What Is the Appeal of Using Dictionaries in Patent Litigation?

Patent litigators should not be afraid of dictionaries. Although two notorious cases—Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005) and Teva Pharms. USA, Inc. v. Sandoz, Inc., 574 U.S. 318 (2015)—may have thrown suspicion at the use of dictionaries in claim construction, a more comprehensive analysis of patent case law reveals that the rumors of dictionaries’ demise are greatly exaggerated.

Federal Circuit’s Hospitable pre-Phillips Approach to Dictionary Definitions

The Federal Circuit has always recognized the important role dictionaries play in claim construction. Before Phillips, in Texas Digital Systems, Inc. v. Telegenix, Inc., 308 F.3d 1193 (Fed. Cir. 2002), the court explained that dictionaries are particularly useful in claim construction because they are “publicly available at the time the patent is issued,” are “reliable sources of information on” “established meanings,” and are “not inspired by litigation.” For those reasons, the court decided that dictionaries are not “extrinsic evidence” and can be used by “trial and appellate judges” “at any stage of a litigation, regardless of whether they have been offered by a party in evidence.”

Texas Digital warned that dictionary use is not without limits. As was standard practice by that time, dictionary definitions inconsistent with the intrinsic record—e.g., the language of the patent—had to be rejected. This was an approach adopted by the Federal Circuit from contract law—which likewise emphasized contractual language over external evidence. But, just as in contract law, the prohibition of using extrinsic evidence to “contradict” the intrinsic evidence is fairly toothless in practice. The fact that the court decides that a dictionary definition is “inconsistent with the intrinsic record” still means it considered that definition in the claim construction analysis. The dictionary definition influenced the claim construction process, even if it was ultimately rejected. In all, the Federal Circuit’s longstanding approach to dictionaries, as of the time of Texas Digital, was relatively expansive.

Effects of Phillips and Teva on Claim Construction Analysis

Two significant claim construction cases arguably obfuscated this open dictionary landscape: Phillips and Teva.

In Phillips, the en banc Federal Circuit downgraded the importance of dictionaries. The court derided them as “less reliable” “extrinsic evidence”—“evidence external to the [intrinsic record of] patent and prosecution history.” The court directed judges to instead look at intrinsic evidence for context, before consulting “extrinsic sources such as dictionaries.”

The Supreme Court in Teva subsequently pronounced that a district court’s “consult[ation of] extrinsic evidence” to “make a subsidiary factual finding” in claim construction “must be reviewed for clear error.” Before Teva, the Federal Circuit treated the entire claim construction inquiry as purely legal, with no deference for subsidiary factfinding. Teva therefore appeared to impose a new deference penalty on the use of extrinsic evidence—which could include dictionaries—in claim construction appeals.

After Phillips and Teva, it became unclear whether introducing competing dictionary definitions transformed the primarily legal claim construction inquiry into a factual dispute, requiring deferential review on appeal. Such a rule would have at least two significant consequences for patent litigation. First, parties would try to frame dictionary disputes as factual ones in district court, potentially increasing litigation costs. Second, by potentially creating another category of deference in claim construction appeals, it could limit the Federal Circuit’s ability to provide the final, authoritative interpretation of a patent.

Dictionaries: Business as Usual After Teva

Despite the uncertainty Phillips and Teva introduced, the use of dictionaries in claim construction has remained largely unchanged from the Texas Digital era. And there are several good reasons to think that neither Phillips nor Teva should scare patent litigants away from relying on dictionaries—at either the trial or appellate levels—going forward.

First, Phillips did “not” “preclude the appropriate use of dictionaries” and reiterated that dictionaries can be used by judges “at any time” even in the initial process of “reading a patent.”

Second, Teva said nothing about dictionaries, and generally classified narrowly what it considered a “factual finding” for claim construction purposes. On remand, the Federal Circuit emphasized this limitation of Teva’s holding—that extrinsic evidence does not automatically “transform into a factual matter” the legal inquiry of determining “the meaning” of a patent term.

Third, an examination of prior Supreme Court decisions confirms that Teva was not meant to be a sea change for how dictionaries are used in legal document interpretation. The Supreme Court has always freely turned to dictionaries when interpreting statutes and contracts. It views them not as evidence but as a legal tool of interpretation. And in the claim construction context in particular, Teva did not disturb the Supreme Court’s decision in Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996), which similarly emphasized the interpretive and primarily legal character of claim construction.

Fourth, the Federal Circuit has continued to treat dictionaries as a routine part of a de novo, legal (not factual) analysis of claim construction after Teva. Recent opinions make no mention of deference for any district court factual findings with respect to dictionary definitions. In fact, many post-Teva opinions do not take note of whether proposed dictionary definitions had been presented to the district court or for the first time on appeal. Or, as commentators have noted, a case might refer to Teva’s clear error standard in general but not specify any standard of review for a party challenging a district court’s use of a particular dictionary definition. Instead, whether the Federal Circuit identifies “error” in this type of analysis typically turns on whether a dictionary definition is inconsistent with the Federal Circuit’s interpretation of the intrinsic record—the same purely interpretive legal exercise that Texas Digital endorsed before Phillips and Teva. Accordingly, by all indications, the treatment of dictionary definitions in claim construction analysis remains unchanged.

Conclusion

In sum, there is little downside in citing to dictionary definitions in claim construction disputes. At best, the court will incorporate that definition into its construction, and at worst, the court will consider it but not use it. And the Federal Circuit, for now at least, shows no signs of analyzing competing proposed dictionary definitions as factual disputes or deferring to district courts’ uses of those definitions. When it comes to dictionaries, it’s “business as usual at the Federal Circuit.”