Actually, the AP likes fair use after all

At least when it's the AP doing the copying! TPM reports that an internal AP memo explains its fact-checking process on Sarah Palin's new book:

"The AP was determined to get the first copy," Oreskes [a senior managing editor] wrote, detailing how the writers learned a store had "inadvertently placed the book on sale five days before its official Nov. 17 release date." "They bought a copy, ripped it from its spine and scanned it into the system so it could be read and electronically searched," he wrote.

I do think this is fair use: it's an intermediate use with the ultimate product distributed to the public being a fair use, indeed a canonical fair use--quotation for purposes of critical review. (Albeit a scoop, a little reminiscent of Harper & Row v. Nation.) But it would behoove the AP to take note that a robust fair use doctrine is good for news reporting, given the AP's high-protectionist stance in other instances.

Another reason I'm glad to be a vegetarian

Sushi fraud: Apparently, if you don't walk around fingerprinting the DNA of your fish--the database is, hilariously, FISH-BOL--you run the risk of ordering tuna and instead receiving escolar, "a nasty fish with buttery flesh that can cause bizarre episodes of diarrhea, accompanied by a waxy intestinal discharge," or an endangered species of tuna. "[R]esearchers from Columbia University and the American Museum of Natural History ordered tuna from 31 sushi restaurants and then used genetic tests to determine the species of fishes in those dishes. More than half of those eateries misrepresented, or couldn’t clarify the type of fish they were mongering."

Alternate post title: California roll, for the inevitable consumer protection lawsuit. HT: Eric Goldman.

Non-survey experts in trademark cases

Flagstar Bank, FSB v. Freestar Bank, N.A., 2009 WL 3837145 (C.D. Ill.)

I am very interested in the use of non-survey expert evidence in trademark and false advertising cases; I think it has a lot of potential to improve the analysis, especially given how manipulable surveys are—reasoning from what we know about human communication and understanding generally may well be more reliable than a claim-specific survey. But it’s not often done, and the judge’s careful dissection of a proffered expert opinion in this case shows that anyone attempting to do it needs to be attentive to the Daubert standards, probably more so than with a conventional survey. I still think a good expert in practical linguistics is worth the investment in an appropriate case. Partially this is because an expert will often be so much cheaper than a survey plus a survey expert, especially given that you can often expect a big, expensive admissibility battle with a survey even if it will probably survive to be admitted. But it’s not just about expense: I really believe that the addition of expert knowledge about how consumers make meaning would improve the quality of adjudication beyond claim-specific surveys.

The underlying dispute: Flagstar is a Michigan-based bank with 175 “banking centers” in Michigan, Indiana, and Georgia, offering checking and savings accounts, home mortgages, and money market accounts. It also operates 104 home loan centers, including 5 within Illinois. They involve mortgage origination only, not deposits or withdrawals, and the loan centers independently decide how to use Flagstar marketing materials in local ads. Prospective customers must undergo a credit check before opening a Flagstar account. It has several registered trademarks and spent $12.3 million on advertising in 2008.

Freestar is a small community bank offering traditional banking services, including checking and savings accounts and home mortgage loans. It has 13 branches, all located within a three-county area of Central Illinois: Champaign, Livingston, and McLean, where Flagstar has no operations. Freestar, which advertises only in that area, adopted the Freestar Bank name in 2006 as a rebranding.

The court ruled on a Daubert motion to exclude proffered expert witnesses. “Dr. Edward Lee Lamoureux holds a Ph.D. from the University of Oregon in Rhetoric and Communication with an emphasis on conversation analysis, rhetoric, qualitative research methods, general speech, and interpersonal communication.” The court accepted him as an expert on “social” linguistics, as distinguished from “formal” linguistics. His report claimed that “flag,” “free,” and “star” are each “strongly associated with some of the most broadly shared values in American culture: Patriotism, loyalty, national identity, and individual rights.” Given the “significant and overlapping metaphorical associations,” he concluded that substantial consumer confusion was likely.

Freestar argued that Dr. Lamoureux failed to employ any identifiable methodology, though he said he consulted “classic texts” in the field of metaphorical association. Except that association is not confusion. He would have been better served to consult the word studies in which people tend to make mistakes about what words they’ve seen on a list when they’re later presented with words semantically related to the words that really were on the list (e.g., put “bone” on the list and people are more likely to mistakenly think that “dog” was on there as well than they are to think that “cat” was there). That would be the missing link in the claim here, which is a variation on the trademark rule—which does not generally require the submission of specific evidence—that similarity in meaning increases the likelihood of confusion. Courts have found infringement based on similar meaning: Lollipops and Jellybeans for roller skating rinks were found confusing. That’s context-dependent; with respect to desserts, lollipops and jellybeans have very different meanings. But precisely because they’re arbitrary for roller rinks, the consumer might easily encode “sweet treat” but not remember which sweet treat served as the name of the business she’d encountered before. Likewise, one could argue—though I am far from convinced on the facts as stated in this case, and I think the descriptiveness/laudatory nature of the terms is important—that the patriotic associations of “flag” and “free” combined with “star” are so similar that, in context, consumers might not remember which they’d seen before.

This is beside the point, because the court noted that, if the expert can’t specify methodology, it’s impossible to evaluate the propriety of that methodology. Here, Dr. Lamoureux didn’t offer any “proposal, theory, or technique” justifying the conclusion of likely confusion. He didn’t footnote or attach supplements explaining the theories on which he relied, or discuss the “classic texts” in his report. The court wouldn’t speculate on what those might be or how they related to his ultimate conclusion. Dr. Lamoureux proposed no theory explaining “how one particular metaphor shared by three words becomes so dominant in customer[s’] minds that it overcomes the many other metaphorical associations attributable to the words.” Moreover, he did not rely on any polls or qualitative research supporting his claim that patriotic metaphorical associations are particularly important post-Sept. 11, 2001, nor—more importantly—did he connect increased patriotic fervor with likely confusion.

The absence of a reliable theory was highlighted by Dr. Lamoureux’s inability to explain how he’d evaluate whether other terms such as “freedom,” “America,” “liberty,” and “patriot” fall under the same metaphorical umbrella. This concern, I think, is related to the descriptiveness/laudatory caution I offered above. There are good reasons to be hesitant to give Flagstar rights over the patriotic metaphor as applied to banks, while monopoly rights over sweet dessert names applied to roller rinks are much less troublesome. To the court, his methodology appeared result-driven.

