Page:ASTM v. PRO (D.C. Cir. 2018).pdf/21

From Wikisource
Jump to navigation Jump to search
This page has been proofread, but needs to be validated.
AMERICAN SOC. FOR TESTING v. PUBLIC.RESOURCE.ORG
Cite as 896 F.3d 437 (D.C. Cir. 2018)
457

likely, the nominative fair use defense is inapplicable and the court need not assess each of the [] factors.” ASTM, 2017 WL 473822, at *23. The district court’s failure to consider the three nominative fair use factors, as we shall explain, was error.

Courts of appeals have disagreed about how exactly to evaluate nominative fair use claims. The Third Circuit, for instance, treats nominative fair use as an affirmative defense to infringement. See Century 21, 425 F.3d at 222. The Second and Ninth Circuits, by contrast, treat nominative fair use as a means for evaluating, for purposes of determining trademark infringement, whether there is any likelihood of confusion at all. See International Information Systems Security Certification Consortium, Inc. v. Security University, LLC, 823 F.3d 153, 167 (2d Cir. 2016) (IISSC); New Kids, 971 F.2d at 308. And, even the courts of appeals that agree about when to test nominative fair use disagree about how precisely to apply the factors. On the one hand, the Ninth Circuit had held that the three nominative fair use factors supplant the ordinary multi-factor likelihood of confusion test. New Kids, 971 F.2d at 308. On the other, the Second Circuit, although “recogniz[ing] that many of the [likelihood of confusion] factors are a bad fit” for nominative fair use cases, has held that the three factors should be considered in addition to the ordinary likelihood of confusion factors. IISSC, 823 F.3d at 168.

The parties have not briefed, and we need not resolve today, which approach our court should adopt. What we can say is that under no formulation can a court ignore the nominative fair use factors altogehter. Where, as here, there is a claim of nominative fair use, the likelihood of confusion analysis remains incomplete without at least some discussion of these factors. Indeed, the particulars of this case show just how consideration of these factors can provide valuable insight both into whether trademark infringement has occurred and, if so, how broad a remedy is needed to address the injury.

Consider the first factor, whether the work is readily identifiable without use of the mark. Assuming that PRO may reproduce some of the technical standards under copyright’s fair use doctrine for the purpose of informing the public about the law, it is hard to see how PRO could fulfill that goal without identifying the standard by its name—the very name also used in the incorporating law.

Likewise, as to the second factor—whether only so much of the mark is used as is reasonably necessary to identify the product—it may well be that PRO overstepped when it reproduced both ASTM’s logo and its word marks but, as it told the district court, it is not wedded to using the logo. See Transcript of 9/12/16 Motions Hearing at 116, ASTM, No. 13-cv-1215 (TSC) (D.D.C. Oct. 18, 2016), Dkt. No. 173, J.A. 3374 (“Public.Resource would take direction from this Court. Logos: yes or no? [PRO] doesn’t care.”). Thus, accounting for this factor may suggest ways of crafting a narrower remedy that better balances the parties’ competing interests here.

Finally, as to the third factor—whether the user has suggested sponsorship or endorsement by the trademark holder—the disclaimers PRO appends to many of its copies of the standards may well fail to adequately eliminate the possibility a consumer would assume sponsorship or endorsement by ASTM, but that hardly means that no disclaimer could cure that risk. Indeed, at oral argument, PRO suggested that it would be “more than happy to modify the disclaimers.” Oral Arg. 24:06–19. And although the disclaimers initially used by PRO were quite barebones, the record contains examples of more fulsome disclaimers it later appended to at