Friday, July 19, 2019

Lysol can't quickly clean up this mess of anti-Clorox ads

Clorox Co. v. Reckitt Benckiser Gp. PLC, 2019 WL 3068322, No. 19-cv-01452-EMC (N.D. Cal. Jul. 12, 2019)

Clorox sued Reckitt for false advertising of its Lysol cleaning products under the Lanham Act and California’s UCL/FAL. The court found tiny parts of the complaint insufficiently pled but kept most of the claims intact.

First, the court determined that Rule 9(b) applied to the false advertising claim, as to the California claims, because intentional misrepresentation was alleged.

Next, Reckitt argued that all Clorox’s misleadingness claims failed because the complaint didn’t allege specific facts proving significant consumer deception.  But proof is for summary judgment/trial, not the pleadings. “[A]t the pleading stage the plaintiff need only allege specific misleading statements and explain why they are misleading in accordance with Rule 9(b).”

On to specifics: Ads: (1)“Bleach Indicator Test” depicting a side-by-side test of Clorox Clean Up and Lysol Daily Cleanser.. One cutting board is first cleaned with Clorox Clean Up and another cleaned with Lysol Daily Cleanser, then an apple slice is placed on each cutting board; the former turns brown and the latter doesn’t. Four individuals labeled as “real people” negatively react to the browned apple, e.g., “I wouldn’t even want to touch this.” Voice-over: “unlike Clorox Clean Up, Lysol Daily Cleanser has only three simple ingredients and leaves surfaces free from harsh chemical residue.” A similar ad used a “bleach indicator test” instead.

This was sufficiently alleged to be false for three reasons: (1) It plausibly necessarily implied that Clorox Clean Up was unsuitable or unsafe for use in kitchens, even though the EPA has in fact deemed the product safe for kitchen use. “[A]n advertisement can be literally false where it necessarily implies to customers that a product is unsafe to consume.” Given the images of food turned brown and people recoiling in disgust, it was reasonable to infer falsity by necessary implication. 

(2) It plausibly falsely indicated that the result of the test depicted—that Lysol Daily Cleanser apparently leaves no residue that would affect an apple slice—was inaccurate. According to Clorox, Lysol Daily Cleanser contains hypochlorous acid, “a form of bleach,” so a “properly chosen and calibrated ‘bleach indicator test’ would detect the residue left by Reckitt’s product,” but Reckitt’s chosen test was deliberately designed not to do so. A “tests prove” claim is false where the tests “are not sufficiently reliable to permit one to conclude with reasonable certainty that they established the claim made.” Clorox plausibly alleged that Reckitt’s test is not sufficiently reliable to permit one to conclude with reasonable certainty that Lysol Daily Cleanser leaves no bleach residue.  (Note the elision of the key question here: what exactly does the ad claim that “tests prove”?  The ad doesn’t explicitly say there’s bleach on one slice and not on the other.)

(3) It was literally false because it necessarily implied that Lysol Daily Cleanser and Clorox Clean Up are comparable products but that the Clorox product leaves a harsh residue. But they have different formulations and purposes. “Although usable in kitchens, Clorox Clean Up is a heavier duty product than Lysol Daily Cleanser” with stronger ingredients not for everyday use, so attempts to “position Lysol Daily Cleanser as a ‘simpler’ or ‘less harsh’ substitute for CCU misrepresents the facts and deceives customers.”  [Gotta admit, less harsh sounds true but also necessarily misleadingly comparative, given the background expectation that a thing will be compared with a thing in its class, e.g., what are the harshest daily cleaners? Clorox alleged that it did market an actually competing product, which makes the misinterpretation even more likely.] LDC couldn’t be substituted for CCU because the former was much more diluted and lacked a surfactant/detergent for intensive cleaning. [This seems eminently correct as to indications but possibly not descriptively true—I might want to know how many consumers are unaware of the difference.] Apples-to-oranges comparisons can be literally false by necessary implication “where non-comparable products are portrayed as otherwise equivalent (except for the superior or inferior aspect being illustrated in the advertisement).” Thus, “plaintiffs can establish literal falsity under the Lanham Act by alleging that two products portrayed as comparable in an advertisement are not actually comparable – that the advertisement omits differences which would have been material to recipients.”  Clorox sufficiently alleged lack of comparability, especially given the availability of a closer Clorox comparator.

Reckitt argued that Clorox Clean Up was in fact comparable to Lysol Daily Cleanser because the EPA has approved the Clorox Clean Up as “Gentle & Powerful Enough for Daily Use,” but there was no authority holding that the EPA’s categorization of products makes those products comparable for Lanham Act purposes. Anyway, this was a factual dispute.

(2) “Spray Away” ads showing an image of Clorox Clean Up and its list of ingredients being sprayed and wiped away with Lysol Daily Cleanser. The ad then flashes banners stating: “Lysol Daily Cleanser[;] Only 3 ingredients,” “Kills 99.9% of germs (when used as directed),” and “No harsh chemical residue.” A Facebook ad juxtaposes the safety warnings of Clorox Clean Up and Lysol Daily Cleanser and concludes with the same statement.  Also sufficiently alleged falsity.

First, this was another plausibly apples-to-oranges comparison of LDC with CCU. Second, the Facebook ad juxtaposing the safety warnings was plausibly likely to mislead consumers because it suggests that the Lysol product is “simpler” and has “fewer safety risks” and “impugns the safety of CCU by suggesting that consumers should shy away from products that require cautionary disclosures.” Clorox also alleged that the active ingredient in the Lysol product, hypochlorous acid, was no gentler than the bleach in Clorox Clean Up.

 (3) “Fake It,” “Game Over,” and “Spray Away” ads: the first said that Lysol could “help protect” children from colds then showed a split screen with the Clorox and Lysol symbols on either side, and “germs” on both sides of the screen; the germs on the Lysol side of the screen disappear while those on the Clorox side remain. An announcer states, “Lysol Disinfectant Spray kills the number one cause of the cold and Clorox Wipes don’t.” Reckitt eventually added in “minuscule type” that the comparison was between Clorox Disinfecting Wipes and Lysol Disinfecting Spray. Two other ads had the same factual claims, presented differently (in one, a container of Clorox Disinfecting Wipes is sprayed off the screen by Lysol Spray).

Plausibly alleged to be literally false by necessary implication. Clorox again alleged an apples-to-oranges comparison of Clorox’s “pre-moistened, anti-microbial wipes that consumers use to clean and disinfect surfaces throughout the home” to a Lysol product in a different category with different usage instructions. Wipes are allegedly formulated to target different germs than disinfecting sprays. “The key difference is between wipes and sprays, not Clorox and Lysol.” Again, Clorox alleged that more comparable products were available—both parties’ sprays (neither of which are approved for rhinovirus) or their wipes (both of which are).

(4) “Strength Test” ads: Employed to pick up kettlebell weights, the Clorox wipe rips immediately while the Lysol wipes holds the weight for several seconds. A disclaimer states, “Dramatization. Based on lab results. Supervised demonstration. Do not attempt.” A Facebook ad said, “Lysol wipes are stronger than competition” and depicted a Lysol wipe intact next to a torn “competitive” wipe that “appears below a version of the famous Clorox chevron, in which the Clorox brand name has been replaced with the words ‘vs leading competitor.’ ”

Clorox’s claim that the ad falsely insinuates that Lysol wipes are “of higher quality, are more durable, and are more resistant to tearing during use than Clorox wipes” failed because Clorox didn’t allege that Lysol wipes are not of higher quality, more durable, and more resistant to tearing than Clorox. [Technically, it should be enough to allege that they are of equal quality—one might even hedge and allege that Clorox wipes are of at least equal quality to Lysol’s.] It wasn’t enough to allege that the Clorox wipes are unlikely to rip under typical use conditions.

However, the court was more receptive to the allegation that the “strength test” conducted in the advertisement was unreliable and therefore literally false. Clorox alleged that the test “is not capable of replication, because neither product is actually long enough to serve as a ‘handle’ when holding a kettlebell weight,” and thus it wasn’t “sufficiently reliable to permit one to conclude with reasonable certainty that [it] established the claim made.” This was more than the insufficient allegation “there was simply no way a valid test would show” results supporting an advertising claim.