Aside from reliability concerns, there were also relevancy issues. Expert testimony may be excluded if the primary facts can be accurately and intelligibly described for the jury and if the jury is just as competent as drawing conclusions from the facts as are expert witnesses. Dr. Lamoureux’s two-page report wasn’t particularly helpful because the trier of fact can also evaluate the meanings of the words comprising the parties’ marks. “When presented with the parties’ marks, the average person is capable of concluding that all three words share a patriotic connotation.” A juror could also recognize differences in meaning. Expert testimony would be helpful in explaining how similarities lead to confusion, or how differences negate likely confusion, but that’s not what was on offer here. Thus, Freestar’s motion to exclude was granted.

On to infringement: the court swiftly disposed of Freestar’s argument that Flagstar lacked constitutional standing for failure to suffer injury in fact. But Flagstar clearly alleged an invasion of a legally protected interest, that protected by its federal registrations; this was sufficient.

On likely confusion, however, Flagstar’s evidence fell short. (Here let me praise the court for including multiple images in the opinion, allowing readers to follow along.)

“Flag” and “free” were distinctly different elements of the mark, contributing to distinct overall impressions, especially given that Flagstar’s formatted mark used mostly lower-case letters, included a slogan (“The new wave in banking”), and was accompanied by a black graphic suggesting the waves of a flag. Freestar’s mark had italicized capital letters and a very different slogan (“Life keeps getting better!”), along with a green A and a star-shaped graphic, invoking shooting star imagery. The slogans were important differentiators—Flagstar’s describes its services, and Freestar’s focuses on the reader.

In the marketplace, Flagstar’s mark prominently features the color red, while Freestar’s pending registration uses black, green, and white. Based on these dissimilarities, a reasonable consumer would not be confused. There was greater similarity with Flagstar’s “Legacy mark,” a mark it still uses on occasion—both marks featured italicized capital lettering, the work “Bank” at the lower right, and a star graphic in the midle of the letter A in “Star.” But the wave graphic, the slogan, the “shooting” appearance of Freestar’s star, and the colors actually used in the marketplace still left the marks readily distinguishable. “A customer is not likely to be confused between the origin of marks with different names and completely different colors and slogans,” especially when one graphic invokes a waving flag and the other a shooting star.

Intent: there was no evidence of intent to pass off, but Flagstar argued that Freestar failed to exercise due diligence in rebranding, given that its registrations provide constructive notice. However, even actual knowledge is not enough to prove bad intent. Bad intent can only be inferred based on similarity where the senior mark has attained great notoriety and is nearly ubiquitous in the area where the junior mark competes. This wasn’t true here.

There was no evidence of actual confusion. Moreover, because of the lack of geographic overlap, Flagstar and Freestar’s services aren’t used concurrently in any area or manner. “An Illinois consumer seeking to open a bank account would need to drive across state lines before encountering a Flagstar banking center. A person seeking a home loan cannot ‘cruise down the street’ and become confused by the presence of a Flagstar home loan center and a Freestar banking center because he or she will not encounter both of these entities within the same county, let alone city.” Customers couldn’t be confused by marks they don’t encounter during their everyday lives.

Nor did Flagstar show that it was reasonable to think that it might expand into Freestar’s counties. Its SEC filings announced an intent to focus on expanding in Michigan and Georgia. Unsurprisingly, there was also no evidence of overlapping marketing channels, given Freestar’s limited local advertising and Flagstar’s failure to show it advertised in those counties. Indeed, Flagstar was unable to confirm that the home loan centers it operates actually use Flagstar advertising materials within Illinois. Though both parties use the internet, the court found no precedent to hold that the maintenance of two independent websites, not linked in any way (such as via metatags), could count as an overlapping marketing channel.

In addition, the court thought that, though banking services are ubiquitous, consumers are likely to exercise a higher degree of care than they do when buying cooking spray or oil changes. Flagstar customers must submit to credit checks; Freestar requires customers to speak with a bank loan officer before obtaining a home loan. Customers wouldn’t carelessly sign themselves up for such “invasive and prolonged inquiries.”

Flagstar also argued that even if banking consumers are sophisticated, they’re still vulnerable to initial interest confusion and reverse confusion. However, there was no evidence that Freestar “lured” consumers, the way the defendant in Promatek Industries, Ltd. v. Equitrac Corp., 300 F.3d 808 (7th Cir. 2002), did by using confusing keywords. (Yes, I know, but let’s just be grateful the court didn’t buy the IIC argument.) And there was no other evidence that anyone would patronize a Freestar bank because of a perceived link with Flagstar.

Nor did the facts support a reverse confusion claim, where a large junior user saturates the market and creates a strong association between its product and the senior user’s mark. Freestar is a small bank, not saturating the market, and there was no evidence that its use, and subsequent goodwill, would hamper Flagstar’s expansion into Central Illinois.

Turning at last to strength of the mark, the court found Flagstar’s mark not particularly strong from an economic/marketing perspective. 175 banking centers and $12.3 million in ads in 2008 is not enormous, and Flagstar failed to provide any evidence about Illinois because it didn’t submit (or apparently even seek) any evidence about how the home loan centers advertise. “A mark cannot bear economic and marketing strength in a place where it does not do business and does not advertise.” The court didn’t even address conceptual strength (which also would have been relevant to reverse confusion.)

Similarity of the products was the only factor really favoring Flagstar: they were identical.

Overall, the court weighed the lack of concurrent use particularly heavily: “Consumers cannot become confused by a mark they will never encounter in the marketplace.” Further, despite the similarities between the marks, as a whole those similarities wouldn’t lead to confusion—and even if there were a material issue on that point, the record was extremely lopsided on the other confusion factors. Thus, Freestar was entitled to summary judgment.

The history of snake oil

From the Smithsonian Museum of American History (yes, it has a blog). And a website for its collection of patent medicines. If you want to know why we have the FDA, take a look. HT: Zach Schrag.

California law allows UCL claim against insurer

Zhang v. Superior Court, 100 Cal.Rptr.3d 803 (Ct. App. 2009)

States deal with nonfederal preclusion as well as preemption issues. Here, the question was whether state insurance regulation precluded a cause of action uncle general unfair competition/ false advertising law. The appellate court, rejecting another appellate ruling, said no. Conduct that violates only the insurance code, which is not enforceable via a private cause of action, can’t support an unfair competition claim. But false advertising is actionable even when insurance-related. This is very similar to the working compromise between the Lanham Act and the FDCA, especially is you take into account the court’s dicta that conduct that violates both laws, UCL and specific provisions of the insurance code, is actionable.

