15th Trademark Scholars’ Roundtable: Session 1: Congress and the Courts (including the role of the Supreme Court)

Introduction: Rebecca Tushnet

What might we derive from things the Court has said about trademark of late?

The purpose of trademark is consumer protection; source-identification is the most relevant consideration but not the entirety of TM law.  [see also Booking: once source identification is shown, all other issues are dealt with as limitations on rights/incorporated into the confusion test.]

Maybe to this we can add in Dastar. Source/origin and control over quality is the core of TM and has special relevance when trying to figure out what rights the plaintiff is claiming to own; sponsorship/affiliation is of lesser importance, albeit still relevant; sponsorship/affiliation might be better handled under 43, which could have stronger guardrails; consider Lexmark as a case about 43(a), which the Court says it is, and not about 43(a)(1)(B) only. In that case we have a zone of interests test and a proximate cause test—adding in a harm requirement at least for claims that sound in 43(a). That could also address some of the Article III standing issues I’ve been encouraging people to raise.

Compare what’s said about source/TM use in Abitron; use in commerce as a separate requirement after Abitron: bona fide is going to give us some trouble but not necessarily a lot since it is explicitly contrasted with not reserving rights, so it could also help identify non-source identifying uses.

I was particularly interested in doing a deep dive on the idea that the Court thought that both Dogan & Lemley and Dinwoodie & Janis were right, when most of us understood the two articles to be debating with each other.

Dogan & Lemley: This article, especially the passage cited by the court, seems to envision a “trademark use” requirement less as an affirmative test for infringement, and more to exempt non infringing uses from liability for an absence of use.

Dinwoodie and Janis taken out of context? They argue that trademark uses are more likely to infringe, and non-trademark-uses are less likely to infringe, but that using this dichotomy as an “approximate” in the interest of efficiency is outweighed by complications that render individual cases (specifically in the example of keyword purchases) distinct. But if Rogers survives, that’s what it does. Meanwhile, valuable speech that is source-indicating gets some variant of the regular LOC test.

The ridiculous approach to dilution: because one statutory provision says use as a mark, the neighboring one that doesn’t use that language must also have that concept: As Graeme Dinwoodie said to me, even if you don’t agree that expressio unis should mean that the noncommercial use exception doesn’t have a use as a mark requirement, you should still easily recognize that the 2006 revision of dilution was designed to expand the defenses, not contract them.

What if we could read that part of the opinion in ways that were less awful?

Use expressio unis to get at the fact that the noncommercial use holding is limited to situations where the challenged use is “used as a mark for a commercial product”; the term noncommercial use can reasonably include uses as a mark for political speech—Robert Kennedy, Jr., is using his father’s famous name to promote himself politically (also highlighting the importance of small differences when it comes to speech in the political realm). Things can be source identifiers, without being commercial source identifiers (United We Stand): confusion but not dilution actionable

Jack Daniel’s didn’t purport to decide the full scope of the “noncommercial” exclusion. It held:

On the [Ninth Circuit’s] view, the “use of a mark may be ‘noncommercial’ even if used to sell a product.” However wide the scope of the “noncommercial use” exclusion, it cannot include, as the Ninth Circuit thought, every parody or humorous commentary.” However, the Court concludes: “Today's opinion is narrow. We do not decide … how far the ‘noncommercial use’ exclusion goes. … On dilution, we hold only that the noncommercial exclusion does not shield parody or other commentary when its use of a mark is similarly source-identifying.” (Reference is to the use of Bad Spaniels as a mark for dog toys). Given that the Supreme Court has also cautioned not to read its decisions as if they were statutes, it’s arguable that use for political speech should fall into the noncommercial use exception, which courts have generally read as covering “noncommercial speech” as that’s defined by the First Amendment. 

What about the use other than as a mark limit in the parody/commentary exception itself? Is there any hope there for parodists? Other scenarios: Howard University Alumni Association should be understood as NFU outside the scope of dilution, as long as it’s a truthful representation—it’s not using Howard as a mark but rather the unitary mark Howard University Alumni Association; the exclusion doesn’t, like other parts of the statute, refer to uses “comprising” the famous mark. There is something intuitive about the idea that someone who names their organization “Americans Against Trump” is not “using” the famous trademark “Trump” “as a designation of source” for their own goods or services—they are instead using the phrase as a whole to express opposition to Trump, which is the opposite of using “Trump” as a source identifier.

