Copyright Society: transformative use (herein mainly of Google)

Transformation of Transformative Use          

Moderator: Julie Ahrens, Stanford Fair Use Project

Mass digitization: at the start of the litigation Google seemed bold, but after 8 years the result was less surprising, at least to some, since digitization is so valuable/expected. Perceptions about what ought to be easily accessible have evolved along with the law.  Subtle but critical shift in fair use evaluations.  It all started with Campbell.

Beyond parody: Dorling Kindersley is a very important case. Not required to discuss the artistic merits of the posters to use them on the timeline; the different purpose—use as historical artifacts—was sufficient when the initial use was artistic expression/concert promotion.  Prince v. Cariou: artist need not have anything to say about the original art if the purpose is different. Message need not be tied specifically to the original work/analyze the original work.  With remand on 5 works, Prince left some uncertainties about how much change is enough.

Panelists:

Joseph Gratz, Durie Tangri LLP

The Campbell mold has never been unique—something much older, can’t be stuffed into one word. The use in question is productive: results in the creation of something new that goes beyond purposes/value of original, and doesn’t supersede the original. Transformative is just shorthand for non-superseding productive use. Goes back to Folsom v. Marsh.  It’s been a good run since Campbell, but maybe the word transformative should be retired and talk about what we really mean, since different uses might not fit in the same bucket even though they are all nonsuperseding and productive—Pam Samuelson’s categorization attempts to group them. 

Four different buckets he’ll discuss: (1) Easily articulable comment on original, incorporated into new creative work.  Campbell, Wind Done Gone. Not all are “parodies,” but there is comment. That’s easy and everyone calls it transformative. Harder cases when incorporated into new creative work where commentary is harder to articulate—Cariou, Blanch v. Koons—commentary on society. We could say the original is used as a touchstone/icon.  Though Cariou photos were probably not iconic.  (Blanch is not iconicity; it’s representativeness/standardness, which also deserves to be available; you’ve got to be able to pick an example.)  Cat in the Hat came out wrong under this standard.

(2) Verbatim copying for the purpose of criticism—not a new song/painting/book, but just to criticize.  Rarely been of the expressive value of the original, but instead of the original or even of the author.  Maxtone-Graham: abortion-related book.  Criticism of Scientology.  Criticism of the message of the song “Imagine,” not the melody/chord structure/etc.  That’s transforming the work as a tool/object of criticism, but transformativeness is distracting: use for purpose of criticism, where license is unlikely. 

(3) Noncritical copying to prove a point: using the work as evidence.  Dorling Kindersley. The work isn’t being commented on in itself, but the work is necessary to make a point about the world.  Zapruder film case fits into this category: raw material in order to make an argument that couldn’t be made without that expression.  Different use than original.  (Why isn’t Blanch v. Koons here?)  These are three distinct categories.

(4) Technological uses.  Google Books, iParadigms (plagiarism detection). They aren’t the creation of new creative works, and they don’t involve criticism/commentary.  What do they have in common? Productive: result in creation of new thing that didn’t exist and couldn’t exist without the original, and they aren’t substitutes. You don’t and can’t go to these new tools instead of experiencing the original in full form. Does the value of that plagiarism tool stem from the expressive value of the inputs? No, its value doesn’t flow from or replace the creative value of the original. That’s how Google Books, Perfect 10, HathiTrust fit into fair use/favored under the first factor.  It’s a big umbrella, but transformativeness may obscure more than it makes clear.

Liz McNamara, Davis Wright Tremaine LLP

Not her position that a significant number of cases have gone wrong on fair use (speaking for herself). Courts have generally done the case by case analysis properly. What concerns her is the rhetoric and the way transformativeness is discussed, unhinged from Campbell—that happened in Google Books.

Where the law was before Google: pendulum swing!  Too often, content owners are accused of living in a time warp.  But content owners are the foremost users of fair use.  (Equivocation in the definition here, but ok.)  In the 1990s, the swing was in favor of too much protection—Harper & Row v. Nation led into Salinger/New Era that made fair use of unpublished works almost impossible, which was what led to Leval’s article and subsequent amendment of © Act.  Overemphasis on whether secondary use was commercial—Sony’s statement that every commercial use is presumptively unfair.  The 6^th Circuit in Campbell relied on this, which was the cause of such concern. Content owners breathed a sigh of relief after Campbell.