Fundamentally, given that insurance is not excluded from UCL coverage, “[t]o construe the Unfair Insurance Practices Act as immunizing insurers from the consequences of misconduct that other business must suffer would simply make no sense.” The court cautioned that the plaintiff could not recover merely by showing unreasonable handling of her claim. She’d need to show that the insurer falsely advertised to the public “that it operated honestly and equitably in settling claims and that it in fact had a policy or regular practice of ‘lowballing,’ delaying, or taking unfair advantage so that its advertising and/or representations were in fact likely to mislead the public.”

Presenting in Chicago

Tomorrow at noon, I'll be giving the Distinguished Professor Presentation at John Marshall's Center for Intellectual Property Law, talking about "Running the Gamut from A to B: Federal Trademark and False Advertising Law," in which I argue that the Lanham Act would make more sense if we spent more time thinking about the relationship between 43(a)(1)(A) and 43(a)(1)(B). You can register here.

Ignorance of 230 is bliss for lawyer ad regulation

A recent opinion of the South Carolina Ethics Bar reveals a not uncommon failure to appreciate the scope of §230. State bars have been struggling to deal with lawyers’ internet advertising, which can reach potential clients in new ways; anything on the internet, not just a banner ad, might in theory count as advertising. And new intermediaries have sprung up to help potential clients navigate, including lawyer rating services that offer profiles of lawyers. Sometimes for a fee, lawyers can “claim” their profiles and add extra information to make themselves more attractive. The South Carolina ethics body ruled that, when a lawyer claims a profile on such a service, she becomes responsible for its content under the ethics rules, including peer endorsements, the service’s own ratings, and client comments. But §230 was written to prevent states from such attributions unless the user herself—here, the lawyer—provides the content at issue, and the ethics body specifically stated that it was relying on the opposite rule, holding the lawyer responsible for statements of others. Section 230 precludes any liability for the parts of the profile lawyers did not create, including the three features highlighted by the ethics body: peer endorsements, the service’s ratings, and client comments.

Damageless music infringement

EsNtion Records, Inc. v. TritonTM, Inc., 2009 WL 3805827 (N.D. Tex.)

Talk about snatching defeat from the jaws of victory. Watch as an apparently unassailable copyright claim falls apart. Plaintiff EsNtion, which I’m just going to call plaintiff, is an independent record label; it sued defendant TM, which is in the business of programming music for radio stations. TM sells subscriptions to radio stations and DJs, including a weekly subscription called HitDisc, which consists of one or more CDs with songs that subscribers use to update their playlist. Plaintiff sued for unauthorized inclusion of its songs on the CDs, also alleging DMCA violations, trademark infringement, and unfair competition. TM disputed a bunch of the allegations and argued that it got the songs at issue from plaintiff, its recording artists, or a promotional company working for it, and that inclusion on the CDs provides a promotional benefit.

Plaintiff pled infringement of 235 songs, but TM argued, without rebuttal, that it didn’t distribute 211 of them. Whoops. Then, for eight of the remaining songs, plaintiff only pled pending copyright applications. The Fifth Circuit follows the (wrong) rule that jurisdiction requires only that the Copyright Office receive the application, deposit, and fee. But plaintiff didn’t even meet that standard, because it only submitted emails from the Copyright Office indicating that “registration claims” had been submitted, but not indicating that the fee or the material being registered had been received.

Then, TM argued that plaintiff didn’t own or have an exclusive license to many of the recordings at the time of the alleged infringement, with no written agreements in place. The court found evidence of licensing agreements that supported plaintiff’s claims for most of the remaining songs, but three more were knocked out that way.

Plaintiff didn’t register before the infringement began, so it wasn’t entitled to statutory damages or attorney’s fees. All that was left was actual damages, and here’s where disaster really struck. Plaintiff’s president was unable to identify damages from the infringement, including lost sales or contract opportunities. Plaintiff also submitted expert reports, but the court concluded that they didn’t support a finding of actual damage, because neither expert opined on that—one “assumed” that plaintiff would prove its allegations of infringement (I don’t get why this is disqualifying, unless the expert actually assumed that plaintiff would prove damages, not just infringement) and the other stated that the infringement “could” cause damage to an independent record label, but didn’t conclude that there was actual damage.

Under Davis v. The Gap, a reasonable royalty is an alternative method of calculating damages where both defendant’s profit and plaintiff’s lost sales are impossible to figure out—why wasn’t that available? Hypothesis, though there's nothing in the opinion or on defendant's site that makes this clear: because these are promotional CDs, the copyright owners who choose to participate in this market might provide them for free, making the market-set royalty $0. Usually, the law leaves it to copyright owners to decide whether they want to participate in a market, and will find infringement even if the infringement increases the value of the work—but there might be infringement without damage.

In any event, there was also no evidence that TM profited from the infringement. It didn’t sell songs or CDs, but only profited indirectly from selling its subscription service. Its revenues didn’t depend on the songs at issue and no one specifically requested them.

Without any damages, TM won summary judgment on all the copyright claims, including contributory and vicarious infringement. Even worse for plaintiff, TM might be entitled to a fee award as a prevailing party (though, having failed to register on time, plaintiff wouldn’t have been eligible for a fee award had it prevailed), based among other things on bad litigation conduct including failing to amend the complaint to remove the songs that never appeared on TM’s CDs.

Plaintiff also brought a DMCA claim for removal of copyright information and provision of false copyright information. TM argued that it didn’t have the intent to induce, enable, facilitate or conceal infringement, as required for liability under §1201, and that it didn’t remove any information. TM claimed that it had no incentive to remove such information and instead makes it as accurate as possible. And, none of plaintiff’s CDs had any electronically-embedded copyright information; some courts have limited CMI coverage to electronic information, though the statute isn’t written to make that the obvious conclusion. Anyway, though the evidence might show that TM threw away the physical copies of CDs sent to it, the song information, including copyright and other identifying data, was input into a database available to subscribers. The court thus dismissed the DMCA claim.

