Monday, October 15, 2012

Nominative fair use of trade dress

Aviva USA Corporation v. Vazirani, No. CV 11-0369-PHX-JAT (D. Ariz. Oct. 2, 2012)

In 2009, Aviva, an insurer, terminated Vazirani’s authorization to sell the life insurance and annuity products of Aviva’s issuing affiliates.  Vazirani, feeling ill-used, registered a number of domain names such as,,,,,,,, and  He used an altered version of Aviva’s website trade dress.  His website had sections such as “Anil’s Background,” “Anil’s Case Against Aviva,” detailing his lawsuits against Aviva; “Aviva’s Class Action Settlement,” linking to a complaint alleging that Aviva deceptively marketed to elderly people; and “Tell us your story about Aviva.”

The court first found that Vazirani didn’t engage in the necessary use in connection with the sale of goods or services to sustain a trademark claim. Such noncommercial use can’t create liability under either §32 or §43(a).  “The distinction between commercial and noncommercial use is particularly important in the context of consumer or editorial criticism and commentary.”  However, when criticism and unauthorized use occur in connection with the sale of goods or services, a successful Lanham Act claim is at least possible. 

Comment: This doctrinal dance would be a lot simpler if the courts admitted that this scenario only makes a false advertising claim potentially successful, where there’s a false factual claim; it remains implausible that consumers would be confused by unauthorized use of a trademark to criticize the owner.  Thus, the court cites cases in which competitors disparaged other competitors, even anonymously or in disguise, and are found to have engaged in commercial speech.  Those cases—which seem to me plainly correct—do create a little bit of trouble for Vazirani in that there’s a commercial element to his vitriol, but the trouble shouldn’t be trademark-related.  The court will eventually get around to this in its nominative fair use analysis, but probably should have started there, because its commercial use analysis sets up a defense for concealed disparagement by a competitor.  (Or maybe, like Citizens United, the courts are just going to create this workaround for Lanham Act claims by allowing anonymous/pseudonymous touting/disparagement to escape the constraints of false advertising law—but I sure hope not.)

Anyway, the court here says that this case is like Bosley.  There was no evidence that Vazirani offered any goods or services for sale on the website, or linked to other sites that did.  Nor did defendants ever attempt to sell the site or domain names to anyone, including Aviva, for profit.  Aviva’s argument that the website attempted to inflict commercial harm on Aviva, and was thus commercial, contradicted 9th Circuit precedent (and good sense).

But what about the fact that the website touted Vazirani’s financial accomplishments and credentials as a seller of products that compete with Aviva’s affiliates?  This could lead them to do business with him (and a quick search would reveal his website).  Also, his avowed purpose was to get money from Aviva through settling his lawsuit and get his contracts with Aviva restored.  (The problem here is that Aviva has mixed an argument that should matter with an argument that shouldn’t—wanting to get money from Aviva isn’t a competitive purpose.)

The court disagreed.  There was no mention of Vazirani’s business website on the anti-Aviva website.  Thus, the court didn’t find it credible that Vazirani was trying to drive business to his other website.  And his demands for settlement etc. were no different than those in the dispute in Bosley, which also didn’t make a critical site into a commercial site.  He didn’t try to hold Aviva’s trademarks hostage; rather, he threatened to publicize information that he said would harm Aviva’s reputation.  His belief that he’d been injured by Aviva and deserved restitution was the same as Kremer’s belief that he’d been injured by Bosley Medical.

As to the argument that Vazirani was a competitor and that the website promoted competing goods or services, the court was equally unconvinced.  For purposes of summary judgment, the court accepted that the parties competed, though it wasn’t clear that this was so (since they appear to operate at different parts of the insurance chain) and anyway there were no references to competing products on the anti-Aviva website.  The only part of the website that could “even remotely be considered relevant to this argument” was in the “Anil’s Background” section:

I’m Anil Vazirani and by any measure I’m one of the most successful financial advisers in America.  I’m a “Top of the Table” member in the prestigious Million Dollar Round Table, I was a 2004 Hall of Fame Inductee into the Society of Senior Market Professionals, and I have been featured and interviewed in leading industry publications.  I’ve been a member of the National Association of Insurance and Financial Advisors for nearly two decades.

And “Anil’s Case Against Aviva” said, “For many years Aviva’s products were among those I used to build diversified portfolios for my clients.”  The court concluded that “[t]o interpret these statements as advertising goods or services that compete with Aviva is not reasonable.”  None were identified on the website; at most there was an implication that Vazirani might currently sell competing products.  There was no contact information or any other information that would allow a visitor to determine whether this was true or how s/he might go about purchasing such products from him.  “To describe as attenuated this path from the Website to any potential offers by Mr. Vazirani to sell competing products is an understatement.”

Aviva argued that Vazirani sent out links to the website in emails sent to “an audience of hundreds of thousands of life insurance and annuity agents in the United States,” and thus some visitors would already be familiar with Vazirani’s offerings.  But Aviva also argued that the emails were materially misleading in that they didn’t reveal that they came from defendants.  (I don’t see how this matters to commerciality, as noted.)  Anyway, it wasn’t clear whether such recipients would be relevant consumers, since they compete with Vazirani.  But, because the emails didn’t contain any reference to or promotion of defendants’ products, the connection to commerciality was at least as “roundabout” as that in Bosley.

