselves. They tend to be problems of TM or ©.  History of design rights in the US: a pure gap filler, which was in the design patent system b/c Congress just listed topics with no conceptual coherence.

Registration systems: might require greater specificity. The Star Athletica © registration, for example, should have been better nailed down. That could contribute to resolving cases earlier. Not sure how to make that work in TM w/o getting rid of unfair competition.

McKenna: de facto/de jure functionality is not a strong distinction.

Buccafusco: that still means expensive fights about what is protectable and what is excluded. Everything is protectable and it’s about scope and it may all have to go to a jury.

Du Mont: © office has started rejecting things for originality.

Fromer: that’s mostly for logos—channeling to TM.

Du Mont: also rejected a float in the shape of a piece of pizza.

Dinwoodie: de facto/de jure could be used in interpretation of separability; could also be put into originality and idea/expression—102(b). Or are you suggesting a freestanding functionality doctrine? If we want a transubstantive functionality doctrine it might make more sense to do it freestanding.

RT: yes, could do it as a matter of 102(b) (facts aren’t listed in 102(b) but we screen them out anyway).

Derclaye: that’s what we do in EU.

RT: and these decisions are always made in context of a defendant who has done something: have they taken something copyrightable? So there will always be a comparison and functionality can be part of that.

Q: would aesthetic functionality count?

RT: yes.

McKenna: Fromer has made the point that courts have never taken seriously the Q of whether the intrinsic function of the article is anything other than to portray its own appearance/convey information—it’s very broad.  Discussion: There’s very little case law playing with aesthetic functionality and whether it counts as functionality in © (in © it can get deemed merger, scenes a faire, etc.).

Reaching consensus on the meaning of functionality is one thing; operationalizing it is another. Features are going to be aesthetic and functional at the same time. The question is where along that curve it has to be. TM says: if a drop of function, it’s out. Design patent: a drop of aesthetics, it’s in. Could have the same question about what function is and different places on the curve.

If you think of functionality as alternative designs available in the market, there’s no reason to distinguish utilitarian from aesthetic functionality.

Buccafusco: the amount we are scared of functionality has to do with the limiting functions of other doctrines in the field.  Functionality in ©, which is long and has limiting doctrines like merger, idea/expression, raises different issues. [Note that TM cases do say you can’t protect an idea with TM, though conceptually that puts the cart before the horse.]

Caponigri: aesthetic functionality as a protection against speech restrictions? People use these for communication.

Dinwoodie: Kalpakian says idea/expression protects freedom to compete, so it’s not hard to locate functionality concepts in ©.

Buccafusco: design patent cases essentially have idea/expression already: Richardson tool case; Ethicon.  Without saying so, the courts are saying that the design patent doesn’t protect the concept. Very narrow level of specificity.

Du Mont: push back a little: factoring things out is problematic.

RT: but substantial similarity is the exact same difficulty.

Du Mont: EU also does that with considering designer’s degree of freedom.

Burstein: Egyptian Goddess actually does a good job of that.

Buccafusco: we don’t necessarily know what’s going on in the field. But design patent data is available. One of the things we should be doing is figuring out what we’d like to know about, e.g., design patent citations. 

Burstein: the prior art citation is very bad; they cite URLs that then disappear.

Heymann: we also don’t know what’s happening in C&D world. Rules that change the effect of what people do before they get to court are more important.

Fromer: even w/registered TMs, it’s really hard to figure out what’s registered—how much on the register is trade dress? The coding is really bad! They don’t care about data consistency, maintenance, etc. We need to push them to code consistently and keep data.

McKenna: we’re asking for data about categories that have only recently been enforced on the law: trade dress as an ontological category is new and undefined—it’s all just stuff that wasn’t a TM, so there was no reason to distinguish packaging from configuration etc. The registration system reflects that—“trade dress” or even “configuration” is a recent phenomenon.

Dastar is a rule about channeling. Defamation is channeling based on the nature of the harm the plaintiff is claiming the defendant inflicted.  Most IP channeling rules aren’t like that. Dastar is an opening for that kind of channeling: if the nature of the harm you’re claiming is that people will attribute authorship of the content to the plaintiff, then that’s a © and © only issue. Would it be easier to configure rules around what the injury claimed is?

Dinwoodie: as RT says, the problem is that we need a coherent concept of the harm.

Heymann: misuse is a group of cases geared at the same kind of thing: you’re trying to enforce your © in a way that © is not for.

Mark Janis: note that the rhetoric of dilution crept into the Apple v. Samsung case—that’s not the kind of harm design patent is for.

McKenna: there was a dilution claim there too, and that may have contaminated the reasoning.

Burstein: that was also about gaming remedies; claiming they needed an injunction for the brand.

Du Mont: is it really a legit purpose of design patent law to let people acquire brand value?

McKenna: no!  Deeper Q: is the sharpness of the defamation rule coming from the fact that we really understand the harm of defamation and so can recognize a defamation claim whatever its label?

Dinwoodie: we can do better than subject matter channeling, but with less consensus on purposes of IP it will be difficult to get as clear as the defamation rule.

Caponigri: Is the harm of © infringement that there was harm to incentives, or is it a more Abraham Drassinower like theory of harm to expressive interests?

Burstein: SCt has said it doesn’t care about overlapping subject matter.

Sprigman: so shallow: how is design patent different subject matter than trade dress?

Dinwoodie: Kohler v. Moen says there’s a difference b/t an invention (faucet) and a sign (faucet).

RT: incentivizing different things/preventing consumer confusion as the cores of patent & ©/TM respectively?

Discussion about how this is a tough and abstact line to draw, and also law & econ has contaminated the idea of incentives. RT: thinks the Trademark Cases are nonetheless good law.

Future directions:

Dinwoodie: the role of transubstantive rules (the flip side of cumulation). Some of us think there are values to that, but we might think about the merits of such rules and design is a good vehicle for that Q generally.

Fromer: the extent to which we use the same terms differently in different fields—Burstein’s faux amis.

Derclaye: Overlaps and remedies. 

Sprigman: Unfair competition as safety valve. Specify doctrinal soft spots of each IP facets that could be eliminated in favor of a meaningful unfair competition approach, e.g. post-sale confusion. Catalog where unfair competition could be useful in solving/reducing overlap problems. There was a theory of unfair competition once upon a time; could that be recovered?

McKenna: people use IP claims to complain their markets have been disrupted; they also sometimes use antitrust or other business torts. Many of those state tort doctrines have recognized the soullessness of those doctrines and built protective doctrines: you can’t bring a claim for tortious interference w/o identifying a separately tortious act—you can’t just show competition. It’s not recovering the theory, but recognizing that we lost a useful tool and reconstructing it for current purposes.

Sprigman: antitrust’s standing/injury doctrines are meant to do that. It puts the plaintiff to an early showing that injury is w/in the zone of interests.

Dinwoodie: Lexmark as innovation towards a similar thing?

RT: Courts let TM claimants get away with standing/injury very easily [b/c we don’t know what goodwill is].

Buccafusco: should think of other areas of law that shape product design: disability law, transportation law, environmental law, etc.