(5) “No Scrubbing” ads: for example, a woman squirts Lysol Power Toilet Bowl Cleaner into a toilet and flushes; the toilet bowl instantly becomes clean without the use of a brush. The ad then depicts the unfavorable results of a side-by-side efficacy test comparing Lysol Power Toilet Bowl Cleaner and Clorox Regular Liquid Bleach. A small disclaimer states, “Cleaning Power vs. Clorox Regular Liquid Bleach on limescale and rust stains. Results after contact with product for 10 minutes, followed by rinsing.” A small banner states, “10X more cleaning power than Clorox.”  In another similar ad, a tiny disclaimer states that the results involve Lysol Power Toilet Bowl Cleaner and Clorox Toilet Bowl Cleaner Clinging Bleach Gel, but that’s is a distinct product from the “Clorox Toilet Bowl Cleaner with Bleach” product referenced in the voice-over.  The “10X More Cleaning Power than Bleach” claim also appears in stores.

Clorox pled falsity: (1) Again, apples-to-oranges: Clorox alleged that the formulation and purpose of the two products differ.  [I just love what you learn about the world in false advertising cases.]  Acid-based cleaners, such as Lysol Power Toilet Bowl Cleaner, are more effective against mineral-based deposits (like rust and limescale), whereas Clorox’s bleach-based Clorox Toilet Bowl Cleaner is more effective at removing organic stains (like mildew and mold). Again, the parties allegedly marketed more directly competing products. Again, Reckitt’s argument that the EPA approved both for use on mineral-based deposits wasn’t sufficient.

Reckitt argued that it added a “disclaimer expressly stat[ing] that the purported advantage relates only to rust and limescale and not to organic matter.” Clorox alleged that the disclaimers were “nearly illegible” and “minuscule” and thus plausibly useless.

Second, Clorox plausibly alleged misleadingness: “[t]hrough its depiction of a woman using a single squirt of the product to render her toilet immaculate,...[the ad indicates that] the Lysol product works instantly to removal all deposits from toilets.” First, “the product requires at least 10 minutes to remove some toilet bowl stains,” and second, it’s not [as] good at organic deposits, so it “cannot render toilets immaculate in seconds, as depicted in Reckitt’s advertising.”

Third, Clorox plausibly alleged that the claim “10X more cleaning power than Clorox” was literally false because Lysol products generally do not have a performance advantage over Clorox. Compared apples to apples, the Lysol product allegedly didn’t have an advantage in removing limescale and rust because they contain the same active ingredient, glycolic acid. So too with the claim “10X More Cleaning Power than Bleach,” because a reasonable consumer would likely believe that the comparison is in fact with its leading competitor, Clorox.

(6) Noncomparative ads: claims included: kills “99.9% of germs”; “kills over 100 illnesses causing germs” and a disclaimer that “Lysol Disinfecting Spray kills germs on hard surfaces when used as directed.”  One ad was ambiguous as to whether the ad was claiming that both touted products could
kill over 100 illness-causing germs, or whether they can do so together.  But it was plausible that it misleadingly implied that either product alone would be enough. Another ad claimed that Lysol Daily Cleanser kills “99.9% of germs” and the visual showed an actor spraying and immediately wiping the surface, whereas the EPA-approved use instructions on the label requires ten minutes of contact with a treated surface to achieve its disinfectant effect. That was plausibly misleading, though it wasn’t a comparison to Clorox.

Materiality: the court quoted precedent that materiality “is ‘typically’ proven through consumer surveys,” even though that is empirically false (this is a problem for the courts of appeal; the court here notes but does not give a reason for the shift in judicial treatment of materiality over time, from presuming materiality for literal falsity—often a matter of common sense—to a seemingly rigid insistence on separate evidence no matter what). Restoring some flexibility, a plaintiff can also establish materiality by showing that “the defendants misrepresented an inherent quality or characteristic of the product.” Clorox plausibly alleged that Reckitt’s advertisements attack Clorox products’ efficacy, safety, and durability. These were inherent qualities and characteristics of cleaning products, and would likely influence the purchasing decision of a consumer.

Injury: Clorox alleged direct diversion of sales and damage to its goodwill.  Reckitt argued that wasn’t enough, but for a competitor who’d been directly attacked it definitely was. In Vincent v. Utah Plastic Surgery Soc., 621 F. App’x 546 (10th Cir. 2015), the Tenth Circuit required the plaintiff to make some showing of “how much Plaintiff’s profits have decreased since Defendants began their advertising campaign,” to “quantif[y] or estimate the decreased in goodwill,” or to “quantify the number of potential customers who allegedly have been lost because of Defendant’s statements,” but the court wasn’t going to do that in the Ninth Circuit.

UCL/FAL: Reckitt argued that Clorox needed to allege its own reliance on Reckitt’s ads to state UCL/FAL claims.  That’s true for the FAL and UCL “unlawful,” but not for UCL “unfair.”  Also, whether the ads would deceive reasonable consumers was a fact question, as it usually is.

Monster Energy can't show falsity, can show tortious interference (but not irreparable injury)

The court denied a preliminary injunction on Monster’s claim that Vital falsely advertised its “Super Creatine” as “creatine,” because there wasn’t sufficient evidence that the two were distinct (in meaning or more significantly in function), and on tortious interference for lack of a showing of irreparable harm.

Monster argued that using “creatine” on the labels of Vital’s energy drink BANG (as part of a promise of “Creatine, Caffeine, CoQ10, & BCAAs”) was literally false, because BANG does not contain creatine, and that “Super Creatine” was impliedly false because BANG actually contains an entirely different molecule, creatyl-L-leucine.

Monster’s expert tested four cans and found no creatine, or only trace amounts, in the products.  Vital’s principal also made public statements that the products didn’t contain “regular creatine.” The expert’s declaration stated that “Super Creatine,” Vital’s term for creatyl-L-leucine, was a “chemically synthesized compound” which is created from the linking of two distinct amino acids, creatine and leucine, with properties “distinct from [those of] the constituent amino acids.”

The court found Monster’s definition of “creatine” to be “overly narrow,” since its own cited evidence established that “[m]any forms of creatine exist in the marketplace,” including formulations combining creatine with other amino acids. Monster also argued that creatyl-L-leucine does not provide any of the benefits of creatine, but its evidence did not establish that fact; it supported “only the more general conclusion that the chemical and physical properties of a new substance formed from the combination of two amino acids differs from their constituent parts.” Monster’s cited scientific journal articles didn’t discuss creatyl-L-leucine, “but both state that at least some of common creatine compounds provide similar or superior effects to those provided by creatine alone.”  The burden of proof was on Monster, and it did not meet that burden.

Monster also argued that BANG doesn’t contain enough Super Creatine to provide the benefits that VPX promises consumers. But testing only four cans of a single flavor of BANG purchased from the same retail store in Southern California wasn’t enough to conclude that the results were representative of BANG products generally. “The Court will not order a nationwide recall on the basis of such a limited sample size.”

Intentional interference with contractual relations: Monster showed that it had valid contracts for specific amounts of in-store shelf and/or cooler space with retailers across the country, and also showed that Vital disrupted those agreements, leading to BANG products being stocked in Monster’s contracted-for retail shelving space in retail locations across the country.  This wasn’t a matter of “wayward cans,” given the scope and duration of the problem and documented statements from Vital’s leadership. One former Vital account manager declared that district and regional managers directed him to use “guerilla tactics” and “be aggressive” in order to obtain shelf and cooler space for BANG at retail stores. “They made specific demands about the location in which BANG should be placed … and instructed [him] to document the space he acquired with photographs,” though he knew that competitors had contracts with retailers about placement. There was also evidence that placement contracts were industry custom, and that Vital’s principal made Instagram posts showing intent, such as one that shows an image of a VPX products placed in front of Monster products accompanied by the text “When in doubt block them out. In life when they tell you there’s no shelf space – make your own shelf space! When multibillion-dollar competitors pay for space retaliate with a vengeance,” and another which states that VPX will “not placate to [a retailer’s] tomfoolery and rigged system of paying for shelf space.”

Defendants argued that Monster failed to show that Vital or its agents had knowledge of the existence of the specific shelving-space contracts between Monster and its retailers with which they interfered. The evidence was to the contrary, including public statements made by the principal which object to the industry practice of paying for shelf space and state his intention to “make [his] own shelf space” and “retaliate” against competitors who pay for space. [Ah, clients and their social media.] “In light of the extent of this practice, Plaintiff need not demonstrate specific knowledge of each contract in question in order to demonstrate a likelihood of success on the merits.”  Likewise, the evidence showed that the “wayward” cans in Monster’s shelving space were likely the result of defendants’ intentional conduct.

