Lemley, Risch, Sichelman & Wagner on Bilski

Bilski v. Kappos was one of the most anticipated patent cases in recent history. Almost all observers agreed that Bilski's method of hedging risk was not patentable; the question was how broadly the Supreme Court's opinion would reach. In the end, the Court rejected a categorical exclusion of all business method patents (although Tom Goldstein has argued that Stevens's concurrence, which would have rejected business methods completely, reads like a lost majority). The Court also rejected the Federal Circuit's holding that the machine-or-transformation test is the only test for determining whether a process is patentable subject matter under § 101. But the Court noted that "the machine-or-transformation test is a useful and important clue," and lower courts have continued to rely upon it. So where does this leave us?

The Stanford Law Review recently selected six Bilski-related papers for its 2011 Symposium on The Future of Patents: Bilski and Beyond. Michael Risch has already shared brief descriptions of the papers over at Madisonian; I will examine each of the papers in a series of posts over the coming week.

First up is Life After Bilski by Mark Lemley (Stanford), Michael Risch (Villanova), Ted Sichelman (San Diego