Here's what the AWH petition had to say -

The district court’s claim construction order included numerous findings of fact about the state of the art at the time of the invention.

Phillips appealed the district court’s summary judgment of non-infringement to a panel of the Federal Circuit. The panel reviewed the district court’s claim construction de novo, pursuant to the Federal Circuit’s prior ruling in Cybor Corp. v. FAS Techs., Inc., 138 F.3d 1448, 1456 (Fed. Cir. 1998) (en banc).

Phillips filed a timely petition for rehearing en banc in the Federal Circuit. The Federal Circuit granted Phillips’ petition for rehearing en banc, posed seven questions that it sought to resolve, and invited participation by amici curiae.

Question 7 is the current rub.

7. Consistent with the Supreme Court’s decision in Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996), and our en banc decision in Cybor Corp. v. FAS Technologies, Inc., 138 F.3d 1448 (Fed. Cir. 1998), is it appropriate for this court to accord any deference to any aspect of trial court claim construction rulings? If so, on what aspects, in what circumstances, and to what extent?

In response to the en banc court’s invitation, both parties argued that the Federal Circuit should afford some deference to a district court’s claim construction. In addition, thirty-three groups filed amicus briefs. Of the twenty-three that took a position on Question 7, all but six urged the Federal Circuit to at least acknowledge the factual components of claim construction and accord deference to district courts on such components. Those supporting a deferential review of claim construction included the American Bar Association, the American Intellectual Property Law Association, the Federal Circuit Bar Association, the Intellectual Property Owners Association and the United States Patent and Trademark Office (“USPTO”).

The majority then proceeded to apply de novo – on appeal – the new rules of claim construction it had crafted and reversed the ruling of the district court in favor of AWH. The en banc majority did not remand the case to the district court for application of the principles or analyze whether the district court had, in fact, followed those principles in its claim construction analysis. Instead, it applied those principles itself to the ‘798 patent.

Judge Newman and Mayer expressed disgust at the en banc majority ruling.

Judge Mayer further decried his own court’s “[persistence] in the delusion that claim construction is a purely legal determination, unaffected by the underlying facts.” App. at 58. To deny the factual components of claim construction was, according to Judge Mayer, to ignore the reality that claim construction “is, or should be, made in context: a claim should be interpreted both from the perspective of one of ordinary skill in the art and in view of the state of the art at the time of the invention.” Id. These inquiries are “inherently factual.” Id. Indeed, the claim construction process includes a great number of inherently factual inquiries, including:
• who qualifies as one of ordinary skill in the art
• the meaning of patent terms to that person
• the state of the art at the time of the invention
• contradictory dictionary definitions and which would be consulted by the skilled artisan
• the scope of specialized terms
• the problem a patent was solving
• what is related or pertinent art
• whether a construction was disallowed during prosecution.

Over ten years ago, the Federal Circuit held that “the interpretation and construction of patent claims, which define the scope of a patentee’s rights under the patent, is a matter of law exclusively for the court.” Markman v. Westview Instruments, Inc., 52 F.3d 897 (Fed. Cir. 1995) (“Markman I”), aff ’d in part, 517 U.S. 370 (1996) (“Markman II”). Three years later, further solidifying its place as the final arbiter of all things related to claim construction, the Federal Circuit announced that it was permitted to “review claim construction de novo on appeal including any allegedly fact-based questions.” Cybor, 138 F.3d at 1456 (emphasis added).

The nut of AWH's argument is that, because claim construction has a significant factual component, it should not be subject to wholesale review by the appeals court.

The Federal Circuit’s rule of universal de novo review for patent claim construction is founded on a fallacy and conflicts with both the Rules of Civil Procedure and this Court’s precedents. The reality is that patent claim construction often and necessarily rests on findings of fact made by the trial court on disputed evidence. But, as the Federal Circuit’s own Judge Mayer wrote, that court’s rule of review rests on “the falsehood that claim construction is a matter of law devoid of any factual component.” App. at 54. The Federal Circuit’s persistent refusal to review trial courts’ factual findings deferentially conflicts with Federal Rule of Civil Procedure 52(a), which requires deferential review on appeal for findings of fact. It also conflicts with the decisions of this Court, which have acknowledged that claim construction involves mixed questions of law and fact and have held that deferential review should be applied on appeal to factual findings in other circumstances of patent claim interpretation.

AWH mentions that there's a 35% reversal rate by the CAFC in patent cases, three times that of other circuits. Though AWH may like to make something of that, that statistical evidence provides no fodder to back AWH's argument. Maybe the reversal rate is so high because the district court gets it wrong so often, thus justifying de novo review.