Plaintiff also brought §43(a)(1)(A) and (a)(1)(B) claims. First, it failed to show harm, as required for constitutional standing. There was no evidence that it lost anything, or that TM gained anything from including the songs. (Not that I mind this foreshortened analysis, but I will note that the typical trademark argument is that loss of control over one’s reputation is sufficient harm in confusion cases.) Even if plaintiff had constitutional standing, it would lack prudential standing. (Again, note that this test is usually only applied to §43(a)(1)(B) false advertising claims, though if you are going to do the ridiculous antitrust-based prudential standing test that has attracted so many circuits then it’s a little less silly to apply it to both branches than applying a competition requirement.) Plaintiff argued that TM engaged in literally false advertising by using plaintiff’s name without permission, causing customers to buy TM’s products instead of plaintiff’s. TM allegedly sold compilation CDs to companies that would otherwise have purchased plaintiff’s products. The court found this argument insufficient, because the evidence didn’t show that the parties competed; TM sold subscription services, not CDs. Nor had plaintiff shown any damages. (Really this claim is barred by Dastar and materiality, not standing.)

Plaintiff’s unfair competition claim failed for want of any independent tort. Plaintiff argued that TM attempted to confuse consumers by using plaintiff’s name on its CDs, but this was just trademark infringement.

Wow.

Jerk doesn't believe in fair use, wants payment for academic quotations. Also, he'd like a pony.

Tarnishing Harvard

From the back cover of the Yale alumni magazine, Nov/Dec 2009. Headline: "Is everything that begins with the letter H only half as good?" The reference is of course to Harvard (and its crimson color). Ad for Coda Automotive.

(As a proud graduate of both schools, I can only offer the best counter-put-down I ever heard, involving one frat boy telling another that frat boy A's fraternity had some really funny songs about frat boy B's fraternity. "Really?" asks frat boy B. "We don't have any songs about you.")

Add another to the overreaching trademark annals

Seattle Trademark Lawyer reports that Car-Freshner has sued Getty Images for licensing for commercial use photos that include Car-Freshner's car deodorant products in the frame. Not that I don't think the claim is ridiculous, but (like the keyword cases) wouldn't the claim be better if it were for contributory infringement? I wonder if anyone's actually used these in an ad.

Reverse confusion claim over highly descriptive mark fails

World Wide Sales, Inc. v. Church & Dwight Co., Inc., 2009 WL 3765881 (N.D. Ill.)

This case caught my eye for a brief moment of insight into the kind of market research large companies tend to have on hand, repurposed for litigation.

World Wide, a small company that made Forever Fresh for the Fridge (a refrigerator deodorizing product) and sold it via infomercials sued Church & Dwight, which makes Arm & Hammer, including Arm & Hammer’s newer Fridge Fresh refrigerator deodorizing product, for infringement—reverse confusion, in particular. Given the highly descriptive nature of the mark—the PTO actually told World Wide to disclaim “fresh” and “fridge”—and the different appearances of the parties’ products, the court had little trouble rejecting the claim.

In running through the factors, the parties disputed the care taken by consumers. World Wide argued that deodorizers are an impulse buy; Church & Dwight puts Fridge Fresh next to conventional baking soda precisely to encourage such “oh, yeah, that looks good” purchases. Church & Dwight, however, argued that baking soda purchases are planned, and cited research “indicating that consumers do not go into the baking aisle of supermarkets unless they are planning to purchase a product,” raising the typical level of care used on low-cost products. The court didn’t agree, concluding that buyers of low-cost refrigerator deodorizers aren’t likely to be all that meticulous in their selection. Given the difference in appearance, though, even a hurried casual consumer wouldn’t be confused.

That baking aisle research intrigues me. I strongly doubt it was conducted for purposes of litigation, and so the questions it asked were likely broader ones. Obviously Church & Dwight would like to get consumers into the baking aisle more regularly, and I wonder what strategies they are pursuing.

Another drug equivalence claim fails to persuade

Graceway Pharmaceuticals, LLC v. River’s Edge Pharmaceuticals, LLC, 2009 WL 3753586 (N.D. Ga.)

Graceway sells prescription benzoyl peroxide, Benziq, for treating acne. River’s Edge markets a benzoyl peroxide product, Benprox, created specifically to compete with Benziq. They’re both available as a 5.25% wash and 5.25% and 2.75% gel. Graceway sued over River’s Edge’s allegedly false and misleading promotion of Benprox as “generically equivalent or otherwise substitutable” for Benziq, and over allegedly false representations on Benprox’s label about active ingredient strength, dosage form, and expiration date.

River’s Edge sent new product submission forms to various pharmaceutical databases, which are key to most retail pharmacy dispensing systems. It was critical to the sale of Benprox that it be listed as a generic or equivalent to Benziq so that it would show up when pharmacists looked up drugs and brand names. As a result of River’s Edge’s submissions, Benprox was linked with Benziq in the First DataBank and Wolters Kluwer databases. River’s Edge argued that its submissions occurred on standardized forms and didn’t include any claims of generic or equivalent status, and didn’t mention Benziq by name. But once the products were linked, it did contact pharmacies and drug distributors to market Benprox as a substitute for Benziq, and entered into drug supply agreements with drug wholesalers to increase Benprox’s market share.

In addition, River’s Edge gave a flyer to Kinray, a wholesale distributor for generic drugs, to give to its customers. It said, “Get your money-saving, quality River’s Edge generic products through Kinray today,” and promoted Benprox as “compet[ing] with” Benziq. River’s Edge argued that the flyer was prepared in a format prescribed by Kinray, which supplied the headline and the table heading “competes with.” It also included a disclaimer that River’s Edge products don’t claim bioequivalence to products they “compete with” unless noted otherwise—a disclaimer that also was on the email transmittal sheet accompanying the new product submission form discussed above.

In a survey of pharmacists, 39% stated that when a database shows that a benzoyl peroxide drug is linked to a specific brand name, it means that the drug is a generic equivalent. River’s Edge argued that the survey was flawed, among other reasons, because it didn’t specify that the question covered unrated, topical products like the parties’. This is an important point: neither party’s products had been tested or approved by the FDA. No NDA for Graceway; no ANDA (used to approve generics for already-approved drugs) for River’s Edge. The FDA publishes the Orange Book of approved drug products with therapeutic equivalence evaluations; neither Benziq nor Benprox is listed and the FDA has not made any equivalency determination.

So: to summarize, what we have is the consequence of a scheme that grandfathered in some drugs. Today, pharmacists (like medical personnel and patients, I imagine) are likely to default to the assumption that FDA standards govern all prescription drugs. If we really took consumer belief seriously, albeit belief generated by an underlying regulatory regime, we would apply FDA standards for approval beyond what the statute actually requires of the grandfathered drugs. And this is precisely the question the survey here might force the court to answer. Can it get out of this conundrum?

