Wednesday, April 18, 2018

Leval on fair use


At a talk today (should be posted in about a week), Judge Leval gave a beautiful explanation of why good faith/bad faith shouldn’t matter to fair use. Boiled down and stripped of its eloquence: (1) A publisher should be able to examine the relevant works side by side and determine whether the use is fair, without investigating the good faith of the second author/the means by which the second author acquired access to the first work. (2) The inquiry should be done on the pleadings/summary judgment, and considering good faith makes that difficult.  (3) The ultimate interest in fair use is the public’s, and if the public is entitled to the new work/new use and benefits from it there’s no reason that the defendant’s means of access should matter.

Judge Leval also insisted that the fourth fair use factor was the key, even more than factor one, and that therefore the result in TVEyes was correct.  I wish he had addressed the record evidence in that case that Fox didn’t license everything—it put video it was later embarrassed by in the Memory Hole.  That seems to me quite relevant even from a market-focused perspective, especially since people looking at news shortly after broadcast, when they’re scanning for topics, can’t be sure what moments will later be awkward for Fox to acknowledge.

Monday, April 16, 2018

allowing TM to be used to describe a different entity could be literally false lawyer advertising


Rosenbaum & Associates, P.C. v. Morgan & Morgan, 2018 WL 1768050, No. 17-4250 (E.D. Pa. Apr. 12, 2018)

Rosenbaum sued various Morgan defendants for false advertising of personal injury lawyering services in Philadelphia.  Defendants started advertising Morgan & Morgan’s personal injury services in the Philadelphia market in December 2015, though no attorneys worked on personal injury matters for a Morgan & Morgan entity in a Philadelphia office. “At some point, an entity called Morgan & Morgan Philadelphia opened an office in Philadelphia but Morgan & Morgan PA denies it has an office in Philadelphia.”  In January 2017, Morgan & Morgan PA hired a relatively inexperienced Pennsylvania attorney as the only attorney in its Philadelphia office, but this allegedly merely occurred “to perpetuate the charade [all Morgan & Morgan entities] are handling personal injury claims for clients in the Philadelphia market.”

The disclaimer “it is not a referral service” was allegedly false or misleading because the Morgan & Morgan entities, in reality, refer all or nearly all of cases from Philadelphia clients to other law firms.  Billboard claims stating that “Morgan & Morgan” has “Offices Philadelphia” were allegedly false because Morgan & Morgan PA denies having an office in Philadelphia and instead states the separate entity Morgan & Morgan Philadelphia owns the Philadelphia office. Similarly, internet advertising that “ ‘[Morgan & Morgan’s] attorneys in our Philadelphia office handle cases in the following practice areas,’ referring to various types [of] personal injury claims, and identifying [Morgan & Morgan] as ‘Personal Injury Lawyers in Philadelphia’ ” were allegedly false for the same reasons.

The Morgan & Morgan entities and attorneys allegedly never intend to represent personal injury clients in the Philadelphia area but refers “all or substantially all” potential clients who contact Morgan & Morgan to another law firm for a referral fee.  The statements "I'm your lawyer,” “We’re all here for you,” and “You don’t pay us unless we’re successful” were sufficiently alleged to be literally false because Morgan & Morgan PA attorneys never represent, never are there for, or never are successful for Philadelphia based clients.

After discovery, Rosenbaum alleged additional statements are false or misleading and also sues other Morgan & Morgan entities and individuals.  The court refused “to examine each statement in a vacuum and strike Rosenbaum’s allegations about certain statements because those statements are not misleading in isolation…. The full factual record of summary judgement is the time to examine each advertisement in context and determine whether they are false or misleading instead of examining the statements in isolation based solely on allegations.”

Morgan & Morgan PA moved to dismiss as to new claims, arguing that billboards advertising “Morgan & Morgan Offices Philadelphia” weren’t literally false because “Morgan & Morgan” isn’t “Morgan & Morgan PA.”  The court disagreed; Rosenbaum alleged that the billboards in the Philadelphia area were literally false when they used the trade name “Morgan & Morgan” through the trademark owned by Morgan & Morgan Global to advertise an office in Philadelphia when Morgan & Morgan Global and Morgan & Morgan PA deny having an office in Philadelphia.

As for which defendants were responsible, Morgan & Morgan Global allegedly permitted its trademark “Morgan & Morgan” to be used in billboard ads stating Morgan & Morgan has an office in Philadelphia when allegedly neither Morgan & Morgan Global nor Morgan & Morgan PA had an office in Philadelphia. Allowing the trademark to be used in an ad with a literally false statement was sufficient to state a claim.

TM scholars' roundtable, part 5


Session 3: The Implications of Tam (and Brunetti)

What are the implications of the Tam Court’s First Amendment analysis for other subsections of Lanham Act Section 2?  What are the implications beyond Section 2?  For example, is dilution protection now vulnerable?  Did Brunetti correctly apply Tam?

Introduction:   Lisa Ramsey
Source-ID function v. other functions; the Court has to decide whether Slants is commercial/noncommercial, and expressive component is noncommercial; as to the commercial, Central Hudson would apply. If your mark is only conveying source ID information, Central Hudson applies.  Regulating expressive component, then strict scrutiny required.

But expressive and commercial are not mutually exclusive.  Descriptive terms can be both.  Comparative advertising can be both. 

Registrations can also chill speech. The Court needed to weigh the harm from registration and denial, but the Court didn’t do so.  Still, can apply the analysis to registrations: a registration for pink buildings can chill others’ use of pink on buildings.

Vague TM laws: infringement, dilution, secondary meaning.

Court rejected theories that would take TM/registration out of the scope of the 1A: government speech.  Could have made a categorical exclusion, but it didn’t do that.

Viewpoint v. content based. Is that distinction made because of the viewpoint of the lawmakers who enacted the laws?  Protecting against confusion isn’t disapproval of the content of the mark itself.  But other cases like Reed say that you don’t look at what the law’s motive was, but rather what the law does—it targets particular content and is thus content based.  You have to compare the speech of the P and the speech of the D.  The remedies target content.

Important to get this right.  There’s a debate about whether © is content based; it is.

Level of scrutiny: Court doesn’t decide what kind of constitutional test should apply in the TM context. Alito + 3 says won’t decide whether TM also regulates noncommercial speech; doesn’t even pass Central Hudson scrutiny. Process: identify substantial gov’t interests at stake, and determines whether laws directly further those interests, and then figure out what’s the harm to free speech. Thomas says strict scrutiny even if it’s commercial; etc. – strong signals.

Rebecca Tushnet: with Ramsey present I don’t have to be the optimistic advocate for working the law pure, so I can be the cynic.  Should Tam be understood as death blow for dilution/tarnishment and more?  Tarnishment seems obviously gone.

Incoherence in Alvarez about whether dilution is even distinguishable from confusion for constitutional purposes/lack of disavowal of USOC v. SFAA.  Can the legislature make wholesale decisions about confusion or must it work retail?  The various parts of section 2 very much depend on that if the interest at stake is only consumer understanding/confusion—e.g., prohibition on  registration of flags & insignia.

Separately, could dilution survive b/c confusion isn’t the only interest the gov’t is allowed to protect?  The appeal of a property concept (and its interaction w/a registration system—here is perhaps the weak point from the perspective of the libertarian-esque Justices; the rights created by registration are very clearly additions to the common law background and thus not “natural” in the way that rights stemming from use are).

Registration system as a whole and Central Hudson: the only interests that a registration system more directly advances than common law are (1) treaty compliance and (2) notice/administrative convenience functions. 

Tam allows you to avoid all this with viewpoint analysis, but Brunetti does not because it decides that scandalousness is content based, which is bass-ackwards given that scandalousness is far more obviously viewpoint based than disparagement, which applied to disparagement from the viewpoint of any targeted group.  I don’t think that “being mean” is a viewpoint (though those who know me might disagree).

Linford: On the treaty point, Eldred & Golan might allow the Court to say that Congress gets to make decisions in light of treaty obligations.  Maybe arguments are stronger for dilution/tarnishment, but their decision to wave off treaty issues in those © cases might signal a willingness to do the same.