Mark McKenna: TM is a creature of common-law courts and we don’t live in that world any more. SCt used to do private law dispute resolution: Kellogg, Mishawaka, Singer—it’s treating it as a tort. Scatter plot: common law reasoning doesn’t look to deduce from a statement of a rule down; you build a rule by accretion over time and each case is a data point. The line you draw never captures all the lines; sometimes you see multiple lines emerge and then you can evaluate why. But law doesn’t develop top-down in this model. For the most part, that’s how we built TM and unfair competition.
SCOTUS is now a constitutional court, not one that resolves individual cases, and that’s now apparent in all these cases. Politics of how SCt Justices are chosen now matters too: they aren’t picked for being good at law generally, but for views on hot-button issues. Most have never really been lawyers. They were movement/administration/academics. They’re not doing ordinary judging.

But when we have a statute with the common law model baked in—tells you very little about what it means to violate the rights at issue—the Justices now have a very different idea of what they’re doing, and they think they’re doing textualism even when the statute is misfitted to that, they don’t have the toolkit to do the common law reasoning. Romag is a clear example: there’s a complicated history, and figuring out how to think about it in light of TM’s evolution over time is hard. But the Court says: we found this one case! 1881 DCt of Alabama. The idea that finding one case in common-law reasoning matters is ridiculous. How can you even engage the conversation when it’s a trump card to say “that can’t be the rule b/c I found one case that says something else.” You can also see it in Tam & Brunetti—little interest in how TM works, the relationship b/t registration and enforcement, even though that’s central to the Lanham Act.

Also JDI and Abitron, where the Court grabs a bumper sticker: focuses on one term in the law. Citing both Dogan/Lemley and Dinwoodie/Janis is a good example: those articles don’t say the same thing.

What is to be done? SCt should get out of business of doing IP, but it’s too late. Even from 20 years ago, it’s a big change—Wal-Mart, Traffix, and Dastar were trying to resolve hard questions and details of doctrine that affect how things work on the ground. The 20 thousand foot view of today is a very big difference. They’d look for some little snippet of text in the Lanham Act; Scalia was a textualist but he could read a law in its entirety. The way we’ve trained a generation of judges to think about law is so different and the suppositions about how law gets done baked into the Lanham Act don’t work any more.

Christine Haight Farley. The Interamerican Convention enters the chat! That’s a missing link that helps explain why so much is unstated in the Act. Rogers had been trying to pass a law, and he didn’t think Congress could enact substantive rights; he was given the opportunity to write a self-executing treaty that was written like a statute, for civil law countries. When the Lanham Act passed, he was convinced it incorporated that already in section 44.

The accident that is 43(a) is also part of this story. It was meant to have a bounded agenda; only in legislative history, when witnesses noticed section 44, they suggested that 43(a) was more appropriate for unfair competition. Rogers would have been floored by what came of it.

Congress delegates to INTA, AIPLA, ABA the role of drafting, and that goes all the way back to Rogers. How did TM use get into an exception to an exception? It was supposed to be a limitation on the dilution right, but INTA said it should move and Congress was none the wiser about the effects.

Codifying 43(a)—but without codifying LOC, spectrum of distinctiveness, etc.—b/c the TM bar sensed a risk of loss, of limitation. No one in the TM bar has an interest in cleaning up things like that, but it means nonexperts don’t get much guidance.

Also helps explain 1962 expansion of confusion; 1988 constructive use; dilution; ACPA. From 1988 on, Congress is doing substantive law, and special-interest revisions like the Anti-Monopoly revision.

What about the courts? The Court seems to think TM law is working really well. There might be little problems but overall it’s a well-oiled machine. It doesn’t see what dramatic changes have occurred in TM law (or that they’re a problem). Only when it’s really doing private law can it see a plaintiff’s overreach. Doesn’t have a sense of the policy landscape. Breyer, Ginsburg, Scalia had more of that sense; Justices now focused on different landscapes. Elster might be a way to get some of that landscape back. Gorsuch is interested in history of old Acts and TM Office doing public policy that’s a little broader than source/origin confusion. [Although he seems wrong on the facts about what they were doing.]

The working relationship of Congress/SCOTUS provides an opportunity to hide the ball about change, especially in 43(a): Congress says “all we’re doing is codifying common law” and Court says “all we’re doing is applying what Congress wrote.” But somehow there’s a big change between Sears/Compco and Taco Cabana. Sears/Compco said there was a right to copy things in the public domain; how did that go away?