The court found that substitutability was governed by state law. The Georgia Pharmaca Practice act allows a pharmacist to substitute a drug that is “pharmaceutically equivalent” to the prescribed brand name product, defined as “drug products that contain identical amounts of the identical active ingredient, in identical dosage forms, but not necessarily containing the same inactive ingredients.” This does not require therapeutic equivalence. By law, if a doctor prescribes a benzoyl peroxide wash or gel without specifying a brand name, a pharmacist shall dispense the lowest retail priced drug product in stock which is, in the pharmacist’s reasonable professional opinion, pharmaceutically equivalent. Based on pricing, that’s Benprox.

As for the dosage form, active ingredient strength, and expiration date, Graceway asserted concerns about the integrity of the manufacturing process (done by Corwood Labs), which underlay much of its argument against substitutability as well. River’s Edge lists a two-year expiration date on its label, on deal sheets to distributors, and on each certificate of analysis. Graceway’s expert found that Benprox failed stability testing due to a change in the drug's color, viscosity, gravity, and benzoyl peroxide strength, falling below specifications six months into accelerated testing. River’s Edge argued that the long-term stability of color, viscosity, and gravity are irrelevant and immaterial; the expert acknowledged that the FDA doesn’t require viscosity and gravity to be measured in a topical benzoyl peroxide product. Moreover, a two-year expiration date is supported by meeting the active ingredient specifications three months into an accelerated test, as Benprox did.

Graceway also argued that River’s Edge misrepresented that some of its products are gels, because Benprox fails to meet the FDA definition of a gel: a “semisolid dosage form that contains a gelling agent to provide stiffness to a solution or a collodial dispersion.” Semisolids don’t flow at low shear stress. But Benprox, because of manufacturing shortcomings, allegedly separates and runs, behaving “more like a lotion than a gel,” and one of River’s Edge’s experts acknowledged that some Benprox gel products “flow like a liquid” and were “pourable.” River’s Edge argued that the FDA Data Standards Manual definition above was irrelevant, and the US Pharmacopeia standard, which is the uniform industry standard, only requires that a benzoyl peroxide gel be in a “suitable gel base.”

Graceway’s expert also examined the certificates of analysis with respect to label strength. Benprox 5.25% wash is expected to contain between 90-110% of the amount listed on the label. But five of the ten batches manufactured by Corwood that he looked at contained over 110%, up to 120%. Three batches of 5.25% gel ranged from 113-119%, while two batches of 2.75% gel contained 120 and 124%.

River’s Edge argued that Graceway didn’t identify an appropriate range of variation for the gel. Also there’s no FDA or USP monograph establishing the appropriate range for a benzoyl peroxide wash. And the FDA doesn’t recognize “wash” as a dosage form at all, so for the court to find such would preempt the FDA’s responsibility. Alternatively, River’s Edge submitted the USP definition of a benzoyl peroxide gel, which used a range of 90-125%, and argued that the court should apply the gel definition to the unrecognized “wash,” which Corwood describes as an opaque gel. Once viewed as a gel, all batches of Benprox wash fall within the USP guidelines of a benzoyl peroxide gel.

The first issue for the court was whether the FDA’s authority under the FDCA precluded its determination of any of these issues. There’s no automatic bar to Lanham Act claims, and in particular courts may evaluate claims of equivalance between non-Orange Book drugs. This case didn’t require the court to interpret the FDCA or FDA regulations in determining truthfulness. However, with respect to the labeling claims, the court wanted to be careful not to get too close to the FDA’s exclusive enforcement domain.

Ultimately, the court found that for it to take a stance on the challenged representations of dosage form, expiration date, and strength would go too far. The dosage form claim was limited to the Benprox gel, premised on the fact that it didn’t meet the FDA definition of a gel. However, the USP standard was the industry standard, and only required a “suitable gel base”; Graceway hadn’t alleged falsity under that definition. The court refused to hold that the FDA definition of a gel is the only acceptable standard for a non-FDA approved drug; that would be a matter for the FDA.

As to strength, the court also rejected Graceway’s claims that Benprox varied too much. The allowable range Graceway alleged was not based on an FDA or USP monograph for a benzoyl peroxide wash, because no such monograph exists, and anyway the FDA doesn’t recognize wash as a dosage form. The court wouldn’t define a dosage form without backing from the FDA or USP. As a result, there was no established range of variation in strength, and no way for a jury to decide whether the percentage in the Benprox wash was appropriate.

As to expiration date, Graceway’s expert conceded that the percentage of benzoyl peroxide three months into accelerated testing was sufficient to warrant a 2-year expiration date. And the other changes in color, viscosity, and gravity were not enough. The FDA doesn’t care about them for topical benzoyl peroxide products, and Georgia law only requires identical amounts of the same active ingredient in the same dosage form for pharmaceutical equivalence. Graceway didn’t demonstrate that color was relevant to those things, and thus to substitutability, and so it didn’t show that color change was material. Viscosity or specific gravity might be relevant to whether Benprox was properly labeled a gel or a wash, but there was no evidence that the difference was material.

As a result, River’s Edge won summary judgment on all the composition/label claims.

What remained was substitutability. The court assumed for the sake of argument that River’s Edge falsely represented generic equivalence, but found that any misrepresentation was immaterial, because there was no evidence that the Kinray flyer affected pharmacies’ decisions to buy Benprox or pharmacists’ decisions to dispense it in place of Benziq. In Georgia, generic or therapeutic equivalence is not required to substitute a drug, only pharmaceutical equivalence. Graceway didn’t provide any evidence that would allow a jury to find that Benprox was not pharmaceutically equivalent to Benziq. Thus, any misrepresentation would not have been material to substitutions. In regards to the New Product Submission form in particular (listing the active ingredient, strength, and dosage form), there was no evidence from which a jury could conclude that any of these representations were false.

Graceway argued that River’s Edge hadn’t conducted any comparison testing to determine the accuracy of its substitutability claim, but it wasn’t River’s Edge’s burden to do so. (If the claim is of the sort that reasonable pharmacists would believe was backed up with comparison tests, then it’s a necessarily implied “tests prove” claim and can be falsified by showing the absence of tests. However, it seems more plausible that a substitutability claim in the pharmaceutical context is reasonably understood to be backed up by some scientific evidence, not necessarily comparison testing, and River’s Edge does seem to have plenty of evidence about the active ingredient in its product, which would satisfy that standard.)