McKenna: TM law is so coherent compared to 1A law.  What are the legitimate interests protected by TM? We are probably going to see cts assert that protection of property rights is a legit gov’t interest/water down scrutiny. But if that’s the case, then we lose Rogers, all the expressive use considerations—there’s no 1A need b/c those doctrines are meant to protect 1A interests.  It’s hard to conceive a way to understand these cases that doesn’t do sweeping things w/in TM unless we start making a bunch of distinctions w/o a difference. E.g., Tam: why doesn’t the Court approach this as a gov’t program; Alito says “those gov’t programs from previous cases are all about cash” (which isn’t true) but even if it were true why would it make a constitutional difference?

Dinwoodie: even more skeptical.  The doctrine seems like a mess, but wholesale/retail framing is useful.  He takes the property point distinguishing enforcement from registration, but is it a get out of jail free card?  Still need to look at less restrictive means.   TM does a lot of different things—industrial policy, etc.—Breyer gloms onto source identification, but were they invited to think about other things? 

Litman: SG raised the issue of Congress trying not to give imprimatur/anti-discrimination norms.

Dinwoodie: the problem w/those rationales is that they run into instinctive speech responses: you can’t suppress just b/c you don’t like.  But there might be issues about other things like notice and treaty compliance. 

Grynberg: Tam oral argument, the focus was on source identification—he wrote about why disparaging terms are inherently less likely to identify source [though that gets us back to wholesale/retail].

Burrell: In Brunetti: why was the court so desparate to get its hands on the 1A issues? Most systems wouldn’t have called FUCT scandalousness.

RT: (1) to get affirmed; (2) because vagueness is the other half of the 1A vise and so many other fuck-type marks had been rejected.

Dinwoodie: how much room is left open for a prophylactic approach?  Gov’t program type conceptualization of what TM is about that would prevent ongoing litigation of any TM issue from a 1A lens.

RT: I talk about that in my Tam paper a bit—given the interest in having a registration and infringement system, the costs are worth bearing.  [It may be that examination on absolute grounds is constitutionally required here; we couldn’t just register w/o examination and grant the rights we do.  But see ITUs.]

Ramsey: focus on regulation of commercial speech and misleadingness.  Gov’t has interest in fighting fraud, and also in promoting fair competition (including use of nonmisleading terms).  Any registration laws that promote competition (functionality, genericness, exclusions of nondistinctive symbols) will be ok, and facilitating truthful source ID is also ok.  Tam says gov’t interest in deterring offensive speech is not sufficient.  Not just protection of property rights, but property rights in language. There’s a difference b/t cars and colors.

Dinwoodie: EU allows property rights even upon application. [Asks about my speculation that under Central Hudson substantive examination might be required; though treaties leave us a lot of flexibility, the narrowest implementation of our treaty obligations easily allows substantive examination.]

McG: why wouldn’t non-examination and challenge post issuance be good enough to satisfy the 1A?

RT: compare to obscenity: it’s the closest analogy I can think of, though still distant—a licensing regime where you absolutely can’t put the burden on the affected parties to make a challenge. This is a regulatory regime very different from most others we’ve seen challenged.

Grynberg: property/TM conflation has been making a mess for a long time.  What’s different now? Why would that endanger Rogers?  [Because of Michelob Oily.]  As long as we maintain a commercial/noncommercial distinction …

McK: the significance of commercial/noncommercial is in obvious flux. [And obvious contradiction among types of law: © commercial is not 1A commercial.]

Realistically, there isn’t going to be significant will to challenge the whole registration system.  And so where it will play out is enforcement. The Court ignored an obvious issue: enforcement is prohibition of speech directly; denying registration is a much subtler impediment.  Surely our current functionality doctrines are not the least restrictive means for dealing w/functionality.  How do we decide whether a particular provision is necessary to promote competition, at which level of generality?  It’s a silly Q.

When you look at enforcement, a lot of times the P is objecting not entirely because of confusion but b/c of use on goods/in ways the P doesn’t like. That’s expressive at least in part.

Dinwoodie: compare to European concept of advertising function, whatever that is, versus source identifying function.

Linford: Sorrell v. INS: speech the state can’t regulate—how far are we from having no distinction b/t commercial and noncommercial speech?  Similar to Citizens United.  Would that have an impact in © and TM or would they handwave on a property basis? Maybe harder to handwave in TM than in ©.  Harper & Row addresses 1A some and still gives Ford a right of first entrance to the market despite speech rights. Campbell = release valve.  If you look at statutory offramps for noncommercial use and commentary, and the fact that we’ve doctrinally imported them into standard TM analysis, a court could limit Tam & Brunetti in the same way Eldred/Golan did—we can have our cake & eat it too.  Would that survive if we really do see shaving of the commercial/noncommercial distinction to nothing?

Diamond: speak more about blurring.

RT: Brunetti says there’s no gov’t interest in promoting some marks over others.  There’s also a Q of what blurring really is/what the harm might be, especially to famous marks.  Also, City of Cincinnati v. Discovery Networks: given the exemptions, the remaining coverage is limited and that small contribution to the overall harm means that the regulation flunks Central Hudson.

Beebe: Rogers step 0: is it an artistic work?  That seems to pick up all the commercial/noncommercial stuff.

Ramsey: consider a nonprofit using a mark to fundraise—that’s more of a TM/commercial use. But offering religious services—e.g., a breakaway congregation—seems so noncommercial that it should be protected.  For some marks it depends so much on context, and so some challenges can be facial while others can be as applied.

Eric Goldman: Denying benefits based on someone’s speech always seemed like an easy case of unconstitutionality to him.  Does think that enforcement is a different question.  Denying © registration to Democrats would also be a problem.  [But that’s not what disparagement was: it denied registrations that disparaged Democrats, Republicans, or nonvoters—and allowed Democrats any registrations except those that disparaged, as others were also denied.  But I understand I’ve lost that debate; the question is what else is viewpoint-based.  Tam is not much help in determining what the next viewpoint-based regulation will be.]

Bill McGeveran: Note that cases raising these issues will be thin on the ground—most dilution claims aren’t litigated and most dilution cases aren’t litigated alone but rather have confusion claims.  TM owners would have to be willing to press their luck.  [LV, I’m looking at you.]  Continued invocation in C&D letters might look better. Cts have a long list of avoidance doctrines, e.g., finding dilution unlikely or a 43(c) defense succeeding.  Won’t have to make other moves w/in 1A doctrine that would be bad.  Harm = vestigial doctrine w/ zombie life in C&Ds.  Remember that parodists no longer lose.  Radiance v. NAACP district court opinion v. appellate opinion—the former looks like a case from the 80s. Judges in the last 20 years want to look like defenders of free speech. These cases about registration may continue to have real-world effects in the registration system, and it’s conceivable they could break out into infringement but thinks the likelihood is really low.

Dinwoodie: motion practice to isolate constitutional issue?

McGeveran: in almost any scenario the nonconstitutional issues would be available on the motion too.  Section 2 bars don’t take away from the core function of TM as judges see it.  To invalidate dilution, a judge would have to take away a piece of TM rather than expanding TM protection as w/invalidating a §2 bar. 

McK: but you’d be supporting the speech of someone else.

McG: whereas allowing registrations increases the scope of TM.  Prediction: it’s a big ask to tell a judge to get rid of a congressionally enacted protection.

Dinwoodie: you could get a dilution claim for an opposition—the Coach case.

McG: thinks that’s more likely.

Grynberg: consider impersonation—could we have a specific claim for impersonating an entity even in a nonprofit context—impersonation as defamation or other harm?

RT: Eugene Volokh has a great paper about, among other things, using defamation to deal w/impersonation cases. Consider also the treatment of false advertising cases alleging false patent claims: courts have imposed a higher intent requirement than ordinary Lanham Act cases; we could do something similar with political/religious cases like United We Stand or religious schism cases, requiring actual confusion (the problem remaining being that they won’t have deceptive intent insofar as they will believe they’re the real thing, and this is a real conflict that we will have to resolve).

Bently: EU takes the position that denying registration puts very little burden on speech, and if there’s some legitimate interest in preventing commercialization of the sign, that’s good enough.

RT: but aren’t those two things the same? If there’s no burden, then there’s no deterrence of commercialization—those are commensurate.

Bently: yes, but we’re ok with that—similar things in patent law. Some is about artificial incentives—we don’t want extra incentives to do undesirable things but we won’t prevent them if they’re truly committed.

Dinwoodie: European/British approach: we have no problem making gov’t judgments of what’s un/desirable where US would leave it to the market.