Dinwoodie: Non TM use is an important variable—something that affects scope of protection not outer boundaries, and that worries him less b/c it’s not a dispositive binary variable. But now you have to analyze it like that (which you see badly in the post-JDI cases); if you fall outside out of the line, the consequence is just in the test/scope of the right. You can see that where we have use as a mark in other tests—KP Permanent—courts have started to use the three elements more as a spectrum—how descriptive is this? We might be getting to the stage in the evolution where TM or non-TM use is a relevant variable, which is consistent w/both articles. Agree that source/affiliation is one way to draw the line, but that is itself not an easy line to draw—standing behind is not an easy line, but as a variable it is less worrisome.

Rescuecom: is it overruled? Only if you believe they’ll stick with their statement about the definition of “use in commerce.” Which he doesn’t; the opinion applies that reasoning to 32 and 43(a) both. Nobody thought that through—it was a bumper sticker, as McKenna says. Rescuecom will survive a bit longer.

[I think you could say that when you claim to have a “mark” subject to 43(a), you use the relevant definition, but when you’re asserting some other claim under 43(a), you don’t have to do so.]

They seem to think the multifactor test is mandated by the statute, but it’s not, so maybe it’s the statute + some principles drawn from private law? Once you go top-down, it’s hard to figure out what other nonstatutory stuff will be in play.

Do we have faith in judicial innovation? RT’s approach is a bit more top-down but we’ve had some success with bottom-up—do you still make the point that the 1962 legislation was trying to do something specific or do you give up? Do you still point out that Congress endorsed Rogers in 2020? Or do you accept the corroded structure and do something else?

Jeremy Sheff: Not just judges trained this way now but how lawyers are trained, the materials they have, and the way they are trained to use those materials technologies. The clerks who wrote Wal-Mart were the first generation to get unfettered access to Westlaw. Before, you pulled a treatise off the shelf who’d been written by someone who read all the cases and condensed them to explain what they mean and which the most important ones were. JDI, which cites articles with “TM Use” in the title, is now the model—that’s how lawyers work now, dealing with huge volumes of materials that they can’t wholly assimilate. The scale is different; the human reviewing cases and synthesizing them is not the model now. We’re encouraged to look for the snippet that matches our claim and no further. This issue is not going to go away.

Similar with registration agencies—a firehose coming at them. Better algorithmic detection/AI won’t necessarily be a better way to deal with that. We should think about how we could train lawyers to be better at this.

Mark Lemley: This isn’t a TM problem. This is a law problem. The Court, across a wide variety of doctrines, isn’t just replacing common law with statutes even when statutes were written with common law in mind. It has embarked on a project of abolishing equity. Not clear why, but definitely doing it, e.g. in Romag. There’s one basic rule of disgorgement: you need to be a bad actor. But the Court doesn’t look at it that way.

Yesterday’s © arguments are the same. Ten years ago they abolished laches, creating a nightmare of 40-year-old claims. Yesterday’s argument was whether the discovery rule applied to the 3-year damages rule; several Justices across the ideological spectrum asked “why should there be a discovery rule in general?” Equity existing alongside the statute is being taken outside the law. Not clear what to do, but you can’t solve it in TM because it’s not a TM problem.

Dinwoodie: maybe it’s worse in TM, b/c you don’t have political commitments at least telling you what will happen.

Lemley: yes, part of the issue is you’ll see a random walk. They weren’t particularly interested in helping TM owners get more $, but they were interested in identifying a couple of words in the statute. Star Athletica is the same way; in the absence of a precommitment around constitutional issues, it’s weird. When the statute was written will make a huge difference too, b/c Congress has responded by writing statutes like the tax code. The earlier statutes will be interpreted in weird ways, see Warhol, b/c they were written against a backdrop that we now pretend no longer exists. Is the solution a European style detailed regime? Don’t love that in TM but that is the obvious response.

Jake Linford: Thinking about Rogers, JDI, and Dastar: is there a throughline about the line b/t © and TM? Maybe JDI is suggesting subtextually that there’s a place for Rogers, and what we mean is works that could qualify for ©. [But the VIP bottle clearly could!] It’s about maintaining lines and Rogers then helps maintain the lines. If that’s right, we use expression as another form of heuristic, like use in commerce/use as a mark. The danger is distortions of the parody/satire line in ©.

Standing makes him nervous b/c of privacy. The SCt uses standing to take power and ability to get relief away from plaintiffs it doesn’t like. TransUnion is a big example; Congress didn’t satisfy Court’s wisdom about who should get relief. Would be surprised, given Tam/Brunetti/JDI/Elster trajectory, if Court said we think TM plaintiffs generally ought to be denied relief based on our gnomic understanding of standing.