Rhetoric has become more shrill/out of the box on what’s transformative, with little regard for content owners—“free on the internet is free to use.”  (I’m not sure which advocates say that.)  Kelly v. Arriba Soft—not necessarily wrong.  But the facts are important: thumbnails were small/poor resolution and could not serve as substitute for aesthetics of full image. Court also assumed, w/o evidence, that search engine actually acted as pointer/drove users to websites where you could get the full image to the benefit of the © owner. Substitution remains a critical component of transformative use, not just in the 4^th factor.  9^th Circuit in Kelly took pains to contrast facts before it, where it found substitution impossible, with Infinity Broadcasting, where it was possible to use original broadcast for entertainment purpose.

Judge Chin/Google Books: what’s at issue is not what’s being served up in the search results, but the fact that Google digitized 20 million books without permission. They didn’t say please or thank you; they just took it. A volume discount on fair use? Judge Chin failed to take into account monetization of data Google collects/Google’s data mining.  (I thought she was cool with commercial fair uses?)  Improperly considered the project educational, looking at ultimate user rather than Google, when law dictates that Google’s use controls rather than ultimate user’s under Texaco.  When uses are systematic and no new expression emerges, we should apply a new and heightened standard, considering not just that the new use is useful, as Infinity said—change in format/enhancing convenience is not transformative.  Instead should ask whether it is capable of substituting for the original.  Kelly wrongly assumed that search engine acted as a pointer; AP v. Meltwater: we now understand how often clicks do/don’t happen.  News service was really being read as substitute. Including links wasn’t a pointer.

Also, are profits being diverted from original publishers?  Rosetta Books established that electronic rights are separate for authors.  Authors are entitled to license them.  Totally removed from the analysis.  Serious loss of revenue—Google strives to keep users on its platform, and 20 million books is an important monetary incentive to stay.  (… If you got to read the whole book, sure.)

Also need to look at whether this is one-off taking or systematic taking of massive quantities of data. 

She isn’t that optimistic about the 2^nd Circuit decision in favor of Authors Guild, but does hope that the court carefully circumscribes its rulings so as not to create unanticipated/significant negative consequences for creators.

Kozinski said that if you want to win in the 9^th Circuit you have to say that to go against you would be the demise of Google. That’s where we are.

Gratz: He’s since said that he’s not worried about Google.

McNamara: reminder about Infinity: media dial-up subscription services, allowing subscribers to listen to radio broadcasts around the US. Marketed for the purpose of auditioning talent, verifying that ads were played, etc.  Evidence showed that users could listen to snippets or entire broadcast.  2d Circuit found infringement.  D had argued that its broadcasts were used informationally, not for entertainment—different purpose.  Court rejected that argument: a difference in purpose is not quite the same thing as transformation. Retransmissions left the character of the broadcast unchanged, without new meaning or message.  Listener could use for entertainment purposes.  (Interesting switch to the effect on the user, which McNamara said wasn’t legit when Google does it.  Also doubtful that many users piece together a few hundred words at a time for original entertainment purposes of the books.)  Societal benefit wasn’t enough to be fair use, citing Texaco.  Not enough of a benefit to outweigh the taking.  Google Books will have to deal with this language.

Ahrens: one thing that seems missing from Infinity that is in more recent cases seems to be the idea of nonsubstitutionary use.

Gratz: we don’t disagree to the extent someone is using the service just to listen to songs, because the songs make them happy—that’s a substitution.  One thing that’s changed in the 2^nd Circuit is how one looks at who the primary/direct infringer is, and who we’re running the fair use analysis on.  We didn’t have the volitional conduct rule in 1998.  Who’s the direct infringer in Infinity? The facts in Infinity could have turned out differently if D was just operating a system that didn’t do anything until someone logged in and tuned their own little radio to play to them.  Is that fair use?  Maybe/maybe not, but then D would be secondarily liable.  If there are legit uses for ad verification, etc., that would be a fair use.  Have to drill down into what the person choosing to activate a technological system is doing.  That’s not actually a matter of changed fair use doctrine.