Burrell: Australia’s scandalous marks provision has been almost entirely hollowed out. Where he sees that as undesirable—case last year where Commonwealth Games Committee appropriated a term from the local indigenous language, while building a stadium in the middle of traditional land.  TM Office says it doesn’t matter b/c we’re banning some terms & reserving rights for indigenous rights/important people.  The Office is pushing the line: if we don’t like it, just ban the use, which absolves us of any responsibility. He’d prefer an intermediate category w/o ban but no registration either.

Friday, April 13, 2018

TM scholars' roundtable, part 4


Session 2, Cont’d

Mid-Point Discussants:                      
Mike Grynberg
Belmora may do very little.  But: Belmora may be bad if it discards important checks, one of which might be territoriality.  Another idea: arguably undermines systematicity of TM, its existence as an autonomous body of law that courts apply. Casually discarding territoriality is problematic in itself and also in understanding what TM is—Jeremy Waldron on the importance of being able to see how a change ripples through a system.  Belmora undermines the idea that consequences flow from not having a TM/not having a registration.  We don’t know what happens when 32 and 43(a) are not equivalent [and I would argue that there’s a hydraulic push—TM owners then claim that they shouldn’t be worse off b/c they don’t have a TM, so 32 has to be at least as broad as 43(a)].  By throwing more up for grabs in inchoate unfair competition, that undermines pro-D thinking in TM law that hasn’t been systematized, such as the 7th Circuit style/plausibility analysis.  Tabari: 9th Circuit talks about what people expect when they see TMs in domain names.  Handwaving about consumer sophistication—but will Belmora allow a second bite at the apple in terms of unfairness claims.  MTM: court says no one thinks there’s a connection w/Amazon b/c of Amazon search results, but would Belmora undermine that by allowing a rump unfair competition claim.

Finally, Belmora exacerbates the problem of TM scope.

Jeanne Fromer: trade secret might provide some insights about unfair competition and its path.  TC starts out as common law in the courts; turns into a multifactor test in Restatement, and then gets codified in a way that offers guidance about what us be found. Uniform Trade Secret Act is more regimented than 43(a), though “improper means” creates uncertainty.

Belmora: the more it’s not applied, the more it will not be applied. Possible to push back on how this is a very limited exception [thinking here of INS v. AP].  Backdoor protection of TM-excluded matter: do we want to understand UC as a matter of not wanting to go beyond TM’s internal limitations, or is it just that TM stops and some other regime starts?  We have ways of thinking about doctrinal overlaps in other areas, such as TM/patent, trade secret/patent.  One of her worries about Belmora: remedies. When TM wouldn’t give a remedy, shouldn’t give that remedy through UC. That might be one way to think about the interface.

McKenna: what’s unusual about Belmora v. old style cases: the latter were things that couldn’t be marks b/c of their nature. That was a recognition that they weren’t susceptible of ownership.  We basically said hands off those things as such, even though they might have caused some confusion. Belmora’s not like that: Flanax can be and is a mark, it’s just not owned by the entity asserting rights in the mark.  You can’t avoid the ownership Q in that case; it’s not about using the TM component v. the nonTM component. You have to confront whether UC will provide a remedy not b/c TM law is silent but b/c TM law has affirmatively said there’s no right.

Fears Lanham Act language that will be cited is that sentence that says that 43 doesn’t require a mark used in the US.  W/o discussion of the reason that someone doesn’t have TM rights.

Dinwoodie: is TM law silent about genericity?  He thinks that’s affirmative too.

McKenna: Generic is a tough one.  Takes the point.  How you understand what the decision is—but still Belmora is about some content unambiguously w/in the scope of TM.  Could be used to argue that, if you have a fanciful mark and no likely confusion, a remedy is still available.

Dogan: Antitrust as a road not taken: choosing between balancing pro and anticompetitive aspects of behavior v. using subjective motivation of the actor.  If a product “improvement” has anticompetitive effects w/o being a better product, there’s no social value. Implicitly, Belmora etc. are courts saying “you’re not necessarily violating rights but going further means you’re not engaging in behavior w/any social value.”  Tea Rose/Rectanus also looks at good faith/reason for adoption; protects the innocent.

McGeveran: reminder that court in Blinded Veterans talks about how the remedy could include requiring a different name as long as “blinded” and “veterans” are part of it.  Allows D to appeal fairly to the desire of the public to aid blinded veterans—the policy reasons why genericness exists.  But still talks about the goodwill P has built up, when generic terms maybe shouldn’t have goodwill b/c they’re not marks.  [McKenna says that’s not necessarily true b/c generic terms could have minority secondary meaning for a subset. Not a mark is not the same thing as lacks source-designating meaning.]

Bone: wants to say that it’s still all TM; where a term has secondary meaning for only a minority of consumers, it’s still a mark and it’s still owned but the remedies change. The cost of separation is that all the remedial flexibility goes to UC, and the breadth of remedy may be unwarranted for every TM instance.  If you said it was all one big thing, called TM for argument’s sake, then you have a series of competing policies to look for a remedy in a given case.  Maybe you have rules of thumb, but at least it’s being done in the same box.

McKenna: haven’t you just mooted the point of validity analysis?

Bone: no, b/c at some point it will make sense for policy reasons that the costs of tinkering w/remedy are greater than benefits in avoiding consumer confusion and you should just have a no-remedy.

McKenna: so you always have a valid mark?

Bone: until the very endpoint where no relief is available—where there is no entitlement to relief, there’s no mark.  Or we could say you still have a mark, but no remedy.  The former communicates freedom to use more clearly. [This really has no room for registration, does it?]

Ramsey: chilling effect of demand letters, but that’s all about notice.

Dinwoodie: you’re making normative choices about security/predictability.

Bone: I’m just not separating into 2 boxes.

Sheff: non-consumer welfare/efficiency based perspectives: Blinded Veterans presents a bit of a problem b/c we have a conflict b/t access to generic concepts and consumer confusion.  That’s a distribution problem, not an efficiency problem, and consumer welfare can’t answer the question of what we should do. We could look at it as a Kaldor-Hicks problem—could try to compensate losers for what they lose from the result we arrive at.  Or we could look at who is good and who is bad—relationships among producers, not just relationships b/t consumers and producers. TM and especially UC should be understood as being (also) about relationships b/t producers and what they owe each other. One reaction is to say that confused consumers will just have to learn; they aren’t thinking about it in the way they should be, and that’s a moral judgment. Or we could consider them vulnerable and in need of protection, and those of us who can tell the difference will just have to accommodate them by giving up the non source identifying meaning to protect those who do see it as source identification. Or we could try to accommodate both to the extent we can. On the producer side, there has to be independent normative content to figure out what producers owe each other. Belmora: easier case that the D is doing something wrong, which doesn’t have to be filtered through what Belmora is doing to consumers.  If we thought more about duties, we might have an easier time thinking about Belmora, though maybe it just collapses into naked misappropriation reasoning.

Burrell: Unconvinced that remedy is problem in Belmora.  Bodybuilding company wants to register Superman for bodybuilding products in Australia; DC opposition initially rejected b/c has no blanket rights in Superman, but when the product is released has a big stylized S on it.  Bad faith is clear and the mark is cancelled on appeal; barring all uses of Superman by this bad faith actor is not a problem, though he understands the problem with slippery slopes/courts losing sight of the nuance.

McKenna: if evidence of bad faith is just that they’re using a mark that’s owned by someone else, that’s still a problem.

McGeveran: presumably there is a robust UC claim, in false advertising or otherwise, if Belmora falsely pretended to be a headache drug that was unavailable in the US.  Having said that, why do we get in a twist if the nature of the bad faith deception has to do with misleading around source? 

RT: because bad faith isn’t an element of false advertising and shouldn’t be.

McKenna: the content of the false statement about the drug ingredient isn’t ownable by anyone else.  But if you admit that there is the possibility of ownership and that ownership has territorial boundaries, we’re artificially dividing up the world.

McGeveran: but what if Unilever, which produces Nuprin, sues instead? Surely we don’t think their false advertising claim is a problem.

Sheff: who is the victim and what is the evidence?  Either the consumer is the victim and the harm to the mfgr is the evidence, or vice versa.

Litman: but that’s not what the 4th Circuit said. It said that if you sue under 43(a) we don’t pay any attention to longstanding rules b/c this section doesn’t mention them. Erosion of territoriality and the idea of evading pesky doctrinal limitations by switching sections. 