The court granted summary judgment on the Lanham Act claims and the coordinate state law claims. It also rejected the common-law misappropriation claim. River’s Edge gained an advantage because it wasn’t burdened with the expenses of development and marketing, and its business plan was obviously likely to do just that, but it’s not illegal to copy an unpatented article.

Buffalo Panel #4/Putting Advertising in Context

William O’Barr, Duke University, Alternative Forms of Advertising Regulation: Comparative Notes from China, India, Brazil, and the US

From most people’s perspectives, law involves writing a will and buying a house, little more. The caselaw doesn’t touch ordinary people’s lives or make up the stuff of law in practice.

Comparative perspective from big countries. China: very highly regulated; India and US: moderate; Brazil: low. Primary methods of regulation are very different. China: government. India: cultural taboos. US: Industry tries very hard to preempt government regulation through various self-regulatory bodies, especially for TV ads, staying one step ahead of regulation. Brazil: the ad agency decides what to do.

Pepsi makes a pattern ad they’d like to use all over the world, but will have to be reshot in various ways for local regulations and ambiance. He showed a popular Pepsi ad with Michael J. Fox; Chinese censors didn’t like it because it showed a number of inappropriate behaviors—running down a fire escape in a nonemergency situation, crossing in the middle of the street, gangs in the city, etc. Chinese version: similar dialogue, very different behavior—he crosses at a crosswalk! Chinese regulators didn’t want to show antisocial behavior. Not an isolated instance.

India: (incidentally, here’s the Indian version of that ad—worth a look, especially in light of the Chinese changes) A private company started competing with the government in condom distribution; people didn’t like how thick the government issue condom was, though it was promoted heavily for population control. Chose the name Kamasutra to emphasize its greater sensuality. Can’t show men and women kissing in movies or on stage; this is a huge taboo. Nor can a man touch a woman’s hips or most parts of her body in an ad. Ad agency works within the context of taboos. The condom ad included a prominent framed picture of the couple—showing that they weren’t engaged in casual sex. Ads can be erotic—he showed a condom ad that probably couldn’t be exceeded in the US—and yet there are still taboos.

Brazil: Creative strategy has become divorced from the product—lost the purpose of selling stuff. “Ghost ad”: made specifically to enter into international competition, but don’t show on TV or in magazines. There’s just no regulation. Ketchup ad: won a competition, didn’t show at home. O'Barr then showed an ad in which boys repeatedly ogled and spied on women, to advertise candy—forget government regulators in the US; this wouldn’t make it past the network censors for concerns over sexualizing kids, objectifying women, general creepiness.

Ramsey: How long does ad approval take in China?

A: Several months. Tedious. Many are rejected and need to be replaced/remade. China has a good side: what the government is requiring is for advertisers to think about whether the values promoted in ads are antisocial and thus contrary to the population’s larger interests. He wishes we could do some of that in the US.

Bradford: OK, so what about niche marketing to homosexuals in China?

A: Wouldn’t get talked about at all: standardization of the ideal family. Same thing with any totalitarian regime: they have an idea of what the images ought to be and enforce them heavily. Also doesn’t allow regional minorities to appear—multiculturalism doesn’t show up in China.

Bartholomew: former student did a presentation on Chinese ads. An ad in which a boy bought two Cokes so that he could stand on them and get tall enough to press the Pepsi button, then abandoning the Cokes, would never be allowed.

A: That’s because comparative ads aren’t allowed. You can’t say your car is best because everyone knows the other brands work fine.

He thinks Brazilian ads are exceptionally cool, and Chinese ads are fascinating because the ads thrive in the midst of this totalitarian state. And why can’t we advertise condoms in this country?

Stauffer: That Brazilian ad is really disturbing, particularly in dialogue with the papers from earlier in the day about the recruitment of women’s bodies to use sales—the social education that boys are supposed to be voyeurs, stick a mirror under the teacher’s dress, etc. Unlikely to be ads in which little girls fantasize about men’s bodies. That doesn’t have the same social meaning.

A: Note that these things are all competing to be the standard for global advertising. The US doesn’t come out very well. US ads are no-nonsense, WYSIWYG, not beautiful, not creative, not generalizable because the American population looks different from every other population—the set of people chosen to represent multiculturalism is unrepresentative everywhere else (blacks, Hispanics, Asians along with whites); too many facts/reasons compared to French/Japanese ads. We’d be wise as Americans to realize how very culturally specific these issues we’ve been talking about today are.

There’s no pretense of multiculturalism in ads elsewhere—they don’t bother to put Muslims in ads in France. American racial codes are different—a code for communicating sensitivity to diversity that is uniquely American, so we show three kids brushing their teeth, one white, one Hispanic, one Asian—an improbable social situation created to show multiculturalism. But there are a small number of niche markets that warrant special treatment in the US—Hispanic, African-American, gay/lesbian; now Asian, though ads tend to make them honorary whites.

Daniel Horowitz, Smith College, David Riesman: From Law to Social Criticism

Lawyer who made his mark as a sociologist in The Lonely Crowd; turned his back on the law. Buffalo was actually critical to his evolution; he came to the law school when the dean was trying to turn it into a national law school, and began to build his career as a sociologist, pushing the boundaries of what was proper to study. He was interested in empirical data, using it to examine the law of finders. Riesman worried about defamation of Jews in Europe. Tried to deal with that in law review articles about the conflict between defamation and free speech. What linked his legal training with his work in sociology? The Lonely Crowd is not obviously a book written by a lawyer.

Horowitz proposes: the link is concern for the conditions under which democracy can continue to flourish under adverse conditions. He began with the problem of political apathy, and tried to figure out how to preserve individual autonomy from attack by totalitarianism in Europe and consumer culture in the US. Often misread as pessimistic.

At several points, he focused on advertising, suggesting that educating consumers might promote autonomy. Wanted “leisure counselors” to educate Americans, especially children, how to consumer, and wanted market research to uncover consumer desires and meet them. Understood the individualizing potentials of mass media, contrary to Adorno and the Frankfurt School, and tried to understand how consumer culture individualizes people and provides a source of resistance to the pressure of the peer group. Feedback and the importance of consumers talking back to corporations and advertisements. The other-directed personality promises flexibility and self-expression, and ads/market research could help Americans resist conformity and seek autonomy. He was countering the pessimism of the Frankfurt School.