McNamara: D wasn’t paying anyone who created the content from which it reaped a benefit. Also, Texaco refutes the notion that the end user matters rather than the commercial actor appropriating the content.  News reporting mission doesn’t let news organizations line their shelves with photocopies of books on journalism.  Equally extravagant to claim that Google Books can line its shelves with 20 million copies of books.  (But in this analogy, the “journalists” can’t take the whole book off of Google’s “shelf,” even if their activities bring indirect profit to Google.  Now we’re back to the snippets.)

Ahrens: but what kind of substitution are we talking about? The original or a derivative market that the copyright owner would just really really like to license?

McNamara: substitution comes in many ways. There can be outright substitution, with news aggregation.  Sliding scale.  There can be substitution if copying usurps a separate existing market that people pay for. 

Gratz: Those are fourth-factor considerations, and don’t go into Q of where the value comes from—expression in the original or what the secondary user added. 

Texaco: not employed as part of a larger whole for some new purpose. Verbatim copying is always the case—it’s necessary to the process of creating transformative works.  Scientist wanted to make hardier copy/copy that he could write and spill stuff on; court said that might well be transformative if that was all you were doing.  But it wasn’t shown that he was doing that.  He was reading it to understand its intellectual content.  Assess the means by which the value of the second use is generated—if the means is the same, copying expression to enjoy expression, then that’s not transformative.

Would this be the same today?  Yes, even after the Second Circuit (hopefully) affirms in HathiTrust and Authors Guild cases.  Just making a reading copy is not by itself transformative, though that same act may also be necessary for a different and transformative purpose/act later.  Public benefit isn’t enough—and that doesn’t even come into transformativeness, though it’s worth weighing in fair use overall.  Totally useless changes can be transformative.   Letting us read 20 million books in their entirety would have a great public benefit, but that wouldn’t make it fair.  Creating something that doesn’t exist before, and that can’t substitute for the original, is different.

McNamara: seat licenses for electronic newsletters—the market has evolved in a way that addresses the Texaco issue.  (And strangles our libraries, but never mind.)

Q: how do you square Cariou with the definition of a derivative work?

Gratz: “transforms, recasts, adapts,” and he doesn’t disagree that the best word for what was going on may have been recasting rather than transforming in the sense of a parody.  Where it departs from what would talk about as a derivative work, is the way the word transformative is used in fair use: fodder or raw material in a way that separates it not for its expression but as a touchstone/icon.  Transformation is not physical (or not only physical) but in purpose, and that means something different.  But that’s one reason we should stop using the term in fair use. 

Ahrens: some markets are not reserved to the copyright owner, like criticism or commentary.  Maintaining that distinction helps preserve free speech.

Q: but it does interfere with derivative rights.  Looking at third party audience’s reaction isn’t helpful to the artist figuring out whether they need a license.  Yes Rasta photos weren’t iconic.  Rather than take his own pictures, he took what was convenient.  There are easily licensable materials.

Ahrens: necessary isn’t the standard; reasonability in light of the purpose is the standard. Nothing is necessary.  2 Live Crew didn’t have to comment on Pretty Woman. 

Q: but no one would ever permit parody.  But a visual reference isn’t parody, and is available for licensing.  (I wrote an article with Bruce Keller about why this is wrong.)

Ahrens: but he wasn’t picking from pictures of Rastafarians.  He was inspired by that specific book. 

Q: McNamara sounds like grafting unfair competition law onto fair use.  Why is making money relevant without a theory of unfair competition (which would be preempted)?

McNamara: there is an existing market for digitizing works.  Copyright Alliance amicus brief: needs to be decided with stakeholders at the table, not unilaterally by Google which drives everyone else out of the market—no one else can engage in this level of mass digitization; even Microsoft backed off. There is a derivative right at issue.  (RT: No, there’s a reproduction right.  This argument about Google’s uniqueness seems a bit in tension with the immediately previous claim that there’s an existing market.)

Gratz: those are fourth factor issues, and asking permission doesn’t matter to fair use.

Q: DJ culture as new business model. Where does that fit in transformativeness?

Ahrens: mashups can be new/not substitutive.  W/music in particular, and sampling law, it’s complicated.  The licensing model matters.  If sampling cases had begun to be decided now, and not in 1992, that might’ve changed the entire landscape.  But it’s not a new business model, but how the genre developed.  There are arguments about how this has harmed the art form; the only people who sample are showing off how much money they have and new entrants are limited in their expression.

Copyright Society meeting: Maria Pallante

Copyright Society of the USA Annual Meeting

[self-promotion: I’m here to pick up an award for Performance Anxiety.]