But there was very little evidence of damages. While it may be the case that BANG’s sales growth has been “fueled by VPX’s theft of Monster’s shelf and cooler space,” Monster didn’t have good evidence backing this up.  There was no evidence connecting contractual interference with “the loss of customer goodwill, the loss of market share, or the loss of profits,” making harm speculative at this stage. So too with irreparable harm.

TrueCar's false claims not subject to car dealers' challenge without evidence of injury

Dependable Sales & Service, Inc. v. Truecar, Inc., 2019 WL 3067115, No. 15-cv-1742 (PKC) (S.D.N.Y. Jul. 12, 2019)

Lots of prior rulings about various aspects of this false advertising claim in the general field of auto sales.

TrueCar successfully moved for partial reconsideration of the previous summary judgment decision holding that the plaintiffs, 108 automobile dealerships, failed to come forward with evidence of economic or reputational injury from TrueCar’s advertising. They weren’t direct competitors and TrueCar’s false advertisements did not make comparative claims, so plaintiffs weren’t entitled to a presumption of injury. Now, the court concluded that without evidence of some injury, disgorgement wasn’t available as a remedy, despite evidence of TrueCar’s willful Lanham Act violations.

There is authority in the Second Circuit for allowing disgorgement to advance the interest of deterrence, even where the plaintiff has not demonstrated injury. However, that line of cases “arose in the trademark and trade dress arena.” It’s worth noting that the statutory language doesn’t make this distinction; Lexmark is an interpretation of the overall structure of Lanham Act liability.  “More recently, in a false-advertising case, the Second Circuit observed that ‘[o]ur precedent permits a district court to award a defendant’s full profits based solely on deterrence,’” Merck Eprova AG v. Gnosis S.p.A., 760 F.3d 247, 262 (2d Cir. 2014), but that was a case of obvious direct competition.

Unfortunately, the court went on to say that “[t]here are important differences in the elements of a trademark claim and a false advertising claim, as well as in the interests at stake…. In a trademark case, a plaintiff may be harmed by virtue of losing exclusive control over its own mark…. The unauthorized use of a mark ‘invariably threatens injury to the economic value of the goodwill and reputation” associated with the mark.’ … Even where a trademark plaintiff cannot point to lost sales, it may still be harmed by ‘a loss of control ... over how the public perceives’ its goods or services.” 

As readers of this post likely know, this assumption is both (1) unsound in a number of situations, particularly outside classic source confusion, and (2) increasingly vulnerable post-eBay and Winter, now that courts have given renewed attention to plaintiffs’ harm stories.  Of course, it would be reasonable to respond that even without a presumption of irreparable harm, a presumption of harm is still justified … but the lost control language is very tightly tied to the discourse of irreparability.  As a few cases (mostly in the 9th Circuit) have recognized, the existence of some unquantified risk of harm (“may still be harmed”) is not the same thing as the likelihood that the harm will materialize.  Perhaps the standard for disgorgement ought to cover any [reasonable? theoretical? plausible? nontrivial? you see the difficulty] risk whether likely or not—I suspect that the level of willfulness might interact with one’s opinion on the matter.

Anyway, in a trademark case, the Second Circuit concluded that the district court should have ordered disgorgement solely to deter willful infringement because the “deliberate[ ] and fraudulent[ ]” infringement of plaintiff’s mark warranted disgorgement of defendant’s full profits, even though plaintiff did not demonstrate lost sales, consumer confusion or damage to good will. Defendant’s “callous disregard for the rights of a competitor” was sufficient. [As opposed to callous disregard for consumers and competition in general through willful false advertising—note too that trademark infringement in a non-sales substitution case wouldn’t come under this reasoning if we take “competitor” seriously, which would actually be a decent way of reconciling the cases.] But a false advertising plaintiff must show injury “by way of lost sales or damage to business reputation,” though a presumption of injury can arise from falsity about a direct competitor.  [The real problem here is that the trademark cases cited largely predate, and definitely have have not actually analyzed, Lexmark and its statutory interpretation. I’m not saying the court is definitely wrong in its holding here—though I might have gone the other way—but I am saying that the efforts to distinguish trademark law are not successful.]

Ultimately, the court reasoned, “[i]t may seem anomalous that a false advertiser in need of deterrence can escape the disgorgement of profits when there are plaintiffs that are eager and willing to pursue that remedy,” but the plaintiffs here didn’t show injury. A differently situated plaintiff [would it have to be an entity that used TrueCar’s business model of contracting with different dealers and offering discounts? It’s not obvious why the false advertising here would’ve hurt them either] might have damages that are small or difficult to establish, and for them disgorgement would be an ideal remedy. Consumer class actions [unless consumers signed arbitration agreements] or the FTC/state AGs are other options for punishing willful false advertising.  

The court declined to exercise supplemental jurisdiction over the remaining state law claims.

Reading list: the dark side of meme culture

I realized that many of the young reporters who initially helped amplify the white nationalist “alt right” by pointing and laughing at them, had all come up in and around internet culture-type circles. They may not have been trolls themselves, but their familiarity with trolling subculture, and experience with precisely the kind of discordant swirl featured in the aforementioned early-2000s image dump, perfectly prepped them for pro-Trump shitposting. They knew what this was. This was just trolls being trolls. This was just 4chan being 4chan. This was just the internet. Those Swastikas didn’t mean anything. They recognized the clothes the wolf was wearing, I argued, and so they didn’t recognize the wolf.
This was how the wolf operated: by exploiting the fact that so many (white) people have been trained not to take the things that happen on the internet very seriously. They operated by laundering hate into the mainstream through “ironically” racist memes, then using all that laughter as a radicalization and recruitment tool.

Thursday, July 11, 2019

Imaginative makeup color names as descriptive fair use

Hard Candy, LLC v. Anastasia Beverly Hills, Inc., 921 F.3d 1343 (11th Cir. 2019)

Hard Candy sued Anastasia, a competitor in the cosmetics industry, for infringement based on one of Anastasia’s “Glow Kits,” flip-open makeup palettes containing four different shades of facial highlighter.  The Gleam Glow Kit included the words “hard candy” in capital letters on the back and inside of the package to designate a peach-pink shade of makeup. “Anastasia says that it chose this name because the product had a ‘shimmer’ that reminded the developer of candies that her grandmother gave her when she was young.” “Hard candy” appeared in marketing materials and social media posts, as well as on the product itself.
front and back of kit


Hard Candy sought “every remedy permitted by the Lanham Act besides actual damages -- that is, it asked for an injunction to prevent future infringement, an accounting and the disgorgement of profits that the defendant made from the allegedly infringing goods, and declaratory relief, along with fees and costs.” The district court denied it a jury trial, held a bench trial, and found no likely confusion/descriptive fair use, and the court of appeals affirmed.

There was no Seventh Amendment right to a jury trial.  Hard Candy argued that it should have gotten a jury because it was seeking to recover Anastasia’s profits as a “proxy” for the damages it suffered due to infringement. Doesn’t matter; disgorgement is still an equitable remedy and doesn’t entitle the claimant to a jury trial.

The standard of review of the confusion finding after a bench trial was for clear error.  Factors that favored Hard Candy: product similarity, trade channel/customer/advertising similarity, and strength of mark (arbitrary and therefore strong). Note that “Hard Candy” was arbitrary for cosmetics generally—or at least, could be used in arbitrary fashion.  Factors that favored Anastasia: dissimilarity of marks/uses and lack of evidence of actual confusion.  The court also found no intent to infringe.

Hard Candy argued that the district court erred by finding that the “similarity of mark” factor favored Anastasia, since the words were identical. But it’s the overall impression that matters, and “formally” identical use of the same letters wasn’t enough. The manner of use matters too.  In the Gleam Glow Kit, the words HARD CANDY” appeared only on the back and inside of the kit, “nam[ing] one shade of a product that is otherwise clearly and prominently featured as an Anastasia Beverly Hills mark.”  It was not clear error to weigh descriptive use, or lack of use as a mark, as favoring Anastasia.  [Despite what the Supreme Court said, this case highlights—no pun intended—that descriptive use just doesn’t work any more as a separate defense.  On the flip side, even if we’ve abandoned use as a mark as an independent limit on infringement, that doesn’t mean it’s not relevant to whether confusion is likely.]  Mark placement, size, and use to identify a specific part of the product are definitely relevant to a similarity analysis.