Aviva argued that gripe sites run by competitors satisfy the commercial use requirement.  But in the cited cases, HER, Inc. v. Re/Max First Choice, LLC, 468 F. Supp. 2d 964 (S.D. Ohio 2007) and Sunlight Saunas v. Sundance Saunas, 427 F. Supp. 2d 1032 (D. Kan. 2006), there was more evidence of self-promotion.  In HER, defendants directed consumers to their own website via email and rerouted allegedly infringing domain names to their own site.  In Sunlight Saunas, which I have criticized on similar grounds, the court found that the defendants “had no apparent reason to disparage [the plaintiff’s] products except to promote their own.” The court also noted that the gripe site temporarily included direct links to competitors and said that “other companies offer the same products without the fraudulent claims.”  By contrast, Vazirani’s website didn’t link to sites offering competing products or promote any such products.  Nor did the domain names reroute to other competing sites.  All the domain names went to the anti-Aviva site. 

Quoting Bosley, the court concluded, “Any harm to [Aviva] arises not from a competitor’s sale of a similar product under [Aviva’s] mark, but from [Defendants’] criticism of [Aviva].”  Which is why this isn’t a trademark case!

Though the court granted summary judgment on the infringement claims based on this analysis, it went on to evaluate likely confusion/nominative fair use, a slam dunk.  The only interesting part here is the trade dress bit.  The court concluded that Vazirani’s use of the trademark and the trade dress “was undoubtedly a nominative use—that is, the mark was used to refer to Aviva and its products and services rather than Defendants and their products and services.” Instead of being used to describe Vazirani’s goods/services, Aviva’s marks were used to identify the target of criticism.  “[T]here is no clear way for Defendants to convey their criticism for a specific entity, Aviva, without naming that specific entity.”  The third factor was also easily satisfied, using an abbreviated likely confusion analysis.  “[T]he entire website is concerned with criticizing Aviva and its business practices.  Therefore, it is not reasonable to conclude that Defendants have taken any actions that would suggest that Aviva supports or endorses the Website or the associated domain names in any way.”  (Of course, a better substitute for the confusion test, as the nominative fair use test purports to be in the 9th Circuit, wouldn’t incorporate a likely confusion analysis, but here the quick look makes the answer easy—and the other two elements of the test irrelevant—in the same way that a true “does nothing else to suggest endorsement” defense would.)

The second factor was merely “a closer question,” even though Vazirani imitated the Aviva website trade dress of a stylized blue font on a yellow background with rays of light in the background:

In certain cases, such distinctive stylizations might be considered more than is permitted under the Ninth Circuit’s nominative fair use test.  See New Kids on the Block, 971 F.2d at 308 n.7 (“Thus, a soft drink competitor would be entitled to compare its product to Coca-Cola or Coke, but would not be entitled to use Coca-Cola’s distinctive lettering.”). Here, however, the stylized logo and distinctive coloring were not used in a commercial or competitive manner, but rather were used solely to identify Aviva as the object of the Website’s criticism.  Further, Defendants embedded their own critical commentary within the logo, such that it read “Aviva Uncovered The Sad Truth About Aviva’s Business Practices.” 

Given that the purpose of nominative fair use is to address the true likelihood of confusion caused by a certain type of use, and that Tabari specifically identified the second factor as indirectly addressing the risk of confusion (“Consumers may reasonably infer sponsorship or endorsement if a company uses . . . ‘more’ of a mark than necessary.”), here the amount of the trade dress used was reasonable.  “In other words, Defendants’ use of the more distinctive colors and font, in light of the very obvious negative commentary directed toward Aviva that is included in the logo, could not reasonably lead to such confusion.”  So, nominative fair use.

On the one hand, obviously!  On the other, does nominative fair use advance the ball beyond (1) a robust defense for expressive uses such as in Rogers and (2) strong protection for comparative advertising/criticism?  Bill McGeveran has suggested that it doesn’t, and (I am paraphrasing, and he might well not endorse this description) that the weird role of quick-look confusion analysis in the practical application of the test shows its incoherence.  I see what he’s saying, but I’m not sure there are less incoherent alternatives out there. 

At some point we simply have to say: we are not interested in making the defendant spend $100,000 on developing a summary judgment record; on its face this is not the kind of activity likely to be found confusing, nor should we encourage plaintiffs to roll the dice and hope they can suppress critical speech on trademark grounds.  As in Wal-Mart, the error costs of case-by-case analysis are greater than the error costs of a blanket rule.  Speech that criticizes the trademark owner is not actionable on trademark grounds; relief if any must be found in the law of false advertising or, where that doesn’t apply, defamation/disparagement.  Properly applied, nominative fair use can do that work across many types of referential uses, and that’s at least a point in its favor compared to having to figure out what to do with something like this, which isn’t really comparative advertising but is kind of competitive.

No comments: