Watershed moment or not, an en banc Federal Circuit was awash in discord on the direction analysis should flow when more than one entity performs the steps that constitute infringement of a method patent. In the combined cases of Akamai Technologies, Inc. v. Limelight Networks, Inc. and McKesson Technologies, Inc. v. Epic Systems Corp., a per curiam majority overruled contrary precedent to hold that infringement claims based on those kinds of divided acts of infringement can proceed under a theory of induced infringement. Arguing that the law required the analysis to go a completely different way, the four-judge dissent said induced infringement claims should still require an act of direct infringement by a single entity. Filing a separate dissent, Judge Newman attacked the majority’s reasoning but not its result. Given the continental distance between the Federal Circuit judges on the policy issues involved, this may be an invitation to the Supreme Court to wade into the Great Divide.
The case reflects a problem that often arises in cases involving method claims when one party induces the infringement of a method patent, but more than one entity actually performs the steps that constitute infringement. In Akamai, the named defendant allegedly performed certain steps recited in a multi-step claim and then encouraged others to perform the remaining steps of the claim. In McKesson, the defendant allegedly set up a system where it never performed any of the steps, but encouraged others to do so. In both instances, prior courts rejected the patent holders’ infringement claims because of the Federal Circuit’s 2007 decision in BMC Resources, Inc. v. Paymentec L.P. BMC held that a defendant cannot be held liable for inducing infringement if there is not a single entity guilty of performing the steps that constitute direct infringement.
Overruling that part of BMC, the majority found that it made no sense to limit induced infringement claims to instances requiring a single act of direct infringement if the accused infringer acts with the requisite knowledge. The court focused upon the fact that the statute does not expressly define “infringement” as the conduct of a single entity and also said that linguistic differences between 271(a) and 271(b) show that what was being defined were acts of infringement, not who was the actor. The court also reviewed the legislative history of the 1952 amendments, which recognized the problem of combination patents in a “modern” technical world.
The majority said it need not address the question raised by the parties in McKesson regarding whether a defendant who encourages others to perform the recited steps of a method claim could be held liable for direct – as opposed to indirect – infringement. The majority said this issue wasn’t necessary to resolve either of these cases because their decision created a way for the plaintiffs in both direct and indirect infringement cases to press forward with their claims. Dissenting alone, Judge Newman argued forcefully that the lack of an answer on the question of divided direct infringement will continue to haunt the court as new fact patterns emerge. She also challenged the majority’s opinion as a tortured reading of the statute, but offered the solution of recognizing direct divided infringement claims under traditional notions of apportionment/allocation of remedy in joint tort situations.
A separate dissent authored by Judge Linn and signed by Judges Dyk, Prost and O’Malley argued that the majority was acting like a policy maker rather than an interpretive court. According to this dissent, the statutory language and prior precedent clearly required a single entity to be responsible for performing all steps of a claimed method. The dissenters said Section 271(a) speaks only in terms of a single entity committing “the infringement” in a unitary fashion and that consequently, the majority opinion amounted to a rewriting of the statute with principles from common law of torts and criminal proceedings. This dissent was willing to accept the notion that patent holders might not have a remedy in instances of divided infringement unless the defendant is vicariously liable for the actions of the entity which performs all or the remaining steps of the method.
With colliding tectonics like these, the Supreme Court may feel compelled to get involved, now or in a future fact pattern. Method patent holders whose claims are based on divided acts of infringement will be well served to build their cases, if possible, around the notions of vicarious liability concepts advocated in Judge Linn’s dissent as a bulwark to a possible Supreme Court reversal. Patent holders will also need to seek additional ways to persuade courts to take up the unanswered question of divided direct infringement, marshaling the type of record suggested by Judge Newman. At the same time, targeted defendants will need to push the bounds of the other limitations of indirect infringement (intent and level of knowledge of patent, degree of “encouragement,” etc.), and not rely on the divided nature of the infringement. Until it is clear whether the Supreme Court will weigh in, both coasts of patent litigation remain—as always—at risk for erosion.