Finally also argued that the district court placed too much weight on the lack of a showing of actual confusion. But, “[i]f consumers have been exposed to two allegedly similar trademarks in the marketplace for an adequate period of time and no actual confusion is detected either by survey or in actual reported instances of confusion, that can be powerful indication that the junior trademark does not cause a meaningful likelihood of confusion.” Here, Anastasia sold almost 250,000 Gleam Glow Kits; the product was touted on social media accounts with millions of followers.  “In this case, there was enough opportunity for confusion to make the absence of any evidence significant. … [H]undreds of thousands of cosmetics consumers purchased the allegedly infringing kit, several times more likely saw it on store shelves or online, and still there is no evidence that anyone, anywhere, was ever confused about whether Anastasia or Hard Candy was responsible for the product.” At a minimum, there was no clear error in weighing this in favor of Anastasia.

Anyway, Anastasia established descriptive fair use. The court of appeals endorsed the view that “it is not necessary that a descriptive term depict the [product] itself, but only that the term refer to a characteristic of the [product].”  And here’s an interesting bit that the court of appeals doesn’t even seem to notice: the use of Hard Candy as the name of the cosmetics brand was arbitrary, but the use of “hard candy” as the name of a specific color was descriptive.  I think this is the right result (though I might’ve called the brand name suggestive), but this case is pretty good evidence that manner of use matters a lot to whether something seems “inherently” to be a trademark. See also Thomas R. Lee, Eric D. DeRosia & Glenn L. Christensen, An Empirical and Consumer Psychology Analysis of Trademark Distinctiveness, 41 ARIZ. ST. L.J. 1033 (2009). 

Hard Candy argued that Anastasia’s use wasn’t descriptive, and that the district court disregarded “the central requirement” of consumer perception of the use.  But the district court did consider how a hypothetical consumer would perceive the use, considering “the overall context; the lettering, the type, the style, the size, [and] the visual placement” of the words on the product.  Anastasia also introduced evidence that cosmetics companies regularly describe shades with words that are not literal color descriptions, like the other three shades in the kit named “Starburst,” “Mimosa,” and “Crushed Pearl.”  This common practice was relevant to reasonable consumers’ perceptions of the use, not just to Anastasia’s subjective intent.  And no evidence of actual consumer perception is required to use the defense.

When © trips courts up: Lack of access to standards makes false ad claim impossible to resolve

Wing Enterprises, Inc. v. Tricam Industries, Inc., 2019 WL 2994465, No. 17-cv-1769 (ECT/ECW) (D. Minn. Jul. 9, 2019)

This false advertising case about multi-position ladders turns out to involve an important copyright issue that the Supreme Court has taken up: when a standard is incorporated into law, should it be readily accessible to the public? Because an older ANSI standard incorporated into OSHA isn’t available to the court, the court can’t resolve whether failure to meet the current ANSI standard also violates OSHA. This matters because the defendant advertised ANSI/OSHA compliance, but plaintiff’s evidence went to whether there was compliance with current ANSI.

The thrust of the federal and state false advertising claims claims is that Tricam represented that its Gorilla Ladders comply with ANSI ASC A14.2, a voluntary industry standard for portable metal ladders that was developed by the American Ladder Institute, but in fact the rungs of its ladders are not sufficiently deep all the way across to satisfy that standard as Wing understands it.  The label affixed to each ladder has an oval icon that says: “manufacturer certifies conformance to OSHA ANSI A14.2 code for metal ladders” and there were similar representations elsewhere.

The court admitted one expert on ANSI conformance and excluded an expert report on materiality, the latter because of the access-to-code problem.  Wing’s survey expert, Hal Poret, conducted materiality surveys: a labeling survey, intended to measure consumer reaction to the allegedly false statement on the label, and an importance survey, intended to assess the importance to consumers of compliance with industry safety standards in general. First, the court rejected the argument that the labeling survey was unreliable because it failed to replicate market conditions; it highlighted the label and demanded consumers spend a certain amount of time looking at it, which might not happen on the retail floor.  “These might be fair points if the survey had been intended to test what message the statements conveyed (as relevant to the falsity element), or whether consumer confusion existed in a trademark case.” But for a materiality study, how the consumers would see the image in the store didn’t affect its relevance, although a jury could weigh divergence from the retail context in its considerations.

The real problem is that Poret’s surveys tested ANSI and OSHA conformance together by eliminating the entire challenged label (and his importance survey referred only to conformance to unspecified “industry safety standards,” not specifically to ANSI); Wing didn’t show that either a combined OSHA/ANSI statement or industry safety standards writ large was relevant to the issues a jury would need to decide in this litigation.

Wing argued that OSHA uses ANSI standards, so that a violation of ANSI is necessarily a violation of OSHA.  And here’s where the access part comes in.  OSHA regulations provide that mandatory provisions (“shall” provisions) of standards incorporated by reference are adopted as mandatory under OSHA and “have the same force and effect” whether they are issued by federal agencies or by nongovernmental organizations. Here, OSHA regulations incorporate by reference “ANSI A14.2-56 Safety Code for Portable Metal Ladders, Supplemented by ANSI A14.2a-77.”

Those last two digits are apparently pre-2000 year codes.  Wing didn’t identify how -56 and -77 differed/overlapped with the 2007 version of ANSI A14.2 the parties were apparently working from in this case.  OSHA regulations, in fact, incorporate by reference different versions of ANSI A14.2. “For example, one regulation that pertains to the construction industry incorporates the 1982 version; another, pertaining to shipyards, incorporates the 1972 version; and others, relating to marine terminals and longshoring, incorporate the 1990 version.”  Wing didn’t confirm whether there was any relevant variation, and “[e]ven if the Court were inclined to do that legwork on Wing’s behalf, the Court cannot independently verify the extent to which the 1956 version explicitly mentioned in the regulations overlaps, if at all, with the 2007 version before the Court by referencing publicly available sources because the ANSI standards are not reproduced in the Code of Federal Regulations and are instead behind a paywall or available for in-person review in another state.” [Apparently the “state” is DC, at the National Archives.]  Testimony from Tricam witnesses was not sufficient to reach the legal conclusion that failure to meet the current ANSI standard, in the manner identified, would also be failure to meet the older standards that actually have the force of law.

Because Poret’s label survey was premised on the assumption that ANSI falsity meant OSHA falsity, it couldn’t test ANSI falsity alone and was not admissible. Likewise, his importance survey tested “[c]ompliance with industry safety standards” in general, but given the multiple sources of industry safety standards and the evidence in this case, that wasn’t relevant—“What happens if, as contemplated above, a ladder that fails to conform to the 2007 version of ANSI nonetheless does meet the requirements of one or more OSHA regulations that incorporate an older version of that standard?”

Without the survey, Wing couldn’t show materiality and summary judgment was warranted. It argued that it could show materiality by showing that “the false or misleading statement relates to an ‘inherent quality or characteristic’ of the product,” and that “questions of safety and efficacy are likely to satisfy automatically the materiality prong.” But the Eighth Circuit has not endorsed the “inherent quality or characteristic” method of showing materiality. And here, without further evidence on ANSI variation, “the most Wing could show is a technical noncompliance with one of multiple potentially applicable safety standards. That is not a compelling context in which to adopt a new approach to showing materiality.”

Nor was the following adequate: (1) testimony from a high-level Wing executive that, in his opinion, compliance statements on Home Depot’s website are “important, otherwise, I don’t believe Home Depot would put it on the website,” (2) testimony by Tricam’s president that an ANSI-certification statement on Home Depot’s website “could be” helpful in differentiating Tricam’s products from hypothetical competing ladders that do not purport to conform to ANSI, and suggesting that an ANSI-certification statement on the product label might be something a professional might want for purposes of OSHA inspections of a job site, and (3) testimony from the chairman of the ANSI labeling committee that “[i]t’s possible” that an ANSI-compliance statement would help a consumer choose a ladder. But these statements were all speculative on their face, and in each case the testimony was qualified by reference to the speaker’s lack of direct knowledge about consumer behavior.