Bartholomew: What episodes from popular culture did he find hopeful?

A: Later essays, yes. Charlie Chaplin, American jazz are commonly cited by writers of this bent. Women & sex in The Lonely Crowd—he moves towards women’s liberation through sexual experience: women are acting in their homes more like courtesans.

Alberto Salazar, York University: Consumer Counter-Advertising Law and Corporate Social Responsibility in Canada

In Canada, the law chills consumer expression/counteradvertising, which prevents democratic demate and social deconstruction and reconstruction of the commodification of life. McDonald’s litigated in the UK over the use of “McDollars, McGreeedy, McCancer, McMurder, McDisease” for 6 years, finally losing at the European Human Rights Commission. Court said: lack of government aid to consumers to prove the accuracy of their criticism against a corporation is a violation of their freedom of expression.

In Canada, defamation law is considered private. Defenses are costly, and there’s no anti-SLAPP legislation. Most people are sympathetic to consumer expression and yet the law doesn’t facilitate it.

Example of problems: discourse on obesity, anorexia, healthy eating. Change in law could stimulate behavioral change, past the “public interest responsible journalism defence,” which is not clearly available to ordinary nonjournalist citizens.

Ramsey: Do firms use TM law as well?

A: Hasn’t studied that.

Ramsey: How rigid is the standard? Negligence?

A: No, truth. Even nonnegligent falsehood leads to liability, and you have to prove truth.

Buffalo Panel #3/New Forms of Advertising Regulation

Laura Bradford, George Mason Law: Sponsorship Confusion

Consumers would prefer more transparent information about sponsorship, but the market is underproviding because of agency and other costs. It’s becoming more and more difficult to tell advertising from organic speech—viral YouTube videos and so on. Search engine results: why do results come up first? More of an issue before Google.

Historically, we have TM owners police deceptive uses of marks. They have more incentive to watch the marketplace carefully. But seller interests in sponsorship disclosure diverge from consumer interests—an agency cost, if we are concerned with consumer welfare. The linkage of a product with a well-known brand tells a consumer that someone with a lot to lose has control over the product: Apple’s reputation sells iPods; McDonald’s reputation sells burgers. But then there’s sponsorship—95% of new product introductions are brand extensions, and cobranding is also on the rise, allowing each brand access to the other’s customers. McDonald’s and Disney market Happy Meals that promote movies and food.

Some buyers punish brands for affiliation—Converse was an indie sneaker, until it was bought by Nike and some buyers abandoned it. Now the Black Spot sneaker, produced by a family business in Portugal, intended to show distance from artificial sameness of mass production, and (some) consumers want this.

Sponsorship and affiliation affects consumer preferences and gives them information about unobservable qualities. Consumers are also interested in learning about compatible products that may not be sold by the TM owner; may want to compare; but they do get useful information from sponsorship. (This doesn’t seem to go to sponsorship of communications v. sponsorship of products and services.)

Sellers: want to preserve the value and meaning of their own marks. Sellers are hurt by lack of common language for disclosing affiliation. They rely on logos/distinctive typefaces to indicate sponsorship, which means that TM owners are forced to police uses in unrelated markets. And she means forced; had to advise clients that they needed to send threat letters even in unrelated markets in order to preserve the right to go after real threats in the future. They have to object when anyone, anywhere accurately depicts the mark/logo. This is a coordination problem: TM owners can’t get together to provide a common language. There’s also an agency problem, because some benefit from uncertainty. There’s no penalty for obscuring sponsorship, as with product placement or guerrilla marketing. Consumers are getting more wary, but that just means that the real harm is the ability to have and trust organic speech.

Another harm: this gives TM owners broad power to pursue other goals. Consumer uncertainty = TM owners can safeguard preferred distribution channels by alleging affiliation confusion, shut down secondary markets in used goods, prevent competitors from using brand names in truthful ads for compatibility or comparability, police brand image by suppressing objectionable speech, etc. Advertisers don’t fully share the costs to consumers of uncertainty and get offsetting benefits.

Since TM law creates the problem by assigning enforcement of consumer interests to producers, TM law should deal with that problem by creating a new mark to realign incentives. The new (s) symbol would indicate a licensed use, and they’d be entitled to current levels of protection against sponsorship/endorsement confusion. Those who didn’t do so would need to provide clear and convincing evidence of confusion and materiality. Asks sellers to stand by their choices. If they want freedom to engage in ambiguous strategies, have to allow others to do so themselves. Shareholder derivative lawsuits: plaintiffs’ attorneys are incentivized to find potential harms, but because of the agency costs—risks of meritless lawsuits—we impose a heightened pleading standard.

Three possible outcomes, all an improvement: (1) everyone uses the (s) mark. Affiliation becomes much more transparent. Everyone learns to recognize it. (2) No one uses it; advertisers want to blur the lines. Heightened standard allows more freedom for organic users. Advertisers who don’t like it can use the (s). (3) Inconsistent use, most likely. Consumers remain uncertain. But the availability of the symbol may change courts’ approach—organic use would be given more breathing room. We’d also get more revealed preferences of sellers—when sponsorship information is seen as critical and when less important, for example by industry.

Katyal: Sees how it would work visually; how would it work orally/in movies?

A: In a lot of cases, you see the logo, so it wouldn’t be that hard to have a little (s) next to it, part of the way it’s depicted in the scene. Consumers would learn it has significance, in the same way that some brand names are blurred out and that practice teaches consumers about legal significance of appearance.

My Q: Seems to me that use and recognition might not covary as she says in (1), which would have consequences for the regime. And I’m not sure how you’d have the current standard and the availability of the (s)—sounds like she thinks the existence of the (s) would change the standard, and wouldn’t it do so even if the plaintiff did use the (s), because the plaintiff would be challenging a use without the (s) and thus have a decreased likelihood of confusion?

A: Yes, she assumes consumers will learn, and that’s key to the regime. And she agrees that it would be harder for everyone to win. Consumers can learn to see the absence of a symbol as an absence of sponsorship, and that would lower the likelihood of success for people who use the (s).

Ramsey: Oral use? Think about how it works with things like trade dress, etc.

A: Needs to be worked out. (Audio disclosure at beginning/end?) That’s advertiser’s job, if they want the protection.

Q: Why not mark when it’s not sponsored and disclaim?