The View from the Copyright Office

Maria A. Pallante, Register of Copyrights, U.S. Copyright Office

Petrella and Aereo through the perspective of the government’s briefs.

Petrella waited 18 years to sue; MGM spent a bunch of money investing in a new version of the film.  Concern is for equity to defendant.  9^th Circuit was too generous to MGM, but should the court consider the delay in some way?  Copyright Office was concerned about threat to three-year rolling statute of limitations.  Middle position between what CO saw as parties’ extreme views, arguing that 3-year period should not be subject to laches, but delay could be considered w/r/t equitable remedies such as injunction, where user has invested in derivative works.  Gov’t prevailed.

9^th Circuit wrongly held that laches should be a complete bar where any part of the conduct occurred outside the statutory period—could deprive owners of their copyrights decades before the statutory term has run.  (RT: Of course, one can be deprived of other property that has no expiration date; other causes of action run out.) 

Dicta: is registration a requirement, or just application?  10^th and 11^th circuits use a registration approach: registration or refusal is required.  5, 7, 9^th use application approach.  Petrella passage suggests that Ginsburg & the 5 Justices who joined her agree with registration approach: registration reduces the need for extrinsic evidence—“registration must be on file with the Copyright Office before the copyright owner can sue.”

Aereo: performance can be most valuable right; 2d Circuit could undermine streaming models. But it’s important to preserve some cloud models, and the SCt focused on drawing a line.  CO: transmit clause makes the relevant audience the ones capable of receiving the performance, and Aereo’s promise to transmit to any paying member of the public makes each transmission one to the public even though each one goes ultimately to a single subscriber.  Could enjoin this infringement without questioning innovative tech for allowing consumers to store lawfully acquired copies.

Observations from oral arguments: Sotomayor asked about other services; Ginsburg asked about whether Aereo’s view was compatible with int’l obligations.  Another query of note: how do you deal with Cablevision? (laughter)

Many current reform initiatives, including PTO/NTIA green paper.  Multistakeholder dialogue on improving DMCA notice and takedown; also soliciting comment on specific legal issues such as first sale, statutory damages, remix.  Signature theme of IPEC (coordinator) was voluntary cooperation—between ISPs, music labels, movie studios. Led to copyright alert system, which just released its first report.  Copyright ecosystem: everyone benefits, so everyone including ISPs, payment processors, advertisers, search engines should contribute to its robustness. Civil and criminal liability is important, but voluntary practices can do a lot. Question: what’s gov’ts role in monitoring/setting practices?

Congressional hearings are underway, on House Judiciary website.  13 hearings by the end of the month.  They’ve had lots of copyright owner groups testify, MPAA to Getty Images to media photographers. Tech platforms; Library of Congress preservation experts; law professors.  Next Congress: more indication of what/whether we’ll have an omnibus bill or connected series of small bills.  Also unclear: what decisions Congress might send to the CO.  Courts are important but don’t weigh the public good; voluntary practices are good but a private sector deal won’t necessarily benefit public policy. SCt confirmed Congress’s role in Eldred. If the CO were to have more authority to update certain things, that might help.  Congressional Q: How to avoid picking favorites when it comes to tech and not allow piracy?  That’s the million-dollar question.

Does the existing Copyright Act adequately protect right of communication to the public under WIPO treaties?  CO believes there’s uncertainty in implementation, which has clouded the certainty required. Legislative clarification of making available would be beneficial; is it possible?

US implemented treaties in 1998 w/DMCA.  Congress and CO concluded that authority was already governed by combination of reproduction, distribution, and public display/performance.  Congress’s intent was to comply with the treaties, and the US takes the position that we comply. But some courts have not understood/been inconsistent, especially in questions over whether infringement requires actual distribution.  Nimmer: both sides recognize distribution right. Point of disagreement: quantum of proof required to demonstrate that distribution took place—uploading, or uploading plus proven downloads. (RT: Um, some of us think that a download is an unauthorized reproduction, not a distribution since no copy is transferred; secondary liability for the unauthorized reproduction takes the place of distribution liability.)  Prof. Lunney disagreed.  Notice of Inquiry: what would the contours be?  If Aereo wins, what is to be done?

Orphan works: years of study.  Picked it up again because of cases applying fair use. Had another public hearing on this.  What should be a good faith diligent search? What should the role of private and public registries be? Should photographs be treated differently, because they’re difficult to trace ownership and a category of works of a very vulnerable population of owners because so frequently targets of unauthorized use?  Mass digitization has been conflated with other issues.  CO is leaning towards view that mass digitization should be uncoupled from orphan works; need legislative solution because even if mass digitization for search/print disabilities is ok, that doesn’t solve other needs if access to works is what the public wants, so Congress should put a framework in place with remuneration for authors.

Small claims: 2-year study. Many small creators want to pursue claims but it’s expensive; we should not pay attention to those who don’t care about asserting their rights because they don’t have a problem, but these small creators do.  Songwriters’ Guild: combined impact of small claims is death by a thousand cuts.  Pro bono assistance for lower-income artists: people who do it wanted alternatives to federal litigation.  We recommended an alternative voluntary system (has to be voluntary, constitutionally) under supervision of the Register but general direction of Librarian, administered through online and teleconferencing, without personal appearances.  Three adjudicators, two with experience in copyright law and one in alternative dispute resolution.  Focus: no more than $30,000 in damages, actual or statutory, but statutory are $15,000 per work or $7500 per unregistered work; registration required before action.  Executive summary bears reading.  We expect a hearing.  Photographers want this; very few can afford suit without this. Dramatist Guild and National Writers Union also agree.

Music:  if there’s a primary focus in policy now, it’s in music. In the middle of 8 hearings/CO roundtables.  Licensing: Commerce Task Force roundtable too.   Lyle Lovett: struggle remains to convince music users and policymakers that music makers should get paid.  Analog/digital divides, etc.

Asking many questions about revenue divisions, opportunities for gov’t to encourage microlicensing and/or standards for ID’ing musical works and sound recordings. Current licensing is broken and should be fixed if not done away with completely: that’s the consensus. Question is what should replace it?  Musical works reproduction/performance licensed together?  Maybe with sound recordings?  Musical work folks want to move to “free market”; others don’t.  Transparency/competition concerns in collective licensing. Some frustrated music publishers have been offering rights to digital services directly. Challenge: how to reconcile competition issues with beneficial aspects of collective management, which is still very important.  DoJ is going to conduct its own parallel review of the operation/effectiveness of BMI and ASCAP consent decrees—whether and how they should be modified.  Comments due Aug. 6!

New bills in Congress: Songwriter Equity Act, changing CRB rates to willing buyer/willing seller and eliminated §114(i) which prohibits consideration of sound recording royalties in setting musical work royalties. Another: require payment of royalties for pre-1972 recordings for any service availing itself of §112 licenses on the same terms.  Other pre-1972 issues would remain before Congress.

CO: we’ve been stuck in time, in budget and otherwise.  Given the changes, we need modernization.  Nimble and forward-thinking; interoperable with marketplace we serve.  We’ve done everything we can do without money: talking/research. The registration system needs a faster/better interface; secure repositories for digital works; identifiers that match how works are created; automated recordation for assignments, licenses etc. that would be publicly indexed; metadata to connect to outside registries; need to finish digitization of pre-1978 works and integrate them into a useful chain of title for public record.

We can’t be fee-funded unless the statute were changed to allow us to recover capital costs for long term projects like IT.  We could if the statute were changed also ask larger © owners to pay a surcharge to make it remain affordable for smaller owners. Under our statute, we can’t raise fees beyond literal cost, but we can keep some deflated. Most have gone from $35 to $55 but we’ve kept it lower to allow for single applications to stay at $35.

Shoutout to Bob Brauneis, who's working on modernizing registration as a scholar in residence.  Parterned with Stanford University Law & Policy Lab; also Barbara Ringer honors program, 2 year paid fellowships like clerking for a judge.

meaningless numbers aren't advertising

MGM Resorts Intern. v. Pacific Pillows, LLC, 2014 WL 2434628, No. 2:13–CV–1404 (D. Nev. May 28, 2014)

Pacific Pillows lets consumers buy bedding products that are used in various hotels including several hotels owned by MGM. Its website allows people to “shop by hotel,” and (at least some) pillows are named after hotels that MGM owns.  MGM also sells pillows and bedding using MGM’s marks for its hotels.  MGM sued Pacific Pillows for trademark infringement, counterfeiting, false advertising, and unfair competition.  Pacific Pillows counterclaimed, to no avail.