Wednesday, July 10, 2019

Competitor's false advertising case against MLM income claims can proceed

Youngevity Int’l v. Smith, No. 16-CV-704-BTM-JLB, 2019 WL 2918161 (S.D. Cal. Jul. 5, 2019)

This Lanham Act case involves, among other things, alleged misrepresentations relating to the MLM aspects of defendants’ business pitched to potential “affiliates.” 

First, defendants argued that plaintiffs failed to establish any damages, justifying summary judgment. But there was some evidence of decreased Youngevity sales during defendants’ false advertising and an expert willing to link that with defendants’ sales generated by ex-Youngevity distributors; also, “an inability to show actual damages does not alone preclude a recovery under section 1117 [of the Lanham Act],” and plaintiffs were also seeking injunctive relief. 

Second, defendants argued that they couldn’t be held liable for statements made by their distributors. There was sufficient evidence for a jury to find that these people (known as Wakaya Ambassadors) were agents for Lanham Act purposes, making defendants vicariously liable.  Their classification as independent contractors in Wakaya’s own Policies and Procedures wasn’t enough to avoid a material issue of fact.  Moreover, defendants didn’t dispute that the Ambassadors “engaged in the allegedly false advertising for the purpose of attracting distributors and increasing sales,” squarely within the scope of their roles. And defendants reserve the right to terminate Ambassadors’ accounts for unapproved conduct. That was enough to go to the jury.

The false advertising claims included allegedly false claims about the potential income of a typical Wakaya distributor, e.g., Wakaya Ambassadors can earn “$6,200 bucks residually in the next three to six months” and “[w]e plan to build several leaders to 10K a month in the next 6 months.” [Ugh. It’s so clear that the US lacks sufficient protections against pyramid schemes. We shouldn’t be relying on competitors to do this, especially not competitors that are themselves MLMs with their own dubious claims.]  One individual defendant (Vaughn) claimed, on behalf of the company, that “when—you get a thousand people joining [Wakaya] a day, that’s $85,000 in a day. If you wanna do it in a week, that’s $85,000 in a week. [Etc.]”  Smith, Wakaya’s owner, observed in deposition that the statement is “inaccurate” because “it claims that Wakaya Perfection pays income for the joining of new people, for new people joining the organization.” Smith also testified that he could not recall any distributor in Wakaya ever making $85,000 in a month and that Vaughn himself does not make $8,500 in a week based on his work as a Wakaya distributor. There was evidence that Vaughn repeatedly asserted that Wakaya Ambassadors could earn a million dollars in their first year. Smith himself stated, “I’m telling you right now you’re going to earn a lot of money....[T]he amount[s] of money you’re going to earn in this program...right now you won’t even be able to imagine. They’re almost incalculable.” There was evidence that other Wakaya distributors also made false statements: one claimed that he was making up to $1700 in one day, then testified that, in fact, he made less than $550 in any single month. Expert evidence showed that, in reality, about 1% of Wakaya’s active associates earn $1000 per month in commissions and less than 3% even earn $500 per month.  There was [at a minimum] a genuine issue of material fact on falsity.

Defendants argued that these claims didn’t relate to Lanham Act-covered “commercial activities,” but of course they did. As the Tenth Circuit has said, “[i]t is [ ] apparent, in the context of the Act’s broad purpose of proscribing unfair competition and the 1988 amendment of § 43(a), that Congress did not intend to narrowly limit the term ‘commercial activities,’ but rather intended to encompass those activities which do not solely involve the provision of services or the production of goods. Proctor & Gamble Co. v. Haugen, 222 F.3d 1262 (10th Cir. 2000). Here, “[a]ttracting distributors is at the core of Defendants’ business model and is a practice with a substantial impact affecting commerce.”

The income claims here were literally false, either on their face or by necessary implication; materiality and misleadingness could be presumed:

The claims are specific, conclusive assertions that a Wakaya distributor will make at least the income that in reality, only 1% of distributors make. Even Defendant Smith’s statement, while not stating a particular dollar value, implies that earning a large amount of income as a Wakaya distributor is an inevitability. Moreover, because the claims are highly specific and present the likelihood of earning unrealistic amounts of money as a foregone conclusion when becoming a Wakaya distributor, the income claims are far outside the scope of mere puffery or opinion.

The relevant consumers were “anyone who might be convinced to become a Wakaya Ambassador based on claims of earning potential above a certain threshold,” and the relevant purchasing decision was to become an Ambassador—which after all, is how defendants make their money.

However, Youngevity’s [rather chutzpadik] argument that Wakaya is an unlawful pyramid scheme was not separately actionable under the Lanham Act, even if operating a pyramid scheme fraud under federal antifraud law. False income claims alone didn’t show that the “rewards” or income that Wakaya Ambassadors received were unrelated to the sale of Wakaya products.

Another alleged falsehood involved the role of alleged billionaire/Fiji Water founder David Gilmour, who Wakaya advertising touted as the founder, owner, and CEO.  The ownership/CEO claims were concededly literally false, and the depositions indicated that Smith was the founder.  Yet “Defendants persist in claiming that Mr. Gilmour founded Wakaya Perfection,” including by claiming that he was an investor (though apparently not in the company itself, but in the island named Wakaya from which some Wakaya ingredients come). The court found that Wakaya’s [implausible] colloquial or symbolic use of “founder” “to refer to one who acts as a kind of figurehead of a venture or project” could be literally false, and there was also evidence that it was likely to mislead consumers, making summary judgment for defendants inappropriate.

However, Youngevity’s claim that Wakaya and its Ambassadors made false claims about the status of Youngevity’s finances failed for want of sufficient evidence; one email written as personal correspondence between associates wasn’t enough to be commercial advertising or promotion.  Two other emails didn’t themselves make negative statements “but rather discuss disparaging communication that the authors allegedly heard about or were on the receiving end of,” which also failed to show sufficient dissemination.

One social media post by “Dave and Barb Pitcock with Wakaya” “certainly disparages Youngevity” but didn’t falsely advertise Youngevity’s financial status. Instead, it described issues the Pitcocks claim they experienced with Youngevity and attempts to explain Youngevity’s alleged behavior by stating, “perhaps [Youngevity] desperately need[s] money.” “While the post does operate to promote Wakaya, it is personal in nature as griping by disgruntled former employees and does not amount to an advertisement about Youngevity’s finances.”

Finally, a former distributor declared that, “[A]fter I joined Wakaya, Barb Pitcock told me in approximately June 2016 on the telephone that Youngevity was struggling financially and would go out of business.” But the distributor had already ended her involvement with Youngevity and became a Wakaya distributor, so that wasn’t a commercial advertisement.

False claims of the origin of Wakaya products: Wakaya’s YouTube page claimed that “Wakaya Perfection is a suite of wellness products by David H. Gilmour, the founder of Fiji water. The uniquely organic products are hand cultivated on the pristine island of Wakaya and made of 100 percent certified organic ginger powder and Dilo oil.”  Smith testified that while this description was accurate when it was first written, it became “untrue” or “not fully accurate” after Wakaya Perfection, LLC acquired the YouTube page and introduced products that included ingredients not sourced from the island of Wakaya. This was enough to create a genuine issue of material fact on literal falsity.

False advertising with respect to safety and health benefits: Wakaya claims that its clay product has “well known benefits,” “may remove toxins from the body,” and is “known” to “neutralize and balance acidic conditions,” “relieve digestion,” and “boost the immune system,” among other benefits. Youngevity’s expert reports that “[t]here is no scientific evidence that would support any therapeutic effects or claims of the consumption of bentonite clays” and that the clay may pose a health hazard because “use of unapproved chelating agents is dangerous and pose a serious risk to human health” and because those who consume the clay may be exposed to unsafe levels of lead and arsenic. This was enough to create a genuine issue of material fact.

Wakaya also advertised that its turmeric product contains six times more curcumin (5.96%) that traditional turmeric (0.92%), but Youngevity’s expert reports identified 2.45% and 3.10% instead via testing, creating a triable issue of fact.  In addition, the only quantitative data supporting the claim to have “a whopping five times more curcumin, the therapeutic agent in turmeric, above all conventional turmeric powders” was based on a test comparing Wakaya’s product with one other brand. Youngevity’s expert tested five other products with ranges from 1.72% to 3.92%.  Wakaya’s rebuttal expert opined that, because levels of organic compounds naturally vary in spices, test results of the percentage of certain compounds in products will also vary, so the testing wasn’t definitive; Wakaya argued that its claims were therefore at best unsubstantiated. [I disagree with this interpretation. There are two facts in evidence, not just one: (1) the percentage in the sample it tested, and (2) that there’s variation, whose range/average deviation is not established—the best understanding of the truth, based on those two facts, is that there is not the claimed percentage in the other lots being sold.  Part of this is an attempt to create epistemological uncertainty: how do we know what’s in any given bottle of a supplement, really? I think the best answer is that you shouldn’t advertise consistency if your own claim is that there’s variation.]  Anyway, the court agreed that the claims were unsubstantiated, but maybe they weren’t false, so Youngevity didn’t get summary judgment on falsity here.

