The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
I'm sorry to say I know very little about patent law, but I know enough to know it's important, and to wish I knew more about Tuesday's Teva Pharmaceuticals v. Sandoz. Teva involved this question: When a trial court interprets a patent claim, and the case is appealed, should the appellate court—the U.S. Court of Appeals for the Federal Circuit—review that interpretation "de novo" (i.e., afresh, with no deference given to the district court's judgment) or for "clear error" (i.e., setting aside the district court's decision only if the Federal Circuit thinks it's clearly erroneous). Legal questions are generally reviewed de novo; factual questions are generally reviewed for clear error; "mixed questions of law and fact," which is to say questions of application of law to fact, are … well, it's complicated; so the issue here came down to how one categorizes claim construction. This is one of those procedural questions on which a lot of substance often ends up turning.
A patent contains "claims" that, per the Patent Act, "point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention." 35 U.S.C. § 112(b). Patent claims, accordingly, delineate the scope of inventions that are—and, conversely, are not—covered by the patent.
Interpreting the meaning of the claims—and thus the scope of the patent—is often a critical question in patent litigation. For example, whether a defendant's products or processes infringe a patent will often turn on the meaning of claim terms. Likewise, when a defendant asserts that a patent claim is invalid because it is either obvious or anticipated in light of prior art, the meaning of the terms is often controlling.
Interpreting claim language is called "claim construction." Two decades ago, in Markman v. Westview Instruments, Inc., the Supreme Court held that claim construction is a job for the judge, not a jury.
To make this determination, a court is to view claim language through the lens of a "person of ordinary skill in the art" at the time of the invention. That is, claims are interpreted as they would be by a hypothetical person who has typical technical expertise in the relevant field at the time the invention was developed, not as a lay individual would understand the language today. In conducting claim construction, a court may consider evidence "intrinsic" to the patent (including the claim terms themselves, the patent's specification, and the prosecution history, i.e., the back and forth correspondence with the Patent Office leading to the patent's issuance), as well as evidence "extrinsic" to the patent (including expert testimony or references in the art).
Since Markman, the Federal Circuit has consistently held that it reviews the entirety of a claim construction ruling de novo, including any subsidiary factual determinations. There long existed some tension as to this holding, however, in light of Federal Rule of Civil Procedure 52(a)(6), which directs that an appellate court may not "set aside" a district court's "[f]indings of fact" unless they are "clearly erroneous."
In Teva Pharmaceuticals USA, Inc. v. Sandoz, Inc., No. 13-854, the Supreme Court reversed this approach. The Court concluded there is no "exception from the ordinary rule governing appellate review of factual matters." Instead, "[a] district court judge who has presided over, and listened to, the entirety of a proceeding has a comparatively greater opportunity to gain  familiarity than an appeals court judge" "'with [the] specific scientific problems and principles'" at issue. The Court thus reiterated a familiar theme of its recent intellectual property jurisprudence: IP cases are subject to the same rules that govern universally.
The Court may have been motivated, in part, by its observation—which came by way of its summary of an amicus brief—that the "Federal Circuit overturns district court claim construction at an unusually high rate."
While the full implications of Teva may take some time to surface, there are a few issues that likely will be the subject of future litigation.
How will the Federal Circuit refine the distinction between factual and legal determinations? Following Teva, the Federal Circuit will have to distinguish between aspects of a trial court's claim construction determinations that are factual in nature, compared against those that are legal. In an effort to provide some guidance, the Court suggested what could become something of a bright line rule for review: determinations as to "intrinsic" evidence—as well as "ultimate" determinations of claim construction— are to be reviewed de novo, whereas determinations of "extrinsic" evidence are subject to deferential review.
But in some circumstances, the "intrinsic" and "extrinsic" categories blur, leaving some open questions as to the standard of review. When, for example, expert testimony contradicts technical information contained in the intrinsic record, a court will likely be making a factual determination that considers, in part, intrinsic evidence.
Likewise, if expert testimony expounds upon prior art references contained in the prosecution history, the division between "intrinsic" and "extrinsic" will not be entirely clean. And is a court's understanding of a prior art reference really subject to de novo review if it is contained in the prosecution history, but subject to deferential review if not?
Will the district courts heighten their focus on factual determinations in claim construction? Under Teva, district courts may have an incentive to increase their reliance on factual determinations in order to insulate decisions from de novo review. Indeed, Teva may signal an implicit endorsement of greater use of extrinsic evidence—like expert reports and expert testimony—for claim construction.
Will the Federal Circuit respond by recalibrating the role of intrinsic and extrinsic evidence? If district courts place increased emphasis on extrinsic evidence in claim construction, the Federal Circuit could respond by stressing the primary role of intrinsic evidence. Indeed, if a court places too much weight, in the Federal Circuit's view, on extrinsic evidence, the Federal Circuit is likely to find this an error of law, obviating the role of clear error review. The proper calibration between intrinsic "legal" determinations and "extrinsic" factual determinations—which resides largely in the purview of the Federal Circuit—will now alter the appellate review standard.
In sum, Teva signals a substantial change in appellate review of claim construction. Over the next few years, there may be a push-and-pull between district courts and the Federal Circuit as to the proper allocation of authority for claim construction. If district courts increasingly rely on "factual" determinations in order to achieve deferential review, the Federal Circuit may well respond by narrowing the circumstances in which such extrinsic evidence is permissible, or by devaluing its weight in arriving at the "ultimate" claim construction conclusion—which remains a legal question following Teva. Whether Teva achieves the policy goal suggested by the Supreme Court—reducing the reversal rate of claim construction determinations—will likely turn on how this balance is struck.