Sheff: if the victim is people in the US, that’s a false advertising harm. If the victim is in Mexico but the evidence is in the US, maybe not. 

Dinwoodie: no reason to think that UC claims should be territorial—10bis indicates that the standard is an international one.

TM Scholars' roundtable, part 3


Session 2:  The Policy and Doctrine of Unfair Competition Claims

Introduction:  Jake Linford: Hoping for a sharper line b/t UC and TM, much as we have a relatively sharp statutory line b/t TM infringement and dilution to the extent we can define dilution.  Can you sensibly plead all 3 causes of action. If you think Kellogg/Blinded Veterans is the right kind of UC w/a limited remedy (ct said unnecessary in Kellogg—Ginsburg points out in BV that the dct should consider whether that’s necessary). A release valve for the hydraulic pressure where there is confusion, and remedies are then cabined. We don’t have that world.

Belmora seems ok from a consumer protection perspective, though it pushes hard on territoriality.  [The existence of a false advertising claim, with its stronger limits on what is actionable in the form of commercial advertising/promotion and materiality, is also important here.]  Checking judges’ work and Beebe on factors: not every multifactor test does balancing for us, and LOC test doesn’t.  On the Q of release valves: Belmora opens some room for arguments that haven’t succeeded and some that have, e.g. DC Comics v. Daily Planet—an unfair competition/free riding claim.  Limited remedy against a bike lock manufacturer or a TV program.  [But the remedy granted these days isn’t limited.]

Belmora’s subsequent use: restatement that 43 claims are supposed to look like 32 claims, but appears to help people get past the motion to dismiss stage.

Jessica Litman: Lanham Act drafters assumed unreflectively that unregistered TMs would continue to be actionable as they had been before the Act, rather than changing what they meant; that’s why they included nationwide priority and incontestability, as a bribe, b/c registration wasn’t required.  Unfair competition crept in late, in 44(h), b/c Congress believed we’d signed a treaty requiring it.  44(i) extended that protection to US citizens at the last minute b/c it was unfair to protect only foreigners. When Congress was passing the statute, it didn’t think of 43(a) as the source of protection for unregistered marks or unfair competition. Envisioned as early as 1920 as a distinct provision about intentionally deceptive marketing rather than as a home for unregistered mark infringement. Less about passing off and more about marking cotton as wool.

We are all vulnerable to the delusion that the law we first learned makes more sense than the law as it has later developed.  Litman was in law school b/t Inwood and Two Pesos.  In the 1980s, the legal system was the system McKenna suggested was sensible: 32 was registered marks w/presumptions of validity and ownership; had started to recognize common law TM infringement under 43(a), but then you had to prove ownership, distinctiveness, secondary meaning (even for inherently distinctive marks), geographic and temporal priority. Before false advertising got complicated (literal falsity/implicit falsity) it was just independently actionable deceptive marketing. The standing requirements were loose—if you could plausibly allege damage from the false marketing, lack of direct competition wasn’t a standing problem.  Two Pesos mucked it all up.  Then false advertising developed epicycles and complications.  But if you look at 43(a) through that lens—2 distinctly different things actionable—then Belmora is surely wrong as a matter of common law TM infringement but it is really understandable as false advertising, intentional deception of consumers.  More useful to understand Belmora as a species of false representation law than a massive expansion of common law TM infringement law destroying territoriality.  Which raises questions about the appropriate remedy: remedy based on the word Flanax would be a TM remedy.

McKenna: but if the remedy against the word comes from (and against) someone who owns TM rights in the word, then it’s a TM back door.

Litman: intentional deceptive marketing. [Though intent is currently not a requirement for Lanham Act false advertising liability.]

Discussant:                  Barton Beebe 
Belmora was shocking at first, then got used to it, then became shocking again primarily b/c of its rejection of simple territoriality.  A global Tea Rose/Rectanus doctrine in which rights extend as far as reputation does, even without use? Important footnote 8; impression is that it was added later.  Plaintiffs who rely only on foreign activity may have proof difficulties. A few isolated consumers who are confused isn’t enough; a case w/o other misleading conduct would face difficulty. It’s different where the cornerstone of the D’s business model is passing off.

If Belmora catches on, is there any reason to hold onto well-known marks/Grupo Gigante?  The latter has a heightened secondary meaning requirement, whereas Belmora apparently gives you a path to success if there’s more than a few isolated consumers.  A few follow-up cases: DC dct suggests Belmora, even if applicable, wouldn’t matter b/c no proximately caused harm; Md. case saying that there are two elements to a claim, proximate cause + injury—the case that motivates the worry that courts will lose track of the details, history becoming legend/myth.  All of the contextual cues could be loss. Another NJ case finding Belmora plausible, though obstacles of proof may exist.  DSC: domestic dispute, no foreign mark—treat false designation the same as infringement, but see Belmora: though not directly relevant here, UC claims don’t always require ownership of a valid mark b/c 43(a) goes beyond TM, citing Belmora/Blinded Veterans.  Already we’re starting to see the details wearing away. What will be left could be bad black-letter law. Optimistic version: Following the European example, appellate courts might treat the distinction w/more sensitivity.

Bone: Nothing much turns on calling Belmora a TM case.  ITC v. Punchgini: not a lost sales case, so it would have to be a reputation case, which then turns on how broadly reputation harms are conceived.  Bad literalism, invoking Lexmark: interpreting the Lanham Act w/o recognizing that it was enacted against a backdrop of common law principles that should be read into the statute, but aren’t.

Ramsey: not as troubled by Belmora b/c has taught well-known marks. It’s an exception to territoriality for the TM owner, but not for the consumers who are confused. 

Sheff: the US registrant in Belmora is using their US property in ways that harm someone else over the border.  Maybe a way to think about it. 

Two Pesos: held that inherent distinctiveness as a question for the jury (where as Litman points out the jury was instructed that a mark is inherently distinctive if it’s not descriptive) and that might not be right. The fact/law distinction might do some work here.

McKenna: Likes Litman’s description of Belmora until the end.  Need a body of law in which the claim is recognizable. Hard to tell in 4th Circuit’s description whether this is a TM case or a false advertising case. Extraordinarily likely that the details will be lost w/o vocabulary to articulate it as a distinct form.  Saying that remedy can cover the mark itself invites everything to be sucked into Belmora.  Reminds me of early version of the Rogers test where TM owners would say “it’s explicitly misleading because it causes confusion”—only in the last 10 years did courts explicitly reject that argument. [As they have meanwhile created that same problem with nominative fair use, turning factor three “do nothing else” into factor three “isn’t confusing.”]  If the evidence that it’s false advertising is the use of the mark itself, then plaintiffs will argue that nothing else mattered (intent, other deceptive acts, nature of the packaging).  If the court made clear that it’s hands off the word, that risk would diminish.

Litman: cancellation is available for marks used to mislead about source.

McKenna: Q is whether use of the mark alone, w/nothing else, ought to be able to satisfy that standard. 

RT: [Others’ approval of Belmora makes more sense for me as a story about source confusion but when I think about IIC and other extensions of confusion theory I get very nervous: another version of Mark’s concern.  What if it had been “if you like Flanax, you’ll love our Forex”?]

I want to talk about 43(a)’s non-exclusion of functional matter: 43(a)(3) says “In a civil action for trade dress infringement under this chapter for trade dress not registered on the principal register, the person who asserts trade dress protection has the burden of proving that the matter sought to be protected is not functional.”  But it doesn’t say what happens if the claimant fails to meet that burden.  Blinded Veterans/Kellogg remedy still available?  And yet everyone has generally assumed that the result is no protection at all, for pretty good reasons from a TM perspective.  First, should we provide limited remedies for functional matter under 43(a)?  Second, could we infer from the opposite conclusion, the availability of limited relief under 43(a) for functional matter, that relief under 43(a) should be presumptively limited?  If not, and if all of 43(a) is just unfair competition, what function does the language in 43(a)(3) serve?

Dinwoodie: language was inserted at a time when 43(a) was understood to be more limited [though I haven’t decided which way I think that cuts]. Notion of use has expanded internationally in the US.

Doesn’t think that using Lexmark will compress the factors in confusion when it’s actually on remand.

Litman: Thinks Tam rejects Two Pesos: registration is substantive, it makes a difference; unregistered marks are not the same as registered marks. Tam doesn’t fill out what it means by that, but if it’s true it’s not the case that unregistered and registered marks are the same, as Two Pesos said.

