Effectively reversing Cybor Corp., which flatly stated that a lower court’s claim construction is a question of law which is to be reviewed de novo by the Fed. Cir., yesterday the Supreme Court held 7-2 that questions of fact resolved by a district court prior to claim construction must be given deference by the Fed. Cir. and only reversed if clearly erroneous.
The basic dispute arose over whether or not the term “molecular weight” was adequately defined — it could be calculated three different ways — or if the information in the specification, including a graph showing the peak average molecular weight met the requirements of s. 112(2). The experts disputed a discrepancy between value shown on the graph’s legend and the value at the actual peak of the curve. Teva’s expert testified that difference was small and was due to the conversion of the data from a chromatogram to the m.w. distribution curve in the figure. The District Court agreed, but the Fed. Cir. rejected these factual findings, without explaining why it was “clearly erroneous.” The Supreme Court vacated the Fed. Cir’s. Judgment of s. 112(2) invalidity and remanded, thus endorsing a “clear error” and not a de novo standard for review.
The ultimate claim construction remains a question of law, and so is reviewed de novo by the Fed. Cir. when it is based solely on the intrinsic record – the claims, the specification and the prosecution history. But the Supreme Court made it clear that the Rule applies to findings of both subsidiary and ultimate facts. Apart from precedent, the S. Ct. found that practical considerations favor clear error review, particularly in patent law, “a field where so much depends upon familiarity with specific scientific problems and principles not usually contained in the general storehouse of knowledge and experience.” [citing Graver Tank v. Linde Air Prods.]. The S. Ct. cited Lightning Ballast for the proposition that Fed. Cir. judges “’lack the tools that district courts have available to resolve factual disputes fairly and accurately,’ such as questioning the experts, examining the invention in operation, or appointing a court-appointed expert.”
The S. Ct. was not sympathetic to arguments that subsidiary factual disputes were rare, but spent two pages reviewing the “complexities” brought on by the Fed. Cir.’s efforts to “treat factual findings and legal conclusions similarly”- that amounted to a conclusion that one round of fact-finding is enough. Of course, there were many more nuances to this important opinion than I have time to discuss here, and I plan to supplement this post tomorrow. One issue raised by the dissent is the potential substitution of one appealable issue, full de novo review, with another – to what extent the claim construction below turned on the resolution of factual issues. The thrills and chills of a new “balancing act” may soon be upon us.