Is a merchandising right part of TM? one PA judge thinks probably not

Pennsylvania State Univ. v. Vintage Brand, LLC, No. 4:21-CV-01091 (M.D. Pa. Jul. 14, 2022)

Penn State sued Vintage for selling screen-printed items with old Penn State logos. Vintage counterclaimed, alleging that Penn State’s registrations were invalid and should be cancelled on account of the ornamentality of the use. The court declined to dismiss the counterclaim for cancellation, reasoning that a symbol on goods does not identify a third-party source of the goods if it merely “creates an association” with the trademark owner.

Penn State has a 1984 registration for the word mark PENN STATE for lots of stuff; it also has a registration for the “Pozniak Lion,” which Vintage Brand alleged was used as a school logo until it was phased out in 1987. Penn State registered this image in 2017 for use on metal novelty license plates and apparel. And it has a 2017 registration for the Penn State seal, which displays the Pennsylvania Coat of Arms ringed by the University’s name and date of founding, for various apparel and drink-related goods.

Vintage pled that each was “used as mere decoration, printed in large font and in a prominent location, and [do] not serve to identify Penn State as the source or origin of the goods”; that “[o]n information and belief, consumers perceive . . . [the marks] to be merely a decorative feature of the goods and not an indication of the source of the goods”; and finally that their “overall commercial impression . . . is purely ornamental or merely a decorative feature[,] . . . [and they] do not identify and distinguish Penn State’s goods from those of others and, therefore, do not function as a trademark.”

Though Vintage had the burden of showing invalidity, it had pled sufficient facts to indicate that it could meet this burden. (The court noted that it understood that Vintage was making an “as applied” challenge to the categories of goods at issue, not a challenge to the entirety of the registrations. I’m not sure this term is applicable. If the registrations for those categories are subject to cancellation, that’s not “as applied”—the registrations would be gone in those categories, presumably on failure to function/abandonment grounds. The reality, as the court noted elsewhere in its discussion, is that scope and protectability are entwined in this case: the key issue is whether Vintage can put the matter across the front of T-shirts, etc., not whether Penn State could get a valid registration for its marks used as marks.)

Using traditional ornamentality analysis—whether the matter was serving “solely as attractive ornamentation” and not “also as a symbol that identifies and distinguishes a single source”—the uses here were plausibly ornamental. Placement considerations—positioning on the goods as if they were the communication/decoration consumers wanted, rather than in the traditional “brand spots”—favored ornamentality. But even with that placement, the matter could still identify the source of goods. Nonetheless, it was plausible that “consumers believe that the essence of these marks is to signal their support for the University, not that the University has produced, approved, or guaranteed the quality of the item.” There was no bright-line rule making it implausible that consumers would perceive the matter as anything but a trademark indicating origin of the goods, as opposed to a signal of their own preferences.

Implicit in the court’s discussion of “affiliation” is the correct point that affiliation can easily be unidirectional. If I wear a Georgetown sweatshirt, I am indicating my support for/feelings of affiliation with Georgetown. But Georgetown is indifferent to me, and no reasonable consumer would look at me and think I represent Georgetown—otherwise all that collegiate licensing, where they let anyone who pays walk out the door with branded gear, is a massive naked licensing scheme. So use to signal the consumer’s “affiliation with” or “support for” an entity does not necessarily mean use that signals formal affiliation on the entity’s side.

The court discussed many merchandising cases and found that, with the exception of Boston Pro. Hockey Ass’n v. Dallas Cap & Emblem Mfg. Inc., 510 F.2d 1004 (5th Cir. 1975), they insisted that a trademark must function as a source-indicator; signifying merely a mark’s own presence on goods was insufficient. Even the Fifth Circuit, the court noted, has subsequently claimed that Boston Hockey was about source confusion and thus a fact-dependent inquiry about source identification (“the analysis turns not on whether consumers tie the symbol to the trademark holder, but on whether they tie the product to the trademark holder”). This is consistent with the basic premise of trademark protection. (The court noted some similar language from the TTAB, and similar walking back.)

Where “the mark itself is the product,” it performs a “non-trademark function.” The key question is whether it also performs a trademark function of signalling source. And this, the court ruled, was a factual question not amenable to resolution on a motion to dismiss given the allegations of the counterclaim, even though Penn State might ultimately succeed on the merits. “[T]rademark law requires more than a mental association between the trademark and trademark holder. … [T]he consumer must instead believe that the trademark indicates that the trademark holder is the source, sponsor, or is otherwise affiliated with the good—a question of fact.”

The court ended with some observations drawn from the scholarly literature criticizing a broad merchandising right. Trademarks aren’t supposed to incentivize creation; they are supposed to indicate source. In the absence of that function, monopolization is a bad, not a good. Dastar provides a hook for questioning a broad merchandising right, given its statement that “origin” “should not be stretched to cover matters that are typically of no consequence to purchasers.” Moreover, in the absence of an “official” claim, the proper remedy may be a disclaimer. The court referred to “recent Supreme Court cases” as supporting this conclusion (citing Bonito Boats, Kellogg, Traffix, and Stiffel)—super interesting to me because First Amendment cases strongly support a disclaimer remedy, outside the narrow trademark context.

The court pointed out that the facts remained for development, but might still support a disclaimer remedy given that consumer confusion might result from misunderstanding of the law, rather than confusion about specific factual situations:

Though the data is now decades old, early N.F.L. trademark cases saw consumer confusion rates above 50%. These results are not determinative here and can certainly be quibbled with at the margin….

But the case for a disclaimer rather than an injunction lies at the base, not the margin. Consumer-survey data taken at the same time as the N.F.L. cases showed widespread belief among consumers that “[n]o product can bear the name of an entertainer, cartoon character, or some other famous person unless permission is given for its use”—a belief that appears to have come from what they thought the law required. The circularity is apparent: the law only offers protection if there’s belief, yet the belief comes from consumers’ (mis)conception about the law. It would seem perverse to award market exclusivity based on a fake-it-until-you-make-it approach. If consumers’ confusion stems from their incorrect belief that goods bearing Penn State’s emblem must be licensed, shouldn’t that belief be corrected, not perpetuated?

Good question! The court indicated its interest in factual development both of the actual confusion question and the source of any consumer confusion, indicating that confusion based on miscomprehension of the law would likely be redressed by a disclaimer.

“The modern collegiate trademark- and licensing-regime has grown into a multibillion-dollar industry. But that a house is large is of little matter if it’s been built on sand.”

I’ve been saying for a while that the transformation of the federal judiciary along ideological but IP-indifferent lines is going to lead to large and unpredictable changes in IP doctrine. Here is one possible example.