Teva Pharmaceuticals USA, Inc. v. Sandoz Inc. (2015)

Is Deference in Claim Construction Review a Good Thing for the Patent System?

By Andrew Williams --

As we reported earlier today, the Supreme Court held in a 7-2 decision authored by Justice Breyer that an "appellate court must apply a 'clear error,' not de novo, standard of review" to the evidentiary underpinnings of a district court's claim construction determination.  Teva Pharmaceuticals USA, Inc. v. Sandoz Inc., No. 13-854, slip op. at 1-2 (U.S. Jan. 20, 2015).  The Court grounded its decision in Federal Rule of Civil Procedure 52(a)(6), which provides that in matters tried to district court, the court's "[f]indings of fact . . . must not be set aside unless clearly erroneous."  But if the answer was that straightforward, why had the Federal Circuit been applying de novo review to all claim construction determinations since the Cybor Corp decision in 1998?  The answer, as Justice Thomas writing in dissent phrased it, lies in whether claim construction really involves findings of fact.

In arriving at the conclusion that the appropriate review standard should de novo, the Federal Circuit's Cybor Corp. opinion relied heavily on the Supreme Court's decision in Markman v. Westview Instruments, Inc., 517 U.S. 370 (1996) (Markman II).  That opinion did repeatedly note the factual components involved in claim construction, referring to it as a "mongrel practice" and observing that the issues can often fall "between a pristine legal standard and a simple historical fact . . . ."  But, the Supreme Court in Markman II did find "sufficient reason to treat construction of terms of art like many other responsibilities that we cede to a judge in the normal course of trial, notwithstanding its evidentiary underpinnings."  Id. at 390.  In addition, the Markman II Court stressed the public policy of uniformity, because "treating interpretive issues as purely legal will promote (though it will not guarantee) interjurisdictional certainty through the application of stare decisis on those questions not yet subject to interjurisdictional uniformity under the authority of the single appeals court."  Id. at 391.  Finally, and perhaps most importantly, the Federal Circuit emphasized the fact that when the Supreme Court affirmed its Markman I decision, it affirmed a decision that used the de novo standard of review.

Nevertheless, the Court today made clear that Markman II did not create, nor argue for, "an exception to Rule 52(a)."  Instead, Markman II stood for the proposition that, much like the proper construction of a document is a question of law, the proper construction of patent should be a question of law, which is a task for the judge.  And, as the Court explained "sometimes, when a written instrument uses 'technical words or phrases not commonly understood,' those words may give rise to a factual dispute" (citations omitted).  Therefore, the Court continued, "Markman no more creates an exception to Rule 52(a) than would holding that judges, not juries, determine equitable claims, such as requests for injunctions."

However, for Rule 52(a) to apply, claim construction must involve findings of fact.  Justice Thomas, in dissent, concluded that it does not.  We will address the substance of the dissent's argument in a future post.

The Supreme Court also addressed the issue of uniformity, perhaps in response to a criticism raised by the dissent.  As Justice Thomas pointed out, "[t]he need for uniformity . . . weighs heavily in favor of de novo review of subsidiary evidentiary determinations."  This is because, with a deferential standard, there is a risk that different jurisdictions construing the same claim term might result in divergent constructions.  The majority dismissed this notion, however, because "[a]fter all, the Federal Circuit will continue to review de novo the district court's ultimate interpretation of the patent claims."  But the Court cannot be suggesting that if differing jurisdictions reach different claim constructions because of different findings of subsidiary fact, the appellate court can simply ignore those findings without identifying clear error.  Moreover, issue preclusion will only be available in limited circumstances.  Ultimately, in this case, the majority ducked the tough questions by concluding that "subsidiary factfinding is unlikely to loom large in the universe of litigated claim construction."

But is this conclusion true?  Or, more importantly, will this conclusion continue to be true?  First of all, as the dissent points out, the notion of issue preclusion as a safeguard was already rejected by this Court -- in the Markman II decision.  More importantly, if litigants were not necessarily incentivized to raise every possible subsidiary factual issue before, they certainly are now.  The result is that the cost and complexity of claim construction is likely to increase, because all parties will now be encouraged to introduce extrinsic evidence, especially expert testimony, in an attempt to maximize deference to a favorable decision.  It is also easy to imagine that trial courts will end up relying on this "subsidiary factfinding" to insulate their decisions from deferential review.

Another unfortunate consequence of this decision is that it could embolden patent assertion entities seeking to abuse the patent system with overly broad and vaguely worded patent claims -- the so-called "patent trolls."  District courts will be more likely to allow additional discovery related to claim construction, which will only serve to increase the cost of patent litigation.  Moreover, under deferential review, district courts may delay claim construction until the end of a case, virtually eliminating early informed settlement discussions.  As the dissent put it, "today's decision will result in few claim construction decisions receiving precedential effect, thereby injecting uncertainty into the world of invention and innovation."  It is this uncertainty that may encourage accused infringers to settle earlier, thereby playing right into the strategy of the prototypic patent troll.

Teva Pharmaceuticals USA, Inc. v. Sandoz Inc. (2015)
Opinion of the Court by Justice Breyer, joined by Chief Justice Roberts and Justices Scalia, Kennedy, Ginsburg, Sotomayor, and Kagan; dissenting opinion by Justice Thomas, joined by Justice Alito

Comments

[Nancy Linck]

Another issue to consider is how this decision will impact the alleged difference in how the PTO construes claims compared to how district courts construe claims. With the focus shifted to underlying fact-finding, entitled to deference, it seems that alleged difference will grow, at least arguably so.

[Skeptical]

How so, Nancy?

I am not certain I understand the point you are trying to make.

I think the possibility exists, but cannot exist 'from the PTO side' because of how the "fact" issue is sliced along the intrinsic/extrinsic lines.

Remember, that ANYTHING happening at the PTO falls into the intrinsic camp.

If the allure here is to be able to bank on extrinsic facts (for a 'better' court treatment), you may see more avoidance of using PTO mechanisms (hold your water, or ammunition - whichever metaphor you like better - so as to switch the factual impact from intrinsic review to extrinsic review.

It will be interesting to see the ripple effects on those post-grant review mechanisms that put the post-grant items into the file record, as that is a mechanism for switching what normally would be extrinsic items into intrinsic items. Unforeseen consequences and all...

[Jim Demers]

The CAFC in this case revisited the expert testimony, and decided that they preferred the conclusions of Sandoz's expert, in contrast to the District Court's decision that Teva's expert was correct.

Isn't that where the Supreme Court found the error? Claim construction, I believe, can still be reviewed de novo; it's the underlying findings of fact that are guarded by the "clear error" standard.