Youngevity also didn’t create a genuine dispute on weight loss: “while the evidence includes testimonials of weight loss during specific periods of time, no promotions or advertising presented by Youngevity promise that users would or will lose weight.”  It’s notable that the FTC would consider these claims to promise that these claims are representative and thus they’d be at least implicitly false, e.g., “So far 100% of People have lost weight on our #Keto #BulaFIT Program” and “While we can’t say that 100% of the people will get results, so far we do have 20 out of 20 that have lost weight.” But Youngevity’s expert report just opined that the claim was “unrealistic, unsubstantiated, and very misleading,” which wasn’t enough under the Lanham Act.  Among other things, the ad claimed weight loss, not that the weight loss would be maintained for any period, and “[c]laims about weight loss are not inherently misleading just because they fail to include data about weight loss maintenance.”

Tuesday, July 09, 2019

Rogers still mostly works in 9th Circuit, despite Honey Badger's best efforts

Caiz v. Roberts, --- F.Supp.3d ----, 2019 WL 1755421 (C.D. Cal. Apr. 17, 2019)

In Gordon v. Drape Creative, 909 F.3d 257 (9th Cir. 2018), the Honey Badger case, the Ninth Circuit began a process that could make Rogers v. Grimaldi as reticulated and difficult as nominative fair use has often become.  Fortunately, the court here sees past the FUD enabled by Gordon and finds that the use of “Mastermind” in connection with rap music is protected by Rogers.

Caiz has a registration for MASTERMIND for for audio and video recordings featuring music (applied 2005, issued 2013).  Roberts is a hip-hop artist known as Rick Ross (who also, incidentally (?), won a First Amendment defense over his use of that name). Rick Ross released an album entitled “Mastermind,” went on a “Mastermind” tour, and allegedly adopted the persona of “Mastermind.” After the court of appeals reversed the district court’s initial finding that the registration was invalid for descriptiveness, the case returned for further proceedings on the infringement and related claims.

Once a defendant shows that its challenged use is part of “an expressive [nonadvertising] work protected by the First Amendment,” the burden then shifts to the plaintiff claiming infringement, who must show “that the mark is either not artistically relevant to the underlying work or explicitly misleading as to the source or content of the work.”  “If the plaintiff satisfies both elements, it still must prove that its trademark has been infringed by showing that the defendant’s use of the mark is likely to cause confusion.” 

Caiz argued that Rogers doesn’t apply to reverse confusion claims, relying on Masters Software, Inc. v. Discovery Communs., Inc., 725 F. Supp. 2d 1294 (W.D. Wash. 2010), which so held because it reasoned that Rogers required a referential use.  Since the Empire case rejected that underlying rationale, Rogers applied. Twentieth Century Fox TV v. Empire Distrib., Inc., 875 F.3d 1192 (9th Cir. 2017).

The uses here were in connection with artistic works, so Rogers applied. “Minimal” artistic relevance is enough.  Six of the nineteen songs on the album make direct use of “mastermind” in the lyrics, and one song referenced the overall album itself by its title (“Mastermind, my 6th LP. Can’t believe we did it. Man, I thank everybody that played a part of this.”), all of which linked the title to its contents.  Plus, there was [completely unnecessary] evidence that other people use “mastermind” as “part of a larger, abstract theme used by artists in hip-hop claiming to be masterminds of music.”  That was enough.

But was the use “explicitly” misleading? Here, courtesy of Gordon, we descend into a theoretical wasteland in which the implicit can in theory be explicit.  Caiz argued that the title of the album was explicitly misleading as to source because Caiz had a registration for musical recordings.  The use of “Mastermind,” Caiz argued, was an explicit reference to him and thus misleading.  As the court noted, the usual rule is that “mere use of a trademark alone cannot suffice to make such use explicitly misleading,” but “each time it made that observation, the junior user had employed the mark in a different context than the senior user.”  [Gordon says this, but of course Rogers derives from a case in which Ginger Rogers, movie star, sued over use of her name in a movie; the Ninth Circuit rejected the Second Circuit’s exclusion of title-v-title disputes from Rogers, but Gordon allows the same argument to be resurrected in much broader, and worse, form.] Under Gordon, “identical usage” could be [but is not always as a matter of law] explicitly misleading.

Gordon added a mini-transformativeness inquiry into explicit misleadingness.  For “identical usage,” it’s now relevant how much creativity the junior user has added.  Worries over confusion are limited “when the mark is used as only one component of a junior user’s larger expressive creation, such that the use of the mark at most ‘implicitly suggest[s]’ that the product is associated with the mark’s owner.”  By contrast, the use of a mark “as the centerpiece of an expressive work itself, unadorned with any artistic contribution by the junior user,” may qualify as explicitly misleading. 

In contrast to the use in Gordon, where the allegedly source-indicating catchphrase was being used as the “centerpiece” of greeting cards [and thus not as a mark, sigh], Roberts used “Mastermind” as one album title out of six albums throughout his career, where he was established as Rick Ross.  “Mastermind” appeared nine times in the lyrics across nineteen tracks, and it was used “through” Roberts’ own artistic expression.  [This discussion really highlights the ©/TM mashup Gordon achieved.  Of course, neither party created the word “mastermind.”  The only way to use the word creatively is to use it.  Gordon was worried about the creativity behind “Honey Badger Don’t Care” and used TM to give the plaintiff a quasi-copyright in a short phrase; that concern is obviously irrelevant here, as it should have been in Gordon.]  The marketing of “Mastermind” the album attached Rick Ross’s persona and history to it. For example,, is titled “Rick Ross The Boss” and introduces him as “Rick Ross AKA Ricky Rozay AKA the Boss AKA The Mastermind.” “In short, Plaintiff’s music lists the artist as ‘Mastermind,’ while Defendants titled one of Roberts’ albums and tours ‘Mastermind,’ while retaining some sort of reference to “Rick Ross” as the artist and incorporating Roberts’ own artistic expression.”  [There are two argumentative threads here: the first one, forced by Gordon, is weird and about the fact that the album has songs in it and is thus more than Roberts just repeating the word “mastermind,” though that would also be a creative, albeit more boring, work protected by copyright.  The second, which goes more to the original Rogers title-v-title exception, is that the uses aren’t in fact the same. Thus the premise of Gordon isn’t even satisfied here and we don’t need all this handwaving about creativity.]

Caiz had no evidence that the use was “explicitly” or even “implicitly” misleading, only a declaration stating that his fans and fans of Rick Ross started asking him if a new album was coming out, “misleadingly thinking Defendant’s album was [his].” That wasn’t enough to avoid summary judgment.  There was no evidence of any explicit statements of association etc. by defendants (citing Novalogic, Inc. v. Activision Blizzard, 41 F. Supp. 3d 885, 901 (C.D. Cal. 2013) (“To be ‘explicitly misleading,’ a defendant’s work must make some affirmative statement of the plaintiff’s sponsorship or endorsement, beyond the mere use of the plaintiff’s name or other characteristic.”)).

It was not enough to argue that the use was explicitly misleading because reverse confusion was likely. Likelihood of confusion analysis isn’t part of Rogers prong two; indeed, avoiding a fullscale likely confusion analysis is the point of Rogers and the only thing that makes it protective of First Amendment interests rather than an additional overlay on a complex and fact-intensive balancing test.  Under Gordon, only after lack of artistic relevance or explicit misleadingess has been shown is a likely confusion analysis appropriate.

Finally, Caiz argued that even if Rogers protected the album name, it didn’t protect use of the term for the tour name or for Roberts’ persona in interviews and other live performances. But these were protected “extensions of the use of the mark in an effort to advertise and market the ‘Mastermind’ album,” just as in Empire appearances by cast members, radio play, online advertising, and live events were protected advertising for the expressive work.