Sheff: could be substantive in some ways and not in others. 

Dinwoodie: Two Pesos on distinctiveness could be the same, while other aspects could be different.

Litman: but cf. Beebe’s argument about how details get washed away. She’s predicting.

McGeveran: FTC also uses “unfair” and that helps you think about it.  Would Belmora’s acts have been enforceable by FTC/AGs as deceptive/unfair trade practices harmful to consumers? Obviously yes: they were misrepresenting their heritage to bodegas/Latino customers. Unfairness has a three part test: practices that cause substantial consumer harm not outweighed by benefits to consumers and competition, and consumers can’t avoid the injury by themselves. A different approach to understanding marketplace behavior. Remedy will often be remedial, not penalty.  What bothers us about Belmora is why Bayer gets to be the enforcer.  Unjust leg up b/c Bayer sells Aleve—but that’s not the court’s primary theory [of false designation; it is the theory of false advertising]; it’s instead that Mexican customers won’t buy in Mexico and will for some reason wait to buy in the US, because that’s totally plausible as how people buy headache relief medicine.  We’re offering up unfair competition if TM theory doesn’t reach consumer harm, but he thinks that false advertising/non-TM sources of consumer protection law can take care of that w/o needing free floating UC.

Linford: Zechariah Chafee writes in 1940 about what unfair competition might be—looks at Keeble, also at FTC.  Sorts out ways of thinking about UC.  Holmesian/Wigmore tort about harm, harm causation, excuse/justification.  Most of the work occurs at the last part.  There’s a presumption in that presentation that P is entitled to something (if you think about TM as property).  Be careful: if I don’t have rights in the US, I don’t have property.  I haven’t secured rights in the space against metaphorical infringement.  Perhaps, Chafee says, we’re trying to figure out how to weight individual and social interests—perhaps what looks unfair is actually beneficial to consumers.  Belmora: not so much b/c there are plenty of sources for this drug.

Bone: Chafee’s restatement of Wigmore is that competition is presumptively harmful & tortious, but it’s allowed b/c it’s privileged.  First Restatement of Unfair Competition therefore speaks of a privilege to compete, not a right, b/c it requires a policy justification.

Ramsey: Analogous to Rogers, you can’t find UC based solely on the use of the TM-unprotectable matter.

Bone: Elegant distinction that doesn’t hold up. Why draw this distinction? Just describes a doctrinal pattern.

Ramsey: competition, free speech.

RT: my problem is the realist objection/nominative fair use problem—an action/inaction distinction is unsustainable when P’s lawyers are routinely so creative.  True, Belmora did a number of things, but it’s hard to imagine Bayer being cool with an orange box; it is plausible that putting Flanax on headache medicine is the core of the confusing conduct and that nothing more is required.

Dinwoodie: institutional Q re FTC: is the existence of another institution to deal w/ conduct evidence that we should leave them alone to do so, or is it evidence that the conduct is nasty enough that we should suppress the conduct at the behest of someone else?  Can argue it either way.

Grynberg: is it really a slam dunk?  Having a new brand in the US, where there is no US owner/the mark is up for grabs, is a consumer benefit.

TM scholars' roundtable, part 2


Session 1, Cont’d
Mid-Point Discussants:                      
Stacey Dogan: pre-20th c, very few cases of claims based on noncompeting goods, and even fewer with actual harm stories.  Can be characterized as chance or as the functions of the markets of the day, but either way courts were confronted w/very few claims.  Hard therefore to describe trade diversion as a principled commitment to a normative concept of TM law. When these cases began to appear, courts used the tools available to them creatively and expanded the doctrine. The blending of UC/TM infringement was creativity in response to what they perceived as the equities of the cases before them.

Doctrinal v. normative: trade diversion is not a normative concept in itself. The normative concept in McKenna’s Normative Foundation piece is Lockean natural rights.  The natural rights to which merchants have claim are rights to their repeated trade, but I’m not sure that if we view Locke/natural rights as the governing normative concept that it has a natural limit. If people are entitled to the fruits of their labor there’s no reason to limit that to expected repeated transactions with existing customers.  Threatens even more expansiveness: the only limiting principles are Locke’s enough and as good; no internal inherent limits.

Role of judicial intuition in definition/lack thereof.  Harm as a defining/limiting principle. That harm-based conception could be based on trade diversion, misinformation that causes consumers to be materially misinformed, or we could think about more attenuated reputational harm that theoretically can result. Judicial intuitions vary in idiosyncratic ways, but at least we could talk about what that harm might be. [Is it really idiosyncratic?  Seems that intuitions have changed a lot over the century.]  But “unfair” isn’t just harm-based. Some versions reflect misappropriation oriented approach. Similar trends in late 19th/20th c. in antitrust, which also lacks vigorous, robust contours for “unreasonable” or “unfair” restraints on trade.  Loose and manipulable.  [But also changed in systematic ways over that period as a result of different underlying theories.]  Opportunities for courts to cloak personal intuitions in what on the surface looks like some kind of formulaic doctrinal standard. 

Leads back to the question of whether we can define UC in a way that has inherent limits.  Confusion doesn’t have those limits, but neither does natural rights.  Two possibilities, which could both be pursued: (1) Shift or insist on a more harms based approach and try to define the harms that the law ought to be concerned w/.  Materiality can be a part. (2) Focus more on limits.  Focus on why some unauthorized uses are good.  Room for more attention to whether we can think of these justifications as providing more limitations on this unconstrained cause of action.  [I teach this as courts wanting to have their cake and eat it: they routinely assert both normative and descriptive limits, e.g., in justifying Rogers v. Grimaldi and nominative fair use.]

Jeremy Sheff: UC/TM discourse happens in the shadow of ideas about property.  Keeble v. Hickeringill—early reports of this 1707 case were that it was ratione soli—he owns the land so he owns the ducks.  A more reliable report, a century later, talks about the livelihood of catching ducks; it’s a lawful way of gaining a livelihood, and D unproductively/wrongly interfered with that livelihood. Diversion through competition is legit, but not what D did.  Innovation in Hickeringill was that in earlier cases, the act wasn’t independently wrongful (as hiring highwaymen to terrorize schoolchildren would have been).  What is it that makes firing on one’s own land wrongful? It’s a period of transition in thinking about rights in things.  Property theory: Harm inheres in disturbing baseline w/o consent of person who has endowment; endowments can be justified on the merits of the endowed.  But the later report in Hickeringill emerges from Hobbes: we start from a baseline of equality, and the rights we have are as expansive as they can be consistent w/equal rights in others.  Constraints in right to carry out a line of business are those that can be imposed on everyone equally.  That doesn’t get us as far as modern concept of confusion, but fairness is always parasitic on some other legal standard, whether that’s threats of violence or something else.  In Hickeringill, the independently wrongful part isn’t conduct but motive—acting on spite. 

“Promising” can provide independent normative content. The kinds of things that everyone carrying out business should be able to rely on is things that signal a promise. What’s wrongful: uses that make a promise that the promisor can’t perform w/o cooperation of others who are under no duty to perform, or interfering w/ others’ ability to credibly make or perform promises they could otherwise have fulfilled.  This is consistent w/a beginning premise of equality for all.

Bob Bone: Why don’t we have the same remedial flexibility with TM as with UC?  From normative/analytic perspective, this doesn’t make any sense.  Abandoned marks: using UC to deal w/remaining source ID meaning—we could do that, but instead we treat them like TMs and are willing to grant full relief.  Belmora just deepened his confusion, b/c put UC up against territoriality.

Does it make sense to have a category of broad protection/injunctive relief for TMs?  What do we do with the rest of them?

RT says: we could have double identity—property like.  You could justify it by saying enforcement costs etc. justify a conclusive presumption of consumer confusion.  Or you could say that what’s going on is unfair to sellers—an equity problem not really about consumers.  But if you do that, it makes sense to have those marks registered—you want to make sure that others have notice of this broad protection.

The problem with the rest of it is how much flexibility there should be. How much do we want to use rules v. standards? What scope for remedies? But practically the first category comes under hydraulic pressure to allow more of “the rest” into the more-protected category.  So he finds it hard to justify even having that first category where harm doesn’t have to be shown, though he kindly allows that my arguments are among the strongest on the other side.