This Court has already concluded that patent claim construction involves factual determinations, and the Federal Circuit’s insistence to the contrary is in direct conflict with the Court’s precedent. In Graham v. John Deere Co., 383 U.S. 1 (1966), the Court directly acknowledged that the determination of what a patent teaches to one skilled in the art – an essential part of the claim construction analysis – is an issue of fact. As the Court stated, “While the ultimate question of patent validity is one of law,” the determination of patent validity “lends itself to several basic factual inquiries [such as] the scope and content of the prior art . . . differences between the prior art and the claims at issue . . . and the level of ordinary skill in the pertinent art.” Id. at 17. In Markman II, the Court likewise acknowledged that the claim construction process is a “mongrel practice” that has “evidentiary underpinnings.” 517 U.S. at 378, 389. And the Court directly acknowledged that “[i]t is, of course, true that credibility judgments have to be made about the experts who testify in patent cases . . . .” Id. at 389.

The same “functional considerations” that the Court applied in Markman II to conclude that judges rather than juries should decide the issue of claim construction support deferential appellate review of the judges’ fact finding – because trial judges rather than appellate judges are better situated to evaluate and make findings based on the evidence. Indeed, it is the ordinary practice for appellate courts to accord deferential review to factual findings made by trial judges – even when those findings are embedded in questions of law that are themselves reviewed de novo. See, e.g., Ornelas v. United States, 517 U.S. 690, 699 (1996); Cooter & Gell v. Hartmarx Corp., 496 U.S. 384, 401-05 (1990); Pierce v. Underwood, 487 U.S. 552, 560-63 (1988).

Judge as opposed to jury decision is orthogonal to higher to lower court deference. AWH's above argument in that regard is logically disjoint.

The en banc Federal Circuit, however, did not even discuss the district court’s factual findings, let alone grant them deference. Instead, the en banc court admitted that it was merely substituting its own findings for those of the district court, suggesting that they would have withstood deferential review.

Revisionist history. Phillips v. AWH was about the scope of the claim term "baffles".

In the first round, the district court had tried to use a means-plus-function claim format for claim construction; a legal error.

In the first appeal, the majority adopted a restrictive definition of "baffles", dragging a described embodiment from the specification into the claim scope when it shouldn't have. The claim construction issue wasn't really one of factual dispute, but how restrictive a definition to grant based upon a priority of evidence consideration, and thereby in interpretation of statutory law.

The dissenting judge argued that the panel had improperly limited the claims to the particular embodiment of the invention disclosed in the specification, rather than adopting the “plain meaning” of the term “baffles.”

The en banc decision basically ended up where the dissenting judge had in the first appeal.

In the en banc appeal, AWH made a purely legal argument, that a broad interpretation would have de facto rendered the claim invalid in light of the prior art; again, there was no significant factual dispute in the core claim construction issue.

Invoking the principle that “claims should be so construed, if possible, as to sustain their validity,” Rhine v. Casio, Inc., 183 F.3d 1342, 1345 (Fed Cir. 1999), AWH argues that the term “baffles” should be given a restrictive meaning because if the term is not construed restrictively, the asserted claims would be invalid.

Here's my devil's advocate to AWH -

Without significant technical/factual tussle, Phillips v. AWH does not provide a substantive foundation for the Supreme Court to address the issue of appeal court deference, precisely because such deference was not at issue in the case, because technical facts regarding claim construction where not in dispute.

There's no doubt that claim construction has a significant factual component. That's why Phillips v. AWH is so important in providing a priority stack in evaluating evidence related to claim construction - it provides a judicial method for legal determination of factual content. What is being determined during claim construction is not a dispute of technical fact per se, but an interpretation of scope applied to a word or phrase, where that word or phrase may be a technical term, but the dispute is only because of conflicting evidence at hand, not really because of actual factual dispute. It's not like claim construction is ever arguing evolution versus creationism. Phillips v. AWH boiled down to how broadly to define "baffles" based on the evidence at hand - does ordinary meaning trump an embodiment in the specification when the claim language does not require reference to the specification for the term to be understood?

With a solid method for evidentiary precedence, practically any claim construction dispute can be decided as a matter of law. The facts should be in evidence (if not, its a legal problem of definiteness); if so, it's simply a matter of picking which set of facts are applicable given the way the claim was drafted. Without a evidentiary priority stack, that process is haphazard - hence all the claim construction reversals prior to Phillips v. AWH.

Patent claims are a mongrel of technical fact in legal wrapping. But that doesn't mean that claim construction is a matter of fact.