The sensible reasoning of this opinion does a lot to indicate that, though Gordon was badly reasoned, it has limited application outside of the scenario in which the defendant basically just reproduces the plaintiff’s mark on the goods for which the plaintiff claims a mark.

Monday, July 08, 2019

Amazon strictly liable for product defect but 230 preempts failure to warn claim

Oberdorf v. Inc., No. 18-1041 (3d Cir. Jul. 3, 2019)

Oberdorf was injured by a retractable dog leash “sold” by Amazon on behalf of a third-party vendor, who shipped the leash directly to Oberdorf. The district court found that, under Pennsylvania law, Amazon was not a “seller” who could be strictly liable for Oberdorf’s injuries, and that the CDA barred her claims.  The court of appeals reversed the first finding and partially reversed the second.

As one might expect, the Amazon contract with vendors allows Amazon almost total control of the relationship between the parties.  The vendor chooses the products it sells and provides Amazon with a product description, digital images, shipping and handling options, and other information it requests. Amazon has sole control over the content, appearance, design, functionality, and all other aspects of its website. The vendor can choose the price, but can’t charge more on Amazon than in other sales channels.  [Hmm.  I can think of at least one wooden jigsaw puzzle vendor that seems not to follow this rule, but perhaps it’s actually some hidden-to-me arbitrage by a third party.] Vendors must communicate through the Amazon platform. When there’s a purchase, Amazon collects the payment and requires the vendor to send Amazon shipping information for each order. Of course, it also charges commissions and fees, and is the vendor’s agent for purposes of payments and refunds.  Amazon can, among other things, require vendors to stop or cancel orders of any product. It further requires that its vendors release it and agree to indemnify, defend, and hold it harmless against any claim, loss, damage, settlement, cost, expense, or other liability.

In Pennsylvania, strict products liability is limited to “sellers.”  The Pennsylvania Supreme Court identified four relevant factors in identifying a “seller”:

(1)       Whether the actor is the “only member of the marketing chain available to the  injured plaintiff for redress”;
(2)       Whether “imposition of strict liability upon the [actor] serves as an incentive to safety”;
(3)       Whether the actor is “in a better position than the consumer to prevent the circulation of defective products”; and
(4)       Whether “[t]he [actor] can distribute the cost of compensating for injuries resulting from defects by charging for it in his business, i.e., by adjustment of the rental terms.”

Under this test, Amazon is a “seller.”  First, Amazon isn’t like a brick-and-mortar auctioneer.  Even though every item on Amazon’s website can be traced to a third-party vendor, the only way to communicate with customers is through Amazon. “This enables third-party vendors to conceal themselves from the customer, leaving customers injured by defective products with no direct recourse to the third-party vendor. There are numerous cases in which neither Amazon nor the party injured by a defective product, sold by, were able to locate the product’s third-party vendor or manufacturer.”  Amazon’s VP of Marketing Business “admitted that Amazon generally takes no precautions to ensure that third-party vendors are in good standing under the laws of the country in which their business is registered. In addition, Amazon had no vetting process in place to ensure, for example, that third-party vendors were amenable to legal process.  After Oberdorf was injured by the defective leash, neither she nor Amazon was able to locate The Furry Gang [the vendor].”  For compensation, it’s Amazon or nothing, weighing in favor of strict liability.

The dissent argued that “[t]o assign liability for no reason other than the ability to pay damages is inconsistent with our jurisprudence,” but it wasn’t just ability to pay that was Amazon’s problem.  “Amazon fails to vet third-party vendors for amenability to legal process. The first factor weighs in favor of strict liability not because The Furry Gang cannot be located and/or may be insolvent, but rather because Amazon enables third-party vendors such as The Furry Gang to structure and/or conceal themselves from liability altogether.”

Second, whether “imposition of strict liability upon the [actor would] serve[] as an incentive to safety”: Imposing strict liability on an auction house wouldn’t help because an auction house doesn’t design or make any particular product; Amazon too argued that it didn’t have any relationship with designers or manufacturers.  However, though Amazon didn’t directly control design and manufacture, it exerted “substantial control” over third-party vendors by virtue of its comprehensive, discretion-allocating agreement with them. “Amazon is fully capable, in its sole discretion, of removing unsafe products from its website. Imposing strict liability upon Amazon would be an incentive to do so.”

The dissent argued that this holding imposes a fundamentally new business model on Amazon because it presently “does not undertake to curate its selection of products, nor generally to police them for dangerousness.”  Echoing what I tell my students (there is no divine entitlement to a specific business model), the court said that Pennsylvania law does not shield a company from strict liability just because it chose a business model that fails to prioritize consumer safety. “The dissent’s reasoning would give an incentive to companies to design business models, like that of Amazon, that do nothing to protect consumers from defective products.”

Third, whether Amazon is “in a better position than the consumer to prevent the circulation of defective products.”  An auctioneer lacks an ongoing relationship with a manufacturer, and financing agencies perform only a “tangential” role in the sales process and ordinarily lacks a “continuity of transactions that would provide a basis for indirect influence over the condition and the safety of the product.” Here, by contrast, “while Amazon may at times lack continuous relationships with a third-party vendor, the potential for continuing sales encourages an on-going relationship between Amazon and the third-party vendors.”  In addition, “Amazon is uniquely positioned to receive reports of defective products, which in turn can lead to such products being removed from circulation.”  Amazon’s website is the “public-facing forum” for listed products; it collects customer feedback.  Third-party vendors are ill-equipped to do this precisely because Amazon controls the channels of communication.

The dissent argued that Amazon wasn’t better positioned than customers (!) to encourage product safety. But even the dissent noted one aspect of Amazon’s power relative to the consumers: Amazon, but not consumers, can eject sellers from the platform. “Imposing strict liability on Amazon will ensure that the company uses this relative position of power to eject sellers who have been determined to be selling defective goods.”

Fourth, whether Amazon can distribute the cost of compensating for injuries resulting from defects: Actually, it already has provided for indemnification in its agreements. And it can adjust the commission-based fees that it charges to third-party vendors “based on the risk that the third-party vendor presents.” By contrast, “Amazon’s customers are particularly vulnerable in situations like the present case. Neither the Oberdorfs nor Amazon has been able to locate the third-party vendor, The Furry Gang. Conversely, had there been an incentive for Amazon to keep track of its third-party vendors, it might have done so.”

This result was also consistent with other Pennsylvania cases, which established that a “seller” need never take title to or possession of the products at issue to be strictly liable. Amazon also argued that it wasn’t a sales agent because it works on behalf of numerous third-party vendors, not a single seller or manufacturer; this is not the law.

CDA §230: Bars some, but not all, of the claims. “The question that we must answer is ‘Would such an addition to the content be part of the editorial function of the Amazon website?’”  Not all of the claims were precluded.  “Amazon’s involvement in transactions extends beyond a mere editorial function; it plays a large role in the actual sales process. … Therefore, to the extent that Oberdorf’s negligence and strict liability claims rely on Amazon’s role as an actor in the sales process, they are not barred by the CDA. However, to the extent that  Oberdorf allged that Amazon failed to provide or to edit adequate warnings regarding the use of the dog collar, that activity fell within its editorial function. “That is, Amazon failed to add necessary information to content of the website.”  Such failure to warn claims were barred by the CDA. But the district court didn’t parse the claims to distinguish between “failure to warn” claims and claims premised on other actions or failures in the sales or distribution processes, so remand was required.

Judge Scirica dissented on the Pennsylvania law issue, but concurred in the “thoughtful” CDA analysis.

Friday, July 05, 2019

Warhol wins on fair use of photo (but should've won on substantial similarity)

Andy Warhol Foundation for the Visual Arts, Inc. v. Goldsmith, No. 17-cv-2532 (JGK) (S.D.N.Y. Jul. 1, 2019)

In some sense, this fair use case is a foregone conclusion; even the terrible Saderup decision made an exception for Andy Warhol, because we all know that his touch (filtered through the actual human touch of assistants) confers new meaning and value on an artwork.  The other thing this case illustrates is that courts are more comfortable with fair use than they are with a true infringement inquiry (did the defendant copy too much protected material from the plaintiff?) when the real problem with the claim is that the defendant copied without taking very much, if any, protected material.  The third factor analysis here even implicitly admits that the plaintiff hasn’t identified any expression that was taken from her photograph of Prince.  If we were really concerned that transformativeness has gone too far—I’m not, but I also think we should be serious about requiring substantial appropriation of protected expression—then one way to deal with that problem would be to take infringement more seriously rather than using fair use as a clean-up tool.  The finding of transformativeness here is in part the flip side of the lack of copying of protected expression: Warhol’s prints were readily able to bear new meaning and message because the expression in the original photo had been abstracted away, not because of a Sherrie Levine-style appropriation.