What you’re seeing is cross-licensing/branding. A move among many of us as scholars to say there should be space for artistic uses, and if we see a lot of Nike collabs and make statements about who it wants to collaborate w/or not, perhaps that sends signals that matter to consumers, and if we want to reflect how consumers interact w/markets, that might lead to rights expansion. Maybe sponsorship/affiliation reflects what brands are doing and how consumers think.

We’re trying to return to a “pure” state of common law: source confusion, double identity. Not sure that’s a better world, but regardless the mechanisms we’re using aren’t fit for purpose, like the First Amendment or standing b/c we think they’re more essentialized and the Court might listen to us about precommitments like that instead of TM policy. Perhaps you can get that, but do we lose something in masking the policy intuition behind a technical tool to get us there.

On Sheff’s point: if you trust the treatise writers, then great. If you don’t, does the search engine help? Maybe, maybe not.

RT: On ©. The bottle is ©able.

On standing: Agrees about prediction that it won’t be much help, but: What is law for? If it’s not for making reasoned distinctions, then it’s nothing. And there’s some value in explaining the inconsistency and challenging the TM harm stories; some judges even listen.

On sponsorship/affiliation: courts don’t inquire into how it’s done, though. If we listened to what brands actually do and what consumers actually see, they see Nike x Common, the dual branding is all over it. But brands don’t signal a collab by having the red wax seal on one bottle of liquor appear on another bottle with no other sign of collaboration. They put it all over! Courts actually ask whether consumers would, e.g., expect a movie company to do cross-branding with a software security company, but they don’t ask, is this how consumers are accustomed to seeing collaborations identified?

Jeanne Fromer: participated in drafting sessions for TMA with Farley & Beebe—were able to advocate for some changes. Allow PTO to unilaterally cancel marks. Another issue: the SG’s office is critical. They often come up with strategies to win that don’t make sense in the larger context of the body of law at issue.

Dinwoodie: SG takes positions they think the Court will listen to, and that methodology bleeds over into other things/methods.

Fromer: hard to know the chicken from the egg.

McKenna: The SG’s JDI brief made egregious legal errors, hope they’re not going there.

In teaching, try to teach students where the rules came from, and maybe that’s what we need to be doing more of. The rules don’t drop out of the sky but evolve. Another big difference along with legal research is that the TM bar used to be small; now the money is good so lawyers not really trained as TM lawyers are litigating these cases; generalists that pick & choose things—you hear that in the arguments. The SG wouldn’t have put in a brief in older TM cases; the Court asks the SG b/c it’s now thinking of itself as a constitutional court.

Dinwoodie: judge discussed how he would have benefited from learning more about functionality from parties; he looked at the dictionary definition.

Ramsey: standing and 1A seem to have a relationship in examining government’s interest and harm/causation stories. Jennifer Rothman has written a paper about history of protecting names of presidents, etc.; may be source of Gorsuch’s interest. So scholarship about history may be important.

Robert Burrell: Australian trend—lot of $ coming in to TM, and so nonexpert lawyers are coming in and litigating, often to damaging effect. See less in UK w/specialist bar and specialist judges. How do we as a community flag “here there be dragons” or at least “there are other cases that have considered these issues”? We can do it in part by embarrassing judges when they get it really badly wrong, but how we persuade other members of the legal community that they might want to find some things out before giving it a go is a question.

Farley: Court loses forest for trees in these cases, and then sees only a cartoon forest. Could we be painting more accurate pictures of the forest in these cases so they see where a case sits and what the impacts will be? Scholars used to write short “forest” pieces, and one of the ways the world has changed is that law reviews like you to solve a problem/make a recommendation, but there’s room to speak for the trees.

Mid-Point Discussants: Jessica Silbey

Compare what’s going on in patent & © specifically: feels better about TM cases than about patent and © generally speaking, which is worth thinking about. The only © case that went well recently was PRO v. Georgia, but © was bad before—Eldred, Golan, Tasini. We did have some thoughtful TM decisions—Qualitex, Traffix, even Two Pesos, Moseley might have been a mistake but got fixed by Congress. But still evidence of hubris: armchair consumer psychology in Wal-Mart etc. Nostalgia isn’t accurate as to cases like KP Permanent, which were doing policy too. So how does the Court do IP more broadly?

