What does the CAFC do to establish clear error? Or is clear error review needed here?

As  to the Supreme Court directions in Teva v. Sandoz, the CAFC must review the factual underpinnings of the district court decision for clear error..  Or does it?


PatentDocs (Kevin Noonan) described some of the issues in the Supreme Court requirements for review:


--

Now, in its Teva decision the Court has decided that appellate review of claim construction is limited by Federal Rule of Civil Procedure 52(a)(6), wherein Federal Circuit review of "subsidiary" facts "underpinning" claim construction must be reviewed with deference to the district court and overturned only where there is clear error in the court's decisions:

In cases where those subsidiary facts are in dispute, courts will need to make subsidiary factual findings about that ex­trinsic evidence.  These are the "evidentiary underpin­nings" of claim construction that we discussed inMarkman, and this subsidiary factfinding must be reviewed for clear error on appeal.

This will arise when "the district court will need to look beyond the patent's intrinsic evidence and to consult extrinsic evidence in order to understand, for example, the background science or the meaning of a term in the rele­vant art during the relevant time period."  See, e.g., Seymour v. Osborne, 11 Wall. 516, 546 (1871).

Such deference is not absolute, however:  the opinion states that where the district court's construction is limited to the intrinsic evidence, i.e., the "plain meaning" of the claim language, the specification and ("when in evidence") the history of prosecution in the Patent and Trademark Office, then the appellate court's review can be plenary and give no deference to the district court's construction.  It is only when there is "extrinsic" evidence (which, up until now has been disfavored; Philips v. AWH) must the Federal Circuit tread lightly on what was decided below and adhere to the deferential standard of clear error.

--

An issue is that Teva made the same argument about Mp to the examiner during prosecution (intrinsic) AND to the court via expert witness Dr. Grant (extrinsic).  Does the fact that the same argument made in prosecution was made a second time elevate the standard of review from plenary to clear error?  The argument by Teva is false because Mp is a mode and is not a species of "average molecular weight."  The invocation of Mp does not offer written description support for the claim element "average molecular weight."

Of the former argument during prosecution:


During prosecution of the '539 patent, the Examiner rejected [] pending claims as indefinite, stating that "the term 'average molecular weight' . . . is indefinite since its method of measurement is not specified, i.e. [M_n], [M_w] . . . etc." J.A. 3245. Teva stated in its response that "[o]ne of ordinary skill in the art, upon reviewing the specification, would understand that 'average molecular weight' refers to the molecular weight at the peak of the molecular weight distribution curve shown in Figure 1," i.e., M_p. J.A. 3258. The claims were allowed. During prosecution of the '847 patent, the Examiner made an analogous rejection over the same claim term, stating that "the term 'average' molecular weight . . . is meaningless as a limitation without specifying its basis, e.g., [M_w], [M_n], etc." J.A. 3220. Teva overcame the rejection by responding that "[o]ne of ordinary skill in the art could understand that kilodalton units implies [sic] a weight average molecular weight," i.e., M_w. J.A. 3229. The only basis upon which the Examiner could have agreed that the'539 patent claims were not indefinite was that "molecular weight" means M_p. In contrast, the only basis for the Examiner's withdrawal of the indefiniteness rejection of the '847 patent claims  [] was that the same term means M_w. Teva's two definitions cannot be reconciled.



The argument as to the '847 was later acknowledged to be false.

Of the '539 argument, the CAfC noted that Mp was not an "average molecular weight."

Is this observation sufficient to establish that Dr. Grant's assertion, as an expert witness credited by the district court,  is clearly erroneous?  If not so, what fact-finding must the CAFC engage to establish the error of Dr. Grant?