Anyhow, “Lynn Goldsmith is a photographer who has photographed numerous rock, jazz, and R&B performers,” and her work “centers on helping others formulate their identities, which she aims to capture and reveal through her photography.”  She uses both interpersonal techniques to establish rapport and photographic techniques with respect to lighting, camera position, and other elements to capture her subjects’ “true selves.” She photographed Prince in her studio on assignment from Newsweek in 1981. He arrived wearing makeup, and she applied more “to connect with Prince physically and in recognition of her feeling [that] Prince was in touch with the female part of himself” while also being “very much male.” He was photographed in his own clothes, except for a black sash that he picked from Goldsmith’s clothing room and wore around his neck. Goldsmith decided to use a plain white background and lit the shoot in a way that emphasized Prince’s “chiseled bone structure.” Goldsmith believed that the photographs from her shoot with Prince show that he is “not a comfortable person” and that he is a “vulnerable human being.”
1984 Vanity Fair portrait
In 1984, Vanity Fair licensed one of these studio portraits for use as an artist’s reference. Goldsmith’s photography agency submitted the photo; Goldsmith herself did not know at the time that the photograph had been licensed for use as an artist’s reference. Vanity Fair commissioned Warhol to create an illustration of Prince for an article titled “Purple Fame,” which stated that it featured “a special portrait for Vanity Fair by ANDY WARHOL.” The credit included: “source photograph © 1984 by Lynn Goldsmith/LGI.” Warhol then created the “Prince Series,” comprised of sixteen distinct works: twelve silkscreens, two screen prints on paper, and two drawings. Although Goldsmith alleged that Warhol bodily copied her photo as part of his creation process, defendant AWF didn’t concede this and it doesn’t matter, because any exact reproduction occurred 40 years ago, well beyond the limitations period.
color version of one of the images

Goldsmith first learned that Warhol created the Prince illustration for Vanity Fair after Prince’s death, when it republished the image online; initially, she told AWF that use infringed one of her colored Prince portraits but, after further comparison, identified instead the black and white photo at issue in this case. She then registered the photo as an unpublished work. AWF makes the Prince Series available for licensing to third parties for use in books, magazines, newspapers, and for other merchandizing purposes. Goldsmith licenses single images of her photography, and has issued 10 or 11 licenses for other photos of Prince in concert and at other venues, but hasn’t editioned or sold any prints of the photo here; she intends to start in the future, when prices will be higher. In 2004, she sold a fine-art print of Prince that she created in 1993 to a private collector who also owns three Warhol works of art.

Fair use factor one: the Prince Series is commercial, but public exhibition of art is in the public interest. Anyway, transformativeness trumps commerciality. The question is whether new meaning/message may reasonably be perceived and the answer is yes.  Goldsmith focused on revealing identity, and her photo illustrated that Prince was “not a comfortable person” and that he is a “vulnerable human being.”

Warhol’s Prince Series, in contrast, can reasonably be perceived to reflect the opposite. In all but one of the works, Prince’s torso is removed and his face and a small portion of his neckline are brought to the forefront. The details of Prince’s bone structure that appear crisply in the photograph, which Goldsmith sought to emphasize, are softened in several of the Prince Series works and outlined or shaded in the others. Prince appears as a flat, two-dimensional figure in Warhol’s works, rather than the detailed, three-dimensional being in Goldsmith’s photograph. Moreover, many of Warhol’s Prince Series works contain loud, unnatural colors, in stark contrast with the black-and-white original photograph. And Warhol’s few colorless works appear as rough sketches in which Prince’s expression is almost entirely lost from the original.
These alterations result in an aesthetic and character different from the original. The Prince Series works can reasonably be perceived to have transformed Prince from a vulnerable, uncomfortable person to an iconic, larger-than-life figure. The humanity Prince embodies in Goldsmith’s photograph is gone. Moreover, each Prince Series work is immediately recognizable as a “Warhol” rather than as a photograph of Prince - in the same way that Warhol’s famous representations of Marilyn Monroe and Mao are recognizable as “Warhols,” not as realistic photographs of those persons.

One could reasonably object to the last sentence as carving out an Andy Warhol exception, but the rest of it is hard to dispute (and provides some reason to think that Warholization is transformative, albeit not a tactic limited to Warhol himself).

Factor two: unpublished status would ordinarily weigh in Goldsmith’s favor, but “the reasons unpublished works enjoy additional protection against fair use - including respect for the author’s choices of when to make a work public and whether to withhold a work to shore up demand - carry little force in this case, where Goldsmith’s photography agency licensed the photograph for use as an artist’s reference.” Anyway, factor two is of limited relevance for transformative works. Favors neither party.

Factor three: Goldsmith argued that the Prince Series works contain the essence of the entire Goldsmith Prince photo.  Her best argument for this was apparently that Vanity Fair told him to use the photo, and thus must have required that he include the expression in the photo.  The court compared this case to the Seventh Circuit case of Kienitz v. Sconnie Nation LLC, and helpfully included the relevant images in the opinion to show why the comparison was apt. The Kienitz court, while—as the district court here specifically noted—criticizing Cariou, found fair use, placing particular emphasis on the third factor. The Warholization-like process employed “removed so much of the original that, as with the Cheshire Cat, only the smile remains,” even though it weighed against the defendants the claim that they didn’t need to use that particular photo “when so many noncopyrighted [sigh] alternatives (including snapshots they could have taken themselves) were available.” Here, by contrast, Warhol was required to use the photo.
Kienitz images
This case was Kienitz plus Cariou: Though the Goldsmith photo had protectable elements, which could include “posing the subjects, lighting, angle, selection of film and camera, evoking the desired expression, and almost any other variant involved,” “these creative elements are almost entirely absent from the Prince Series works.”  The cropping was different; Goldsmith’s photo included much of Prince’s torso.  The Prince Series softens, shades, or traces over the sharp contours of Prince’s face that Goldsmith emphasized in her photo. The 3D effect of the photo, produced by the background and lighting that Goldsmith chose, was replaced by “a flat, two-dimensional and mask-like figure of Prince’s head,” and mostly on a loudly colored background; the Warhol works that were in black and white “especially crude and the creative features of the Goldsmith Prince Photograph are especially absent.”  Here you see the flip side of transformativeness in the factor three analysis: “Ultimately, Warhol’s alterations wash away the vulnerability and humanity Prince expresses in Goldsmith’s photograph and Warhol instead presents Prince as a larger-than-life icon.”

The pose and angle of Prince’s head were copied, but “such a pose cannot be copyrighted” because copyright law “protect[s] only plaintiff’s particular photographic expression of [a] pose[] and not the underlying ideas therefor.” Several non-Goldsmith photographs also captured Prince in a similar pose, “indicating that the pose is not particularly original.” The distinctive way in which Goldsmith presented Prince’s uncopyrightable facial features was absent from the Warhol works, which each contained “little, if any, of the copyrightable elements” of Goldsmith’s photo (which is why this should be a non-substantial similarity case).  Heavily favors fair use.

Factor four: Goldsmith argued that the Prince Series harmed her licensing markets, which overlap with AWF’s licensing markets.” Her evidence didn’t show market substitution.  Although her photos and Warhol’s works have appeared in magazines and on album covers, “this does not suggest that a magazine or record company would license a transformative Warhol work in lieu of a realistic Goldsmith photograph.”  The licensing market for Warhol prints is for Warhols, not for portrait photos like Goldsmith’s. One collector who owned three of Warhol’s works of art also bought a fine-art print of Prince from her. “But as AWF persuasively argues, this does not suggest that the collector bought the works for the same reason, perceives the works similarly, or believes the works are substitutes for each other (the fact that the collector owns both of them suggests the opposite).” The court declined to rely on AWF’s expert report or on one of its fact witnesses on the market, and didn’t rule on excluding Goldsmith’s expert; even taking his opinions into account on licensing, the fourth factor favored AWF. “The evidence shows that the Prince Series works are not market substitutes that have harmed - or have the potential to harm Goldsmith.” And we’re done.