Methodological shift: deciding cases on administrative agencies, jury questions. It will be busy making claims about separation of powers, role of jury in rights/property claims; the idea that won’t have effects on statutory schemes like TM is ludicrous. It will affect the TTAB. The cross-disciplinary effects of its methodological decisionmaking are profound. So what the court says v. what it does: SEC v. Jarkesey—it’s going to say that the ALJ cases that sound like common law should be brought in court before a jury. What is the role of the jury in TM? We can talk about courts’ power, but will it throw everything to the jury? Full of its own self-worth as judges being gatekeepers. Judicial supremacy: this Court is all about its own power, so it will say juries are important, but really it is creating more gateways for it to decide cases. Where is factfinding required? Distinctiveness, abandonment, etc. Where’s the influence that advocates might be able to exert on the divide b/t law and facts?

Appellate courts: there will be tension b/t Fed Cir and regional circuits. As Fed Cir decides IP cases v. 2d/9^th, will we see different kinds of SCt decisions w/r/t Fed Cir? Insiders versus outsiders: con law phenomenon is spreading. Opportunity to figure out who those insiders are, including the SG.

Helping TM to be its best self, whatever that might be.  Provocation: TM restatement! Less controversial than ©; who gets to decide gap-filling questions and what justifies them being the decisionmakers? The development of the common law in TM is its own force, and that calls out for the idea of competing treatises, which a Restatement could be.

Bill McGeveran

Who is a good guardian of public interest in things like speech and competition? A common law approach assumes judges will guard the public interest. A delegating structure assumes that admin agencies will do so. More specific statutes now are Congress reacting to those responses and trying to embed a view of the public interest into the statute itself. The Lanham Act is clearly a product of common law, but Congress passed plenty of other statutes in the 50s, 60s, and 70s. Where is the public interest being guarded in those statutes? Antitrust is another capacious common law statute, but there’s been enormous normative struggle—Chicago School monopoly definition of monopolization, and then countermovement. Big, dominant, coherent approaches to the public interest permeate antitrust lawmaking. Also, FTC and DOJ really matter. SEC is similar: the regulator really matters. The text of the statute doesn’t cover all the regulations, where the public interest is guarded. Privacy/consumer protection: FTC has jerry-rigged a guardianship of the public interest; California also passed a specific law with a specific agency (which Eric Goldman loves). There’s an understanding of where the public interest is supposed to reside.

What about IP? The statutes have been amended to make them more and more specific; © and patent have had comprehensive rewrites since they were created in the common-law era. But they all have this issue: where is the public interest in a common-law structure implemented by the statute? It’s supposed to be the courts, b/c the statute doesn’t embody the specifics. But there aren’t a lot of other old-fashioned statutes with gap-filling common law that are operating anymore. The registration process has some of that, but registration isn’t the primary or only mover here. Nobody has their hand on the tiller.

RT: [Point about sending everything to the jury: it’s easy to say that won’t happen, because there aren’t enough courts and juries, but another possibility is that the system really won’t work/we’ll get decisional paralysis: Indian judicial system has seized up—in practice, you just don’t have rights. Will businesses be ok with this? Compare responses to mass arbitration—they do have incentives to change the law when the law stops working for them.]

McKenna: there’s a big difference between wholesale judgments about what’s on the whole more likely than not—Wal-Mart and the other classification cases—we make those wholesale judgments b/c of the countervailing concerns like competition; very different from the Court’s current approach, which is that everything is retail and has to be decided using concepts that won’t be filled in by the Court. Very different concept of institutional actors. Generalization is itself valuable, but that feels like it was abandoned.

Registration’s getting more important, so maybe the PTO will be more of a guardian of the public interest, b/c Amazon has forced that role on them.

Sheff: Not sure the allocation is the core problem; we’re interested in the outcomes/the system and it’s not clear that who has decisionmaking authority is controlling that.

Dinwoodie: dissent in Samara in 2d Circuit: don’t let juries decide how competitive the economy is going to be. Worry about juries—think about who’s in the best position to do the weighing/it’s easier to control the judge’s interpretation of the law.

Surveys: A growing skepticism of surveys? Sotomayor’s concurrence; others may well be on board. In the rest of the world, common law in particular, there has been pretty significant trend against the use of surveys. Maybe that’s one of the issues where the way you develop facts and prove facts may affect outcomes. JDI survey did a lot of damage in that case.

Linford: we don’t have instructions for TM cases in most circuits; that is a leverage point. Court-assigned experts as in Germany?