Dinwoodie: Industrial property/metes and bounds claiming/notice functions—needs a right that’s relatively fixed and static. But one of the harms we’re interested in is an intangible that is very dynamic, consumer understanding, which means we want a second regime, tort-like. We then have to worry about the circumvention of the rule-based static regime by the flexible regime.

McKenna: Some early TM cases cite Keeble—productive use of your own property v. destructive use. Analogy that TM cases make: lying to divert isn’t productive use of your own property, it’s destructive of someone else’s property.  Equal rights theory works for this Lockean reasoning too. 

Two threads in conversation: realist—rules don’t matter b/c market pressure will change results—and rule-based: we need rules to cabin some expansions. We have to think that doctrine matters in some meaningful way to limit claims. Though nothing violates antitrust law any more, everything is UC over on our side.

Blinded Veterans is a perfectly sensible case if you understand that UC intervenes precisely when someone does not have a TM. The whole point of the remedies is that UC would only have something to say w/o a TM; the nature of the remedy has to account for the fact that you don’t have a TM; you can’t get any injunction on the use of the words Blinded Veterans, but only about the manner of use. Problem: b/c we’re so used to emptying UC’s contents into TM, every P wants the benefits of TM—no proof of harm requirements, no constraints on remedies—without the constraints of having to demonstrate that it has a TM.  Why do we have that dynamic?  UC needs to be limited and distinct or it’s a second bite at the apple for a failed TM claim.

Dinwoodie: do you want this to be about registered/unregistered?

McKenna: pretty hard—have to also decide what registrable should mean to answer that question.

Dinwoodie: European solution strongly encourages registration, but w/o limiting very much what is a registrable mark.

RT: Theory matters/does theory matter?  (1) Evolution of the law is not just random, even if it has been path dependent.  Consider treatment of comparative advertising in the US and Europe. Antitrust point: change in theory changes/narrows what’s actionable. (2) But in TM/UC can we do better than intuition at the edge cases?  Why do so many of us find the 7th Circuit cases rightly decided but wrongly/unreasoned?  (Is it fair to say that the 7th Circuit is actually always only deciding UC cases?) What if anything is the matter w/plausibility instead of nominative fair use etc., which have their own difficulties with definition. Is my objection merely aesthetic?

Dinwoodie: Derek Visser interviewed Dutch judges who do pretty much what the 7th Circuit does.  Not just aesthetic, but about structuring affairs predictably.  [But has it worked?]  If they’d reason a little more, then the doctrines could work—Beebe’s empirical work.

McKenna: 7th Cir. hides the normative ball: makes a determination that some things are plausible or not, but implicitly limiting the things about which they ask the questions. Court says there’s no risk of passing off but doesn’t talk about the other things TM law now covers (sponsorship/affiliation) and doesn’t say why it’s not talking about them.

Dinwoodie: a lot of these cases aren’t close.  [I think this is a matter of excellent legal writing on the judges’ part—I don’t see the 7th Circuit getting different kinds of cases than the 6th or 4th.]

Beebe: a tiny part of it is strong judges.  They all want to write like old English judges.

Litman: they also write their own opinions. If you won’t let clerks write, you’ll be [less doctrinal?]

Beebe: not consistent w/ the rule of law!

Dinwoodie: They’re so far away from 50/50 cases that he’s comfortable with that kind of thing.

Ramsey: we want certainty, which is why we’re attracted to registration, and to inherently distinctive marks that don’t need secondary meaning.  Double identity, but that creates needs for things like exceptions for comparative advertising. Vagueness is a big concern in free speech law.  Definitional balancing in 1A could be used for “TMs” and then messy case by case analysis for everything else.

Grynberg: Not just do rules matter, but what exactly is the source of the rule?  The reason Belmora looks so weird is that it discards what we’d thought is a systematic rule of TM, helping make TM coherent—you only have rights when you use a mark in a territory.  UC: consider the kinds of rules that we think are defining this discrete concept. 7th Cir. and Belmora: does it interfere w/courts’ ability to engage in hand-waving that is often aesthetically displeasing but helps them get to the right result?  Going to the jury as the same kind of thing.

Linford: does the idea of unproductive use open up a space for policing what’s unproductive, as in ©’s fair use analysis? 

Burrell: Do we want to confine TMs to a more limited category?  History in the UK, where we experimented w/ TM that had to be invented/fancy word. Series of really ugly cases on what that meant—don’t underestimate the costs of linedrawing. Geog. place names: Phoenix—is that geographic?  Stone?  Eureka?  If they’re in b/c they have other strong meanings, where do you go from there?  Political economy of proposed solutions: McKenna’s solution is not going to happen b/c offices aren’t going to lose ½ their revenue and we aren’t going back to unharmonized rights in Europe.  Maybe think about TRIPS?

RT: ISSCC in 2d Cir. as opposite of 7th Circuit: you have to consider all the factors and all the types of confusion that might exist—a useless/counterproductive formalism.  If those are the choices, maybe I need to reconsider my opposition.  Can we hope for McKenna’s ideal of discussion of the reasons for rejecting a theory in a given instance?

Mark Janis: ISSCC is like the 7th Cir, but in more words—at the end you’re told to consider a whole garbage can of elements and not told how to balance them.

RT: not sure that’s true: 2d Circuit is sending it back for more factfinding, while the 7th Circuit has seen everything it needs to give you an answer. They’re both unsatisfying for sure but for different reasons!

Bone: doesn’t like Sheff’s promissory theory.  The seller doesn’t think it’s an enforceable promise, and the buyer’s opinion is usually not enough to make a promise, unless it’s a normative conclusion about an implied promise.

Sheff: that’s a concern, but to the extent the promise is often/always implied, that’s something you could replace w/the word reliance, aligning w/materiality proposals by others.  Empirically, when are consumers relying on aspects of marketing in forming beliefs about what a product or service will do for them?

Dinwoodie: the problem with, e.g., geographic terms kicked out of TM, is less problematic if there’s no cliff (if there’s still UC protection).  TRIPS does allow some flexibility as long as what you’re doing is protection.  Registration on basis of secondary meaning: could put on presumptions/burdens that were different for registered inherently distinctive marks.  If you could show PHOENIX is being used nongeographically, you could get the higher level of protection [at least as against geographic uses]. 

Beebe: thinks that the multifactor test is good v. global plausibility reasoning, especially in the age of Trump.  It’s not great that judges write their own opinions.  7th Cir. could go out of control very easily in another set of circumstances; that’s why statutes & technical TMs are more attractive.

Dinwoodie: but one of the strongest points from your article is that they aren’t following black letter law!

Beebe: it at least exerts a general background discipline. The judge can’t get away with as much.  Belmora reduces everything to two elements where it used to be five (falling w/in the general terms of the statute + proximate cause), which itself just reduces to proximate cause.  Seems risky when we take out elements and reduce the issues to less analytical ways of proceeding.

Dogan: Rules & legal realism can complement each other.

McKenna: at least it allows us to criticize the decisions.

Dinwoodie: and yet Beebe has shown that they’re not telling the truth about using the multifactor test.

Litman: the rules are way more expensive. If you’re in the 7th Circuit you don’t need a survey unless the judge says s/he wants one.

Jeanne Fromer: why has UC not swallowed everything up?  Is it judicial discretion?  Lawyers/clients limiting the claims they bring?  Or is there some internal limit we haven’t seen?

McGeveran: it has, it’s just changed its name to TM law.  UC may mostly be vestigial.

Beebe: how do we teach 43(a) differently from 32?

Grynberg: didn’t spend much time on the difference until Belmora.

[RT: I teach 43(a) mostly as trade dress where you get to pick the contours of what you’re claiming based on what the defendant is doing.]

Shari Seidman Diamond: Cognitive consistency argument: interprets factors in line w/what the ultimate decision is. One would like the 7th Cir. to use the factors just so it can go back and forth in its reasoning (reflective equilibrium!).  Easterbrook was very frustrated b/c he couldn’t get them to print the pictures of the Top cans in color b/c then the obvious lack of confusion would be even more obvious to all.

Sheff: could collect evidence on subject matter, etc. for 32 v. 43(a) claims.  Belmora is a weird 43(a) case precisely because it is a word mark.

