Old Tricks/Bad Habits at the Federal Circuit

The Federal Circuit used to get a lot of flak for failing to defer to factual findings of the tribunal it was reviewing. My (highly unscientific) sense is that such criticism has eased somewhat, but I was reminded of it when reading the court's recent precedential decision in Rai Strategic Holdings v. Philip Morris Products, reversing a PTAB's finding of inadequate written description. 2022-1862, 2024 WL 500642 (Fed. Cir. Feb. 9, 2024). The court itself described the case as turning on “unique facts” and whether skilled artisans would consider the claimed range “to be part of the invention.” It sure seems like the Federal Circuit is up to its old tricks.

The issue in Rai was whether the inventors adequately possessed a “heating member . . . having a length of about 75% to about 85% of a length of [another recited structure.]” The specification disclosed ranges of about 75-125%, about 85-115%, and about 90-110%. The Federal Circuit walked through precedent reaching different outcomes depending on whether “a skilled artisan would have understood the [underlying] application as disclosing an invention with the range between these endpoints.” As that discussion demonstrates, simply describing a range in the specification and claiming a narrower portion of that range is not necessarily sufficient to support a written description.

The patent challenger argued that the PTAB's decision was adequately supported by the specification and its expert's testimony focusing on the difference between the midpoint of the ranges from the specification (100%) and that recited in the claim (80%). From this difference, the expert concluded that a skilled artisan would not conclude that the inventors possessed a range that went no higher than 85%. The Federal Circuit responded by pointing to a prior decision rejecting a “mere comparison of ranges” and the application of “mechanical rules [as] a substitute for an analysis of each case on its facts.” But that testimony was an “analysis” grounded in the “facts.” Similarly, the Federal Circuit distinguished what it characterized as the “predictable art[]” in this case from that at issue in another case. As the court acknowledged elsewhere in its opinion, however, the weight to give the relative predictability of the technology at issue presents a question of fact that should be the province of the factfinder (here, the PTAB).

I'll leave it at this: If, as the Federal Circuit acknowledged, its “determination [in Rai] is highly factual and dependent on ‘the nature of the invention and the amount of knowledge imparted to those skilled in the art by the disclosure,'” aren’t those exactly the circumstances that should prompt the Federal Circuit to defer to the trier of fact? 

In so holding, we note that our determination is highly factual and dependent on “the nature of the invention and the amount of knowledge imparted to those skilled in the art by the disclosure.”

passle-net.s3.amazonaws.com/...