Nearly 20 years ago, the Supreme Court (in Markman v. Westview Instruments) considered whether the construction of a patent claim during litigation is a matter of law reserved entirely for a federal court judge, or subject to a Seventh Amendment guarantee that a jury undertake that exercise. The Court unanimously held “that the construction of a patent, including terms of art within its claim, is exclusively within the province of the court.” District courts have followed that holding by conducting hearings (in the absence of juries) in which the litigants may present factual evidence and expert testimony, and weighing that evidence in view of the litigants’ arguments before construing claims for trial of infringement and invalidity.
Shortly after the Court’s unanimous holding, a divided en banc panel of the Court of Appeals for the Federal Circuit (in Cybor Corp. v. FAS Technologies, Inc.) concluded that a district court’s claim construction must be reviewed, on appeal, for correctness as a matter of law (de novo). And, in late February 2014, a different, divided en banc panel of the Federal Circuit (in Lighting Ballast Control LLC v. Philips Electronics North America Corp.) confirmed the propriety of the Cybor standard of appellate review and refused to replace that standard with one that would offer deference to a district court’s claim construction.
The debate dividing the Federal Circuit is not likely over, although it should be. Many, including those in dissent, believe that the Cybor standard is not consistent with Markman. While Markman may well support the holding in Cybor that the ultimate question of the scope of a patent claim is a question of law, that does not mean that a district court’s subsidiary factual findings underlying its construction may be reviewed de novo on appeal.
There are of course instances (many, indeed) where the scope of a patent claim may be discerned by simply reviewing the patent itself. In those instances, the circuit court is as capable as a district court in properly construing the claim scope — and, likely better capable because of its exclusive appellate jurisdiction over patent disputes and its vast experience (relative to district courts) in construing claims. De novo review in those instances promotes the consistent application of the patent laws and serves as a necessary check against the prospect that two different district courts may construe the same patent differently. But, opponents of the Cybor standard theorize that de novo review is a mistake in other instances where factual findings regarding, for example, the level of ordinary skill in the pertinent art of the patented invention, the definition of that art, and that art’s accepted meaning of certain claim terms are relevant to the construction. In those other instances, they argue, a district court’s fact-finding ought to be owed the same deference as any other factual findings in accordance with the Federal Rules of Civil Procedure, and nothing in Markman suggests otherwise.
Markman did not address the allocation of authority between trial and appellate court judges, so that decision does not necessarily require a de novo standard of review of all district court findings underlying claim construction. But, as explained below, its practical effect counsels that review standard. And, despite opportunities since Cybor to address that issue, the Supreme Court has consistently declined to do so. For example, it has denied certiorari petitions in cases involving no factual findings or credibility determinations, as well as in cases involving extensive factual findings and credibility determinations, where the standard of review might have influenced the outcome on appeal.
Perhaps the Court knew full well of what it was deciding. But, if it did not, then, according to Cybor’s proponents, belatedly repudiating the Cybor standard now would likely cause far more turmoil than good. Thousands of disputes have been resolved in view of the Cybor standard, and there appears to be no practical alternative to that standard. Further, it is not certain that the Court is capable of offering an alternative standard, as it has repeatedly failed to provide one in the context of determining a standard for deciding patent eligible subject matter.
Cybor’s opponents prefer a review standard affording deference to factual findings underlying a claim construction. But, applying that standard would make it impossible to reconcile meaningfully inconsistent results in different cases interpreting the same patent if those results were deemed to be based on supposed factual findings entitled to deference. Not every patent is the subject of multiple lawsuits in different jurisdictions, but enough of them are where this is hardly an isolated concern.
Set aside (or accept, as you must) the inevitable satellite litigation that such deference would spawn. Given the possibility of different results based upon which among the nearly 700 district judges interprets the patent, a deferential standard of appellate review would incentivize the type of forum shopping that Congress created the Federal Circuit to eliminate. When the same judge presides over separate trials involving the same patent claim, each accused infringer may persuade the judge of a different claim scope supported by different factual evidence. Thus, not only are there impediments to judges affording one another comity in dealing with the same patent, judges would not be able to defer to their own prior determinations regarding the same patent.
Interpretation of a patent through claim construction is practically no different than interpretation of sentences in any other legal document (e.g., statutes, contracts, etc.), which is subject to de novo standard of review. That review standard applies in many situations outside of patent litigation, where a district court must interpret facts, often animated by expert testimony. For example, interpretive issues of contracts and statutes are reviewed de novo on appeal, even if the district court must weigh competing expert testimony on the subject. Claim construction may well be less fact-intensive than contract- or statutory-interpretation. The latter two entertain factual consideration of the contracting parties’ or legislature’s intent. In claim construction, however, the patentee’s intent is of no import and, instead, the objectivity of the ordinarily-skilled artisan is key.
There is no doubt that the definitions of the artisan and the art are among the factual determinations that underlie the construction just as they underlie certain legal conclusions of patent validity (e.g., obviousness, enablement). But, there is minimal (if any practical) conflict between the de novo standard of review for claim construction and a deferential standard of review of the most pertinent factual issues underlying determinations of patent validity. When trying validity, the most pertinent questions are of historical fact — i.e., as of the patent’s filing date what did the ordinarily-skilled artisan know, or could that artisan have practiced the invention based on what the patent teaches? — and the answers are precisely the type for which deference is likely suitable. And, if there is a conflict, then practically a circuit judge is just as capable of resolving the fact as a district judge.
The principles of deference here do not, however, apply in the context of claim construction, which Markman held is solely within the judge’s province. And, even if the deference principles could be argued to apply, we can look to disputes outside of patent law to appreciate that circuit judges are not only capable of, but also often charged with, de novo review of facts. For example, factual findings underlying conclusions of constitutional violations (e.g., free speech, due process, unreasonable search/seizure) are routinely reviewed de novo, even if that review involves interpretation of complicated factual situations. Because de novo review of those findings is unexceptional, it is hardly offensive to apply the same de novo standard of review to facts underlying claim construction.
Opponents of the Cybor standard hypothesize various litigation inefficiencies attendant in de novo review. But, there is hardly any shortage of solutions to those inefficiencies that do not require changing a standard of review that has been used to resolve thousands of disputes over the past fifteen plus years and provide uniformity to the application of the patent laws. If claim construction is indeed outcome determinative, litigants can set their egos aside, stipulate to invalidity or infringement, reserving their right to appeal, and jointly moving the district court to bifurcate infringement liability and damages determinations. Plowing ahead to adjudicate issues that are moot if the construction of a claim is reversed is as suspect as hypothesizing the inefficiencies attendant with the Cybor standard.
It is not clear that the Cybor standard is practically wrong. And even if it is, we now have a significant and established collection of case law applying that standard in view of the Supreme Court’s repeated refusal to review that standard and Congress’s inaction. The Federal Circuit’s reliance on stare decisis to maintain that standard can hardly be faulted. Recently, in oral arguments in another patent law case (about attorney’s fee awards) before the Court, the Chief Justice asked, “Why shouldn’t we give some deference to the decision of the court that was set up to develop patent law in a uniform way? They have a much better idea than we do about the consequences of these fee awards in particular cases.”
The Court’s review of the Cybor standard may well be inevitable in view of the Federal Circuit’s recently articulated division over the propriety of that standard. Cybor’s proponents can hope that the Supreme Court posits the same question, recognizes the practical benefits of that standard, and confirms that standard is correct. Cybor’s opponents, on the other hand, should try to find a practical alternative.