In trademark dispute between guitar manufacturers, Fifth Circuit reverses district court’s pretrial order excluding evidence of decades-old third-party-use evidence, which defendants sought to introduce at trial in support of genericness defense, and remands with instructions for new trial.
Plaintiff Gibson Inc. and defendant Armadillo Distribution Enterprises Inc. have been involved in the guitar manufacturing and distribution business more than 50 years. In the 1960s, Gibson introduced its “SG” model, which offered a slimmer-body guitar with horns protruding from the connection of the guitar’s neck and body. In the late 1970s, Dean Guitars Inc. released two guitar models, which grew in popularity in the 1980s as famous guitarists began playing on Dean-brand guitars. In 1997, Armadillo purchased the Dean brand, and it has continued to produce and sell Dean- brand guitars. In 2013, Armadillo released its own “Gran Sport” model guitar, which, like Gibson’s SG model, contained a body with horns extending from the guitar’s neck.
In 2004, Concordia Investment Partners LLC, a holding company for Armadillo, filed a trademark application with the U.S. Patent and Trademark Office for the registration of a guitar with a “winged-silhouetted headstock design without any brand name on it.” Gibson opposed the application based on a likelihood of confusion with its own “dove-wing headstock” design, and the Trademark Trial and Appeal Board declined to issue Concordia a trademark. Thereafter, the parties attempted, but failed, to draft license agreements allowing Armadillo to use Gibson’s shapes.
In May 2019, Gibson sued Armadillo and Concordia, asserting trademark infringement and counterfeiting claims and alleging that the defendants infringed four of Gibson’s trademarked guitar body shapes, one trademarked guitar headstock shape and two Gibson word marks. Armadillo asserted a counterclaim seeking cancellation of the Gibson trademarks based on genericness and also asserted genericness defenses based on third-party use.
Prior to trial, Gibson filed an omnibus motion in limine, seeking, among other things, exclusion of all evidence relating to advertisements or sales of third-party guitars prior to 1992. Gibson argued that third-party-use evidence at trial should be restricted to the five-year period preceding Armadillo’s acquisition of the Dean brand in 1997.
The district court granted Gibson’s motion and excluded third-party-use evidence based on Federal Rule of Evidence (FRE) 403 and the Federal Circuit’s 2018 decision in Converse, Inc. v. Converse, which held that third-party use “older than five years should only be considered relevant if there is evidence that such uses were likely to have impacted consumers’ perceptions of the mark as of the relevant date.” As to FRE 403, the district court held that “concerns regarding judicial efficiency and possible confusion with the jury compel[led]” adherence “to a cut-off date” for third-party-use evidence.
After trial, the jury returned a verdict largely in Gibson’s favor, finding, among other things, that Armadillo had infringed various Gibson trademarks and marketed counterfeits of a Gibson guitar body shape, and that none of the Gibson body shapes’ marks should be cancelled due to genericness. The district court granted a permanent injunction against Armadillo and Concordia prohibiting the sale of the infringing products. Both parties appealed.
On appeal, Armadillo argued that the district court abused its discretion by excluding decades of third-party-use evidence that predated the registration of the Gibson marks. According to Armadillo, because this evidence was critical to both its primary defense of genericness and its counterclaims seeking cancellation of the marks, the district court’s wholesale exclusion of this evidence was not harmless error.
The Fifth Circuit agreed with Armadillo and reversed. The panel held that the district court abused its discretion in excluding all evidence of third-party use of the Gibson trademarks prior to 1992. Although the abuse of discretion standard of review for evidentiary matters is an exacting one, the Fifth Circuit panel determined that Armadillo cleared that threshold by demonstrating substantial prejudice resulting from the exclusion of evidence.
The panel disagreed with the district court’s reliance on both Converse and FRE 403 in its analysis. As the panel noted, the court in Converse did not assess third-party-use evidence older than five years for purposes of a genericness analysis. Instead, Converse addressed whether that evidence constituted the best evidence to determine whether a mark attained secondary meaning in the minds of consumers. Converse did find that third-party uses older than five years could be considered relevant if evidence existed that those uses were likely to have impacted consumers’ perceptions of the mark as of the relevant date of the alleged infringement. Converse also did not mandate a strict five-year limit for evidence of third-party use, as the lower court appeared to hold; instead, it instructed courts to examine whether older evidence is relevant.
Here, Armadillo was prepared to present evidence that older third-party uses of the guitars’ body shapes dating back to the 1960s would have affected consumers’ perception of those shapes, such that they were either generic or “incredibly weak.” Because the district court excluded this evidence without providing Armadillo with the opportunity to demonstrate its relevance, the Fifth Circuit held that the district court abused its discretion.
The Fifth Circuit likewise held that the district court abused its discretion by excluding this evidence based on FRE 403. The panel acknowledged that exclusion of evidence under FRE 403 is an extraordinary remedy requiring a district court to balance the probative value of the proffered evidence with the risk of countervailing concerns. As to probative value, the Fifth Circuit observed that the district court already determined that Armadillo’s third-party-use evidence was relevant at several points during pretrial discovery. As to prejudice, the panel held that exclusion of the evidence caused substantial prejudice to Armadillo’s primary defense. In particular, as a result of the court’s ruling, the jury heard little to no evidence supporting Armadillo’s theory that the marks were generic at the time of Gibson’s registrations. That defense was critical, the panel held, because generic marks are categorically excluded from trademark protection, as Armadillo argued should be the case with the Gibson marks. Accordingly, in the panel’s view, FRE 403 provided no basis for the wholesale exclusion of pre 1992 third-party-use evidence.
Because the genericness defense and counterclaim could not be divorced from the rest of the issues on appeal, the panel determined that a new trial was the appropriate remedy.
Summary prepared by Frank D’Angelo and Elena De Santis