The Sherman and Clayton Act claims were dismissed because there is no antitrust law left.  MGM’s allegedly false advertising on its pillow tags was also not actionable under the Lanham Act.  Pacific Pillows lacked a  “concrete and particularized” injury, because all it alleged was that MGM’s advertising created “significant confusion in the marketplace” and harmed Pacific Pillows, which was too vague and speculative.  Even with standing, the claim failed. The allegedly false advertising was an incorrect registry number on pillow tags, but a registry number isn’t a commercial advertisement.  “The incorrect registry number is simply a series of letters and numbers that means nothing to consumers. The number does not give typical consumers any information about the pillow, and it certainly does not influence customers to buy the pillow.”  Even if other parts of the tags (the logo) were advertising, that didn’t make the registry number a commercial ad.  Also, Pacific Pillows failed to plead that the registry number was likely to influence purchase decisions or cause injury to Pacific Pillows.

Separately, Pacific Pillows alleged that MGM falsely advertised a pillow named “Aria,” because “Aria” wasn’t the exclusive pillow used at the Aria hotel, and that a pillow similar to the “Aria” pillow is only used in a small number of rooms. Pacific Pillow also lacked standing to bring this claim.  Further, it didn’t allege a false statement of fact: it didn’t allege that MGM advertised that a customer who purchased “Aria” would get the same pillow that is found in every room at the Aria hotel. Since both sides agreed that a similar pillow is found in some rooms in the Aria hotel, there was no falsity.

Common-law unfair competition failed too, as did wrongful interference with contractual relations/prospective economic advantage.  Though MGM allegedly contacted Pacific Pillows suppliers, and thus did intentionally interfere with contractual relationships, Pacific Pillows didn’t successfully plead that this interference was improper.  MGM notified the suppliers that they were supplying a company that was allegedly infringing MGM’s marks.  Protection of its trademarks was legitimate.

(pot)head of liability: Hershey's sues marijuana candy producer

Here's the story.  I've been waiting for this for years, ever since a NYT story that included similar illustrations.  As with e-cigarettes, an unregulated field full of small players tends to have a lot of playing fast and loose with all laws.  Take this quote from the story: "'I think it's just a simple market strategy - mimicking brands that are pre-existing,' said Vivian McPeak who is the Executive Director of Seattle Hempfest." Still, is anyone really likely to be confused?

Reefer's Cup and Reese's

HT Patricia Williams/The Trademark Blog.

unregistrable means unprotectable by sec. 43 as well

Renna v. County of Union, N.J., 2014 WL 2435775, No. 2:11–3328 (D.N.J. May 29, 2014)

This is a well-written opinion whose legal conclusions are in part obvious but nice to have down in print and in part quite striking: I think for the first time, a court clearly rules that a mark specifically excluded from federal registration on policy grounds may not take advantage of §43(a) either.  If I were the attorney for Washington’s slur-named football team, I would be paying attention.

Renna, an outspoken critic of the County Board, produced a public access television show, the “Union County Citizen’s Forum.” The show displayed a graphic illustration depicting the Seal of the County of Union with a spotlight shining on it, symbolizing “the self-proclaimed mission of the show to shine a critical light on the workings of the Union County Board of Chosen Freeholders…. It would appear in the background as Renna read Board resolutions or interviewed guests on the air.” 

Screenshot of spotlight logo

A tidbit from the opinion:

This Seal, by the way, may be the only official County seal to depict a murder. The central illustration is an artist’s re-creation of the fatal shooting of Hannah Caldwell during the Revolutionary War. Hannah resided with her husband, “rebel pastor” James Caldwell, in the Connecticut Farms area of what today is Union Township. Her death in 1780, allegedly at the hands of a British soldier, was used to further galvanize public sentiment against England and its royalist sympathizers.

The County Attorney sent a C&D to the television station, warning it to stop displaying the seal and claiming trademark infringement.  “Whether the County has tried to bully a constituent is for the public to decide.” What it didn’t do was assert any legitimate rights. Declining to reach the First Amendment issues, the court found that Union County had no trademark rights in its seal, and Renna’s display in connection with her TV show did not infringe.  (If I were the County, I’d get out my checkbook for fees.  The Rutherford Institute represented Renna.)

The County’s C&D to the Township of Cranford, which operated the public access channel, said:

Re: Unauthorized usage of the seal of the County of Union Union County Citizens Forum

To Whom It May Concern:

It has come to the attention of the County of Union that your entity is using the Seal of the County of Union without proper authorization. The County of Union demands that your entity cease and desist use of the Seal of the County of Union in any way including, but not limited to, displaying it in the background of all television shows with exception of the prerecorded unedited meetings of the Board of Chosen Freeholders. Please be advised that the County Seal is also a pending trademark, therefore you are committing trademark infringement.

Renna discussed the letter with the TV station manager, who asked her not to use the Seal graphic.  Meanwhile, the USPTO refused the County’s pending trademark registration application, “because the applied-for mark consists of an insignia of a U.S. municipality.” 15 U.S.C. § 1052(b).

When Renna resumed taping (after an unrelated delay), she replaced the old logo with a new one, a spotlight shining on a photograph of the Union County Manager.   

Revised logo

This resulted in another request from the Cranford Township laywer to stop using that logo. Then the management requested that Renna sign an indemnification agreement, which made her nervous and caused her to seek legal counsel. 

Her counsel then sent a letter to the Union County Counsel who’d sent the C&D, pointing out that the County was wrong in a number of ways, with the consequence of chilling Renna’s First Amendment rights.  Counsel requested that the County withdraw its demands.  Instead, the County doubled down, claiming, “For your information, this Seal is in fact now trademarked.” The County maintained that the seal was “trademarked under Federal law”; that by state statute, the County was required to maintain a seal; that use by others was unauthorized; that State law “prohibits a third party, such as your client, from registering a mark that uses such an insignia”; and that there was “no impairment of your client’s first amendment right.” The County “simply is protecting its official seal, protected by trademark, from direct and unauthorized use.”

Meanwhile, the County failed to respond to the office action denying registration, and the registration was abandoned.  The County tried to revive its application, but of course revival would’ve been futile, and the application was again abandoned.

The court ruled that, given Section 2, 15 U.S.C. § 1052(b), the Seal was not a “valid and legally protectable mark.”  Section 2(b) provides an absolute bar to registration of matter that “[c]onsists of or comprises the flag or coat of arms or other insignia of the United States, or of any State or municipality, or of any foreign nation, or any simulation thereof.”

The County argued that this barred registration only by nongovernmental parties, not by the government body itself.  “The USPTO, however, correctly and authoritatively interpreted Section 1052(b) as an ‘absolute bar,’ one that ‘does not list any exceptions that would allow for countries, states or municipalities to register their own flags or insignia.’”  The Federal Circuit has agreed with the PTO.  In Re City of Houston, 731 F.3d 1326 (Fed. Cir. 2013).  While G-d  and the Supreme Court know that the fact that the Federal Circuit came to a conclusion doesn’t mean it’s right, the reasoning itself is sound. The Lanham Act must be read according to its terms, and not as a mandate to protect the public from “pirates and cheats” of all stripes. “And it would be particularly inappropriate to read a silent exception into section 2(b), because the surrounding provisions demonstrate that Congress knew how to express such an exception when it wanted to.”  

In a footnote, the court noted that it was bound by the statute’s words.  But it also defended the lack of any special exemption for government entities as consistent with the Lanham Act’s purpose.  “The registration bar was not enacted to protect official prerogatives or preserve official symbols from desecration. Rather, the bar represents a more general determination that state insignia are not appropriate subjects of trademark law at all. Trademark law concerns itself with goods and services in commerce.”  McCarthy explains the bar as based on the theory that government insignia shouldn’t be registered as symbols of origin for goods and services because such insignia “ought to be kept solely to signify the government and not to be sullied or debased by use as symbols of business and trade.” The court concluded that it could “easily imagine the absurdities that would arise from the trademark registration of flags and insignia, which are intended to be used by the citizenry.”