Silbey: public interest location: in © different industries have forced amendments over time; that’s somewhat policed the different stakeholder interests. What is the equivalent in TM for that kind of industry diversity that could create wedge issues? LKQ/repair parts/must-match for trade dress. Off-ramps are hard to convince Court of when they think that courts are the best places for deciding these questions and they do it well.

Where are the lines the Court is drawing that take it out of fact questions? Commercial/noncommercial? What other things are available doctrinally? 1A is clearly a line; property rights is another line. We’ll see more takings cases. Note that Feist, a huge © case, turned on Stevens being mad at telecom companies. O’Connor wanted to deny cert, then wrote the unanimous majority. Interesting to see the case file in Wal-Mart. Where did Scalia get those ideas about the balance of the system?

Dinwoodie: public interest considerations factoring into the analysis can be a means of constitutional avoidance, but the Court seems unwilling to do that—forces it to the nuclear options.

Farley: In JDI, examples were effective: Blatt said the survey should be relied on because we don’t want judges deciding these questions because judges don’t understand anything about whether panties are related to teen girl magazines.

Do plaintiffs waive jury trials? [PI/SJ is probably most important.]

Lemley: sometimes it’s because all you want is an injunction/a PI resolves the case. If it’s a disputed factual question it’s not appropriate for PI, because who knows what a jury would say? That framing, if you can fit it into PI jurisprudence, which the Court is messing up as it abolishes equity, could serve as a bulwark not in speech cases—where you do want to lower costs—but in other cases where the D might otherwise just be shut down.

Farley: INTA says its objective is to protect the public through the use of TMs/avoiding deception.

Eric Goldman: the platforms could be a countervailing force. But they can’t. They’re sidelined by the techlash. Google fought for keyword advertising but couldn’t win that today.

Burrell: Judges are pretty good about “this shouldn’t get monopoly protection in the first place,” and that might be the 90s cases. But once they’ve decided “Jack Daniel’s is a trademark” they get much worse at the scope of the right. Could be a shift in the types of cases.

Dinwoodie: British judicial enforcement of TMs until 1994 was much more balanced view of range of interests at stake in TM: often former TM barristers that wrote/enforced law. They had represented clients on both sides. More sensitive to balance. We’ve lost that.

Ramsey: worked on INTA’s model TM guidance; really interesting—a lot of them were fine with speech-protective limits for registration and enforcement. Put into the language of functionality “any feature that adds substantial non-source-ID’ing value” but someone on the board of INTA didn’t like it and so it came out. Compromise is possible, but can be derailed by 1-2 people w/a lot of influence. Restatement model might be better to avoid those roadblocks.

Failure to function as a guardrail for the public interest, not just for speech but for cases like Lettuce Turnip the Beet. But we need some clarity about when secondary meaning should be enough to overcome failure to function, as with Nike/Automotive Gold, and when it shouldn’t be, as with DRIVE SAFELY. [Maybe part of the answer is preexisting meaning: there will always inevitably be non-TM meaning for “DRIVE SAFELY” but not for Nike/the BMW icon, and the ratio of TM to non-TM meaning, or the ability of TM meaning to drive non-TM meaning out of the market, might matter.]

Linford: Dastar as response to Court’s recognition of Eldred/Golan being extreme—strongest statement that © can’t be perpetual is found in Dastar, not a © case.

McKenna: it’s true that the SCt cases we describe as more sophisticated are about policing what counts as a mark, but it’s never taken a straightup infringement case as the massive expansion of what counts as infringement occured.

There was a settlement over validity v. infringement doctrines but TM use puts that in a blender. We’ve committed to a system where everything is in the scope of TM rather than using public interest to set boundaries for protection, and that makes it hard to figure out how to use the public interest in the retail inquiry.

Dogan: consider differences between arbiter of public interest and advocate for public interest. INTA represents a consistent value, whereas on the other side there are different interests and different ways in which expansive TM rights threaten values we care about. There are clusters of interest groups for competition, and for speech, in uncoordinated ways. Think about whether there’s any way to coordinate and make more coherent this set of values threatened by expansive TM rights. Courts and Congress respond to what they perceive as broader cultural norms—SOPA/PIPA; to some extent in courts protecting Sony VCR. How to make values salient to broader public? Economic costs, speech costs, etc.

Farley: one challenge is people don’t know when they will become interested. Small businesses face particular challenges. No representatives of small business in dilution hearings.

Leah Chan Grinvald: right to repair as one of these, fighting for 15 years.