A: That’s kind of the law now. The burden is on the user to show nonconfusion, and that’s hard in context. There’s no easy language to disclaim sponsorship either. Federal Expresso coffeeshop—not claiming affiliation. Because names are seen as indicators of sponsorship, no easy way to say “we’re not affiliated,” and disclosures/disclaimers often don’t work.

Q: Wouldn’t it be burdensome to have all the logos around sports arenas add an (s)?

A: Not a big one, and if they find it burdensome they don’t have to do so.

Stauffer: proposed EU directive: product placement would have to be generically disclosed with a black dot on the screen. Could have a tone for radio.

Why are you confident this won’t be coopted by the antibrands as well? The Yes Men will use the (s).

A: Then they’re likely to be confusing!

Errol Meidinger, University at Buffalo Law School, State University of New York: Branding Corporate Responsibility with Marks of Rectitude

Margaret Chon just published a piece on a similar topic; he’s been talking with her. Home Depot execs came to their building and found a giant banner: Stop selling old growth wood! Using their logo, too. Home Depot protesters brought inflatable chainsaw to the parking lot: thanks for helping Home Depot destroy the world’s old-growth forests! Homedepotsucks.com.

What should Home Depot do? Initially resisted getting certified, but ultimately adopted. Contestive branding: not just about the image, but the image is part of the regulatory regime being crafted.

Forest Stewardship Council: an example of corporate responsibility institutions. A set of institutions using activist targeting of corporate brands; standard setting institutions that accredit certifying organizations and attempt to balance North/South power. They certify products and communications. FSC is legislative: the certification requires meeting various criteria on biodiversity, monitoring, indigenous rights, and so on—looks like what a state would do except that no single state institution would be able to do it.

When Home Depot agreed, the Rainforest Action Network ran an ad in the NYT thanking them. Used the ad to attack other companies that weren’t yet on board.

Pour encourager les autres: After the Network turned its attention from Home Depot, it took only two phone calls to get Lowe’s to follow.

This triggers the growth of private competing certifiers—SFI, a US industry initiative for alternative “green” certification. Eventually, the industry coalitions turned into a worldwide certifier, PEFC. It’s a tournament of competing programs. He posits this has reshaped the forest policy arena, in which environmentalist- and industry-driven certifiers are setting global policies and NGOs and states are at the periphery.

Similar things have been happening in many other sectors: fisheries, organic, coffee, apparel, tourism, carbon footprints, energy efficiency, animal welfare, flowers, etc. He showed some beautifully designed logos, of which the Marine Stewardship Council was my favorite because of the checkmark integrated into the fish.

IP strategies: the Marine Stewardship Council tries to ensure the fish comes from a sustainably managed fishery. The policy is implemented through the logo, and shaped by the fact that it’s implemented through the logo. It’s registered as a mark in Australia, Canada, EU, Iceland, New Zealand, Sweden, Switzerland, and UK. Does license agreement with each user for a fee. On-product requirements: logo, plus statement about meeting MSC’s standards, plus chain of custody code number. MSC owns the mark but delegates the power to license it to MSC Int’l for IP purposes. Main concern is use by nonlicensed organizations. MSC denied its permission for use on a brochure attacking farm/acquaculture salmon—wants to give gold stars, not black marks. They are looking into a standard for aquaculture. They don’t allow any association with campaigns, and demand permission required for all uses.

He sees strategic concerns about future use of the mark ending up being a not insignificant inhibition on public speech.

Another element: certification wars: “Don’t Buy SFI: Certified Deception by the Same-old Forest Industry.” One entity filed a complaint with the FTC: there’s no chain of custody; it’s an industry organization; standards are deceptive because they’re not ecologically protective but littered with holes. (Reminds me of the “Animal Care Certified” case.) SFI filed a countercomplaint against the Forest Stewardship Council and the US Green Buildings Council arguing there’s too much variation in the standards—sometimes it can be certified if it’s grown in a lenient environment but not in another one; imperfect auditing/chain of custody. Also making antitrust allegations because the USGBC has accepted the FSC as the sole source for certification for LEED green building standards, with allegedly anticompetive effects. FSC also doesn’t admit prospective members who publicly criticize the FSC.

And then there’s the question of the WTO: the WTO gives presumptive validity to relevant international standards. One hope is that these standards will be absorbed into international law this way.

Competition for moral authority. NGO v. industry fight, NGO v. corporate. IKEA prides itself on buying FSC wood, but doesn’t display any FSC labels in its stores. Is that a problem, for IKEA to take green branding and put it into the IKEA label?

Ramsey: This is an area where nonprofits are offering up information—complicates the commercial speech account/claims that TM law should be limited to commercial speech. You mentioned prohibitions on use by campaigns: is that justified?

A: He doesn’t make a commercial/political distinction. Maybe it’s not so bad to disallow the use of the logo in political debate—you can talk about the Council.

Ramsey: but bloggers often like to have the logo there (heh).

A: If the point is to brand rectitude, to withhold the use of the logo is a little odd—making moral claims signified in a special way. (Well, just because they say that’s their rule, doesn’t mean it’s the rule by which bloggers must govern themselves, as I believe I am demonstrating—the moral claims are made by application to fish, not to blogs.)

Rebecca Tushnet, Georgetown Law: Ad Creep, Astroturf, and Other Challenges of the Attention Economy

I have a very rough draft in which I talk about my own crackpot theories of why ads are colonizing every aspect of existence. The bulk of the paper is given over to defending advertising law’s ability to follow ads where they go, even in the realm of user-generated content, subject not to First Amendment constraints but to §230. I argue that §230 does in fact prohibit liability for advertiser adoption of pure user-generated content, creating a possibility of regulatory arbitrage (since the user will be subject only to state-law defamation standards while the advertiser who “said” the same things would be subject to strict liability under the Lanham Act), and that we might want to think about revising §230 in this specific circumstance, though opening up §230 might be so risky that it’s not worth the cost. By contrast, I argue, contrary to Eric Goldman and in agreement with Paul Alan Levy, that the FTC’s new endorsement guidelines are fully compatible with §230, since they make the advertiser responsible for a blogger’s content based not on the provision of internet access (nor even associated payment/consideration for same) but on other agency principles.

Questions focused on what counts for the FTC’s purposes—not a free book, but a free Playstation—and aforesaid crackpot theories of how ads, like porn and protest, are in a dynamic of getting more and more extreme so as to overcome our exhaustion and ennui.