McKenna: need to distinguish b/t UC as a mode of reasoning or UC as a doctrine/body of law.  The former, which is what we are doing now, is very different from and broader than what UC as a body of law was.  We shouldn’t use the term UC to describe the mode of reasoning.  Belmora is problematic, but at least the paragraph at the end emphasizes that a remedy should take into account traditional TM principles relating to Belmora’s ownership of the mark.  They should have said from the beginning that this is an UC case not a TM case, and they can’t until they get to the remedy case b/c they don’t have the language to do so.

Dinwoodie: complicating factor is that the cancellation action goes forward in Belmora too and w/o a registration they may not have a mark.

McGeveran: so much similar language in Blinded Veterans.  Specific actions increasing likely confusion, but not b/c of a “mark” as such.  Intent, confusion, skullduggery; remand to figure out a remedy.  They sound like appellate judges trying to not grow the territory.

TM Scholars' roundtable, part 1


Trademark Scholars’ Roundtable, 10th year!
Session 1:  The Distinction Between Trademark and Unfair Competition Law

Introduction:   Mark McKenna: Until mid 20th c, could have said that TM and UC were distinct legal claims that vindicated the same right; thus cts said that TM was a species of UC.  All UC aimed at dishonest trade diversion. Use of another’s TM was an egregious example, but other things w/the same dishonest effect fit under the broader UC umbrella.  TM was a more limited category, not conceived of purely in source designating capacity: it was an ontological category of words/symbols that unambiguously indicated source, arbitrary/fanciful. Descriptive terms weren’t TMs; you could bring an UC claim against second users and the nature of the argument was the same, passing off, but not by copying a TM.  A subject matter distinction.

UC cases did involve uses that might have had legit explanations—e.g., descriptive use—so they had add’l proof requirements, such as secondary meaning and in most circumstances an intent to pass off.  UC was also usually characterized by more limited remedies.  Kellogg—SCt contemplated remedy directed not at name or shape but packaging, but here Kellogg had adequately distinguished its products.  Injunctions were highly tailored and would often include long verbal descriptions of who was making what and courts found them adequate (maybe just normatively adequate).

First federal laws were TM by definition—only technical TMs.  TM and UC then became jurisdictional—TM could be under federal statute, but UC common law.  Pre-Erie, would have been general law of UC, but post-Erie they had to admit that it was common law, so that was a big motivation for the Lanham Act.  Over the 20th c., some significant trends. First, we lost the idea that trade diversion was a central concern of UC, fundamentally altering the conceptual integrity of the whole system, b/c unified concept of nature of harm was lost.  Once TM covered non-diversion, UC became a slop bucket of ideas about things people were doing that we thought were bad. When Congress first started having hearings about child labor and convict made goods, some people suggested that a hook for Congress was unfair competition—producers in other states were at a cost disadvantage to those who were child labor users.  At first hearings (1916), that was dismissed out of hand b/c it’s not passing off, but those theories got more credence as the years went on and unfair competition is increasingly mentioned in the legislative responses (1930s).  Suspects, though doesn’t know, that you’d see same evolving thinking about narrowness of UC to a term that we apply to things we don’t like in hearings about FTC.

Problem: no new concept is put into place to substitute for trade diversion.

Second: we gave up on subject matter limits to TM proper.  Some of that is about registrable matter.  Unregistered marks become enforceable as marks under the Lanham Act.  Courts interpret Lanham Act more broadly than drafters expected.  Disuniformity concerns: courts invented the idea that the law reached unregistered and unregistrable matter; assimilated UC into TM.  Abercrombie mashes up categories of TM law with proof requirements for things that aren’t TMs.  Secondary meaning = requirement for TMs, when it began as a proof requirement for things that weren’t TMs.

Those two together leave us w/free floating doctrine, as in Blinded Veterans, that would be sensible under another system, but look strange in the modern scheme. We used to have a property-like system of TM as such, with UC as an equity system.  These days there’s a lot more pressure in identifying subject matter; the things we’re especially bad at IDing the boundaries of, like trade dress & complex marks, wouldn’t have been in TM at all in the past.

Belmora: once upon a time, it would have been entirely natural to think that even though Bayer lacks TM in Flanax, it had a right for a limited remedy for things like packaging and labeling. The reason Belmora strikes people as such a strange case is that it draws on concepts that no longer fit our system. B/c so much has been pushed into TM, the Q of rights you might have in non-TM matter is hard to ask.

Most of this is in motion in the 1920s-1940s.

Lionel Bently: The UK didn’t embrace the UC language/concept at all. Why not and what effects did not doing so have compared to the US?  Worth starting w/observation that language of UC becomes widespread from 1890s (or earlier).  US cts cite lots of case law from England in developing this distinction—buses using the same livery; 1843 case about boot polish, using label & name; Stone Ale for beer where Stone is a place in Staffordshire, but had secondary meaning and was protected.  So US doctrine is citing all these British cases to establish distinction b/t TM and UC, but British didn’t see it that way at all. British saw it as one big melange called passing off.  Engagement w/foreign countries, trying to develop bilateral relations was part of this, as well as potential multilateral treaties.  1883 Paris Convention: one British commentator thought that TM had to be defined before any progress could be made.  Ontological distinctions of Americans never really stuck in the UK.

Why not: Lots of reasons.  Has come up w/11, but won’t go through them all.  TM registration in 1875 allowed registration of only a narrow set, not even word marks.  1883: invented marks; 1905 became a general definition w/some exclusions, e.g. of geographic marks.  The effect was huge—significant uptake, and developed same disinction b/t technical TMs and UC, except done as registered TM infringement v. passing off, where there was “fraud.”  If you wanted non-fraud protection, you needed registration. 

Also, there was a House of Lords decision in 1891, Mobil v. McGregor, a combination b/t traders from which someone was excluded.  Lower ct had said that courts can’t distinguish b/t fair and unfair competition, and that was reflected in the House of Lords decision—a big signal from the top that UC wasn’t fit for purpose (at least for an imperial legal system).  Related circumspection about courts’ definition of broad legal concepts.  Fear of conflict with existing statutory regimes, e.g., copyright and patent.  Australian High Court’s rejection of INS v. AP—someone overlooking a racecourse can broadcast what occurs at the races: we have all these statutes and we shouldn’t undermine the legislature. 

Also, Germany & France had UC laws in the 1890s/1900s, and thus might well have contributed to a higher level of skepticism in the UK.  In the 1800s, Germany didn’t protect TM well, and there was a lot of counterfeiting; the US didn’t protect copyrights.  Towards the end of the century in both countries, there was a sense of a need to present the country as moral, and its businesses as moral.  Herbert Spencer was a big author for this purpose, especially in the US.  You see in American literature and case law a surprising level of pride in embracing the notion of UC.  Dennis v. Thomas, 1899 Delaware case—principles of high business morality gives “luster” to the system.  A sense that the law needs to control business behavior and also express a moral vision of US capitalism.  The British didn’t seem to need that; cases in Britain at the same time say (in case about competing seed companies that both used prizes and one copied language from the other) that the court doesn’t sit to judge “according to any high standard of honor” what should be allowed—“everything is fair in trade, as it is in war.” Totally different view.  [Talk about an imperialist legal system!]

Did it matter: Variation in US from 1920 through the Lanham Act, sometimes the UC label had effects on outcomes and sometimes less so.  In the UK, we can see more expansive tendencies from time to time even w/o UC.  We find deception in all sorts of places a right thinking human wouldn’t find deception, as a means of expansion. 

There are a lot of commentaries that say UC is just the American term for passing off—a mere terminological difference? But other people are building big conceptual frameworks to explain the relationship b/t UC in the sense of passing off and all these other laws like anti-monopoly laws more generally. Callman in the 1940s etc. start to elaborate ambitious systems for fitting these things together within a principled notion of regulation of competition.  Read first 120 pages of Callman’s 1945 book! It’s amazing, though obviously wrong.  He has an idea of competition as a relationship—very Germanic, duty of care based. You have to abide by the rules of the game—lots of sport analogies.  The umpire is the consumer, so you mustn’t mislead the consumer.  You must compete on your own feet, rather than misappropriating. 

He thinks that UC was sometimes just passing off and sometimes not, at different times/jurisdictions.  INS v. AP is the big point for misappropriation. Most INS commentary says it was a one-off and everyone just distinguished it.  But more contemporary commentary wasn’t sure—some very broad statements of the rule.  When we all leap quickly to Cheney Bros. and later cases, remember that at the time it seemed to signal some big shift.  That is a shift away from UC as passing off to UC as any form of misappropriation.  Erie, and even before Erie, INS was a casualty of the legal realists.