In its communications with Renna’s counsel, the County didn’t acknowledge the USPTO’s authoritative rejection, but instead claimed that the seal was “now trademarked under federal law,” which—given the earlier reference to a pending mark, which had to mean registration—“carried the misleading implication that some intervening event had solidified the trademark status of the Seal” when the opposite was true.  To that misleading implication, the county’s lawyer added another, claiming that New Jersey law barred third parties from registering insignia.  But the state statute didn’t say that—it said what the Lanham Act says, imposing a categorical bar.  The letter then added a “non sequitur”: “Indeed, N.J.S.A. 52:2–4 makes unauthorized use of the State seal a crime.”  While this is true, the state seal wasn’t at issue.  “It is hard to discern any purpose, other than general intimidation, for the citation of this criminal statute in an official communication to a citizen, even one represented by counsel.”  The stated basis for the C&D was illusory.

Okay, so what about § 43(a)?  The County argued that its seal was a protectable mark “because its letters and symbols signify only one thing: the County of Union and the services the County provides,” as well as being fanciful/inherently distinctive.  The doctrine says that the Lanham Act “protects unregistered marks to the same extent as registered marks because trademark rights emanate from use and not merely registration.”  Registration is not a prerequisite for protection—but validity is.  The court concluded that there was “a difference between a mark that happens to be unregistered, and one that cannot be registered as a matter of law” (emphasis added).

Section 2 was best interpreted as determining that certain marks “are inappropriate subjects for trademark protection.”  It followed that an unregistrable mark was not actionable under §43.  There are no cases finding unregistrable marks protected under §43, which made sense because of the broader purpose of §2 to distinguish “fit from unfit” subjects of trademark protection. Section 2’s bar is substantive, not a function of the registration process.

Two Pesos supported this result, using §2 standards to determine validity in a §43 case.  Though §43 is broader than §32, that doesn’t mean it applies to “a different kind or class of mark.”  Justice Stevens’ concurrence further elaborated that “§43(a) is properly understood to provide protection in accordance with the standards for registration in §2.”  Wal-Mart likewise applied §2 analysis about distinctiveness to a §43 case.  (Consider whether a deceptive mark could get §43 protection.  The court’s reasoning suggests that there’s a super-unclean hands defense whenever a plaintiff’s mark is itself confusing/deceptive.) 

This all made sense.  Consistency was an “independent virtue”: as McCarthy said, “What the law does not need is a separate set of different substantive trademark rules followed in § 43(a) cases.... A plethora of different rules and standards provides neither predictability nor consistency, both hallmarks of a rational and democratic legal system.”  Plus, the whole point of §43 was to provide unregistered marks with the same protection as registered marks.  “A mark should not earn extra protection because it is not registered. The Congressional scheme would be scrambled if Section 43(a) were used to protect marks that could never have received any protection as registered marks.”

Thus, Renna’s use could not violate any County trademark rights; there were none.

Independently, even assuming that the County had trademark rights, trademark law has a particular purpose, to which it should be confined: the use of a mark in connection with furnishing goods or services.  “Trademark law is not properly employed to stifle discussion. Microsoft Corporation could not use trademark law to enjoin a consumer from saying ‘I prefer Apple® to Microsoft®,’ nor could Apple do the reverse. Such comments simply have nothing to do with protecting a purchaser mistaking the source of goods.”  Use in the course of criticism is not infringement.  Even if the County considers Renna’s statements inaccurate, trademark law “protects only against mistaken purchasing decisions and not against confusion generally,” which must be corrected with counterspeech or, if appropriate, a libel claim.  This rule implements First Amendment values.  And if that’s true of the limits of trademark as applied to a private company, it is much more so with respect to a governmental body. 

But the court didn’t have to decide any First Amendment issue, though Renna’s arguments on this point were persuasive.  Renna’s use was clearly expressive and political, “entitled to the highest degree of constitutional protection.”  If the US couldn’t criminalize the desecration of the US flag, “an artful extension of trademark law” to quash political expression using a county seal would be “both unwarranted and Constitutionally risky.”  Not before the court was whether a state or municipality could enjoin deceptive misuse of official insignia, as opposed to expressive statements.  The county submitted no evidence that any viewer had been or was likely to be deceived by Renna’s logo.

Comment: And now we see the sub rosa encoding of a confusion analysis.  But I’d say this makes more sense than many such moves (*cough*Louboutin*cough*), since the relevant confusion in an official insignia case would be confusion about the source of authority.  The risk of deceptive use of insignia is not that people will buy a T-shirt from the wrong person; the risk is that they will obey or give money to someone believing that someone to be an official/have official sanction.  That’s just not a trademark concern, but it is reasonably addressed by laws against impersonation.