In the UK: noncompeting goods. 1898, Kodak bicycles infringed Kodak cameras, as passing off. Walter v. Ashton, the Times newspaper was infringed by Times for bicycles (again).  Showing the flexibility of passing off to expand well beyond what we’d think of as diversion of primary markets.  Having said that, 1947 we see a case on personality merchandising, McCulloch v. May, the High Court says that a radio presenter’s name isn’t infringed by use on breakfast cereal—you need a common field of activity for there to be passing off.

Discussant:                  Robert Burrell
Differing legal climates. Rise of legal realism in the US. Affects culture in which Schecter was writing, as Beebe has written: problematized the idea that you could throw around the label of property. It’s a conclusion, not a reason.  Never had the same influence in Commonwealth jurisdictions.  Earlier years of 20th c—passing off as protecting property in goodwill was taken very seriously by commentators.  That’s one reason why UK didn’t embrace UC—they thought they were protecting property.  However, bear in mind that these things were often close calls.  We now think of an Australian High Court case [Victoria Park, 1930s?] as the final rejection of UC in the Commonwealth, but it was a 3-2 decision.

We tend to write out of legal history: there might be a significant element of chance.  1891 McGregor case says there’s no law of UC in the Commonwealth, but a load of unregistered TM cases where you want to find the D liable.  You have to reach for some other theory.  If the House of Lords had dealt with a bad guy in 1890 maybe we would have embraced UC.  By the time the Australian High Court decides against UC, INS has fallen out of favor even in the US.

UC as nation-building: INS had an idea about nation-building.  Commonwealth had a similar dynamic, there about ensuring that other bits of the Commonwealth remained in the fold.  Imperial control, and leverage in international negotiations to create a law of the Empire.  1890s: clear imperial policy to ensure TM law uniformity, but that was when relations b/t imperial gov’t and self-governing dominions was becoming increasingly fraught. W/TM, though, happy to follow UK lead, so there was no need for Privy Council to try to harmonize through development of the common law—the self-governing dominions were playing nice and adopting UK law; no point in inflaming things w/massive expansion of law.  There was also possibly a concern about being overwhelmed: if there’s a broad law of UC, will courts throughout the empire keep coming back to the Privy Council for further instruction about what that means—a docket control concern.

Parliamentary sovereignty: no limit on what it could do. Not unimportant that in Australia, which did have those considerations, Victoria Park was in the 1930s/after the 1920s, a key moment of Australian law in which there was a massive expansion in what the federal parliament could actually do. There would have been no impediment to federal regulation of things like UC.

Bob Bone: Why create a category of technical TM infringement? Why did trade diversion drop out?  His answer to #1 involves a particular conception of property rights; sometimes concerns about difficulty with granting rights w/nontechnical TMs. 

Second, trade diversion was never a normative component, it was a standing requirement so that the firms had a reason to enforce consumer harms.  If you drop trade diversion as a normative requirement, you get reasons to protect technical and nontechnical TMs (w/more proof for the latter), but that distinction isn’t stable as soon as you start to Q the assumptions of technical TM infringement. And that begins to happen very early in the 20th c., even in the late 19th.  People recognize that once you drop out the property theory, TM and UC look very similar.  A lot of folks saying that UC is about fair dealing in the competitive market, as well as a leftover idea of UC as passing off.  That latter is largely doctrinal.  People are struggling to see how much further it can be pushed, and they recognize that UC is exactly what it says: fairness in the marketplace [but not competition?], and those principles have normative extensions, such as INS v. AP.  It’s a matter of overcoming 19th c conceptualism, which becomes legal realism; you see the boundary b/t TM and UC is porous/disappearing, and then you have the normative justification (fairness in market) for a further extension. 

Graeme Dinwoodie: Q of judicial role: extent to which courts in UK felt constrained by the statutory monopoly.  In the 1970s, in passing off cases, court says it can take inspiration from what the legislature is doing and protect under passing off.  In the US context, there’s a more developed sense of the Lanham Act as encompassing the broader subject matter—something like a delegation statute allowing the judiciary to develop the scope of rights.  Is the line b/t infringing a registered mark and UC the same as the line b/t registered & unregistered marks?  Related goods as a line?  Is any case against an unrelated good part of the slop bucket, or can they be trade diversion, given the development of businesses over the 20th c.?

McKenna: The problem was abandoning trade diversion and putting nothing solid in its place. Searching for a way to understand a theory of protection against noncompeting goods that is not limitless.  Closely related goods: he thinks that’s ok, not just vague sponsorship/affiliation. 

Dinwoodie: once late 20th c. becomes all about endorsement, that’s hard.

Beebe: confusion isn’t a limit?

McKenna: confusion about what?  You need a theory about which forms of confusion and among whom are relevant.

Bently: thinks the registration system was much more ambitious initially—designed to replace passing off.

Burrell: agrees, 1875-1900, people were speculating about whether registration was an absolute precondition.

Stacey Dogan: but it’s also important if there were no cases saying that.  We think the 20th c showed a big change, but many of the old cases rejected broad claims b/c there was an interest on the other side—the D wanted to use a geog term or their surname. There aren’t a lot of cases from the 1800s where the court says there’s a lot of confusion and no justification for the D but nonetheless we won’t intervene. Part of it is changes in commercial context through the 20th c.

Jake Linford:  Path dependence/hydraulic pressure.  Could be caused by good D’s att’ys, but cts don’t like leaving harms that they see unaddressed.  [Depends on whose, right?  Poor Joshua!]  Somewhat persuaded by affiliation/sponsorship confusion as a response to hydraulic pressure: competition could mean more than it used to mean, and that’s why such confusion is actionable. There are reasons to be troubled by that, but it is a plausible story for how expansion happened.  If trade diversion is just makeweight/standing, why can’t consumers sue for their own confusion?  Path dependence keeping consumers out of court?  Why shouldn’t consumers have more power in this space? But trying to vindicate that, w/o carefully defining confusion, leaves a lot of room for rights overexpansion.

Lisa Ramsey: Is it even possible to have a TM theory that isn’t limitless/supplies the necessary rules?  One option is a theory that guides most issues, but uses other considerations to limit it, such as free speech/competition.

Jessica Litman: in the original and 1988 legislative history, there is significant evidence that Congress (some members at least) thought that consumers could go to court, but courts slammed it down—all but unanimously didn’t want it in the courtroom. 

Ramsey: California state law.

Bill McGeveran: Almost all UDAP laws allow consumer standing.

McKenna: in a bunch of old cases, cts say TM/UC is about harm to competitors, while consumers have their own claims in fraud & deceit.  There are even a pair of parallel English cases with consumer/competitor Ps.

Bently: there are different accounts of what’s supposedly the first TM case; some accounts say it’s brought by the customer and others by the competing trader.

Dinwoodie: it might be possible to hollow out the concept of competing goods in the same way so it also doesn’t do work.

McKenna: competition is in UC for a reason.

Ramsey: “deceptive” trade diversion does more work.

McKenna: that’s another thing that happened: “deceptive” is a lot more limiting than “confusing.”

Bone: we know the theory; the problem is that it becomes really weak, if supported at all, as we push it, and the empirical basis of the theory may not be very strong. But the theory of reputational harms from association is perfectly cromulent.

Dinwoodie: once you accept that the claimant has the possibility of expansion, it’s not just lost sales but lost potential sales that matter.

Bone: all that fits the core theory; it just goes too far.

McGeveran: the content of what courts substitute in for direct competition/other traditional limits is much more inchoate than what went before: makes it easier for the next bout of hydraulic pressure to lead to expansion.

Beebe: material conditions change; the rise of brands.  Current analogy: globalization and Googlization—tech and social conditions change, and perhaps that helps explain Belmora.  Political conditions also matter: the Depression, the New Deal, concepts of antitrust/trust—how this affects concepts of TM law.  They trickle down into the law, just as current conditions do.  We’ve analyzed TM through a democratic system, which affects why we ask why consumers don’t have standing.  What would a communist/authoritarian TM system look like, where the individual serves the state and we don’t care about consumers, it’s just secondary.  How would you articulate that version of TM law or unfair competition outside a liberal democratic system?  [China?] [There’s also a really interesting book covering Nazi copyright law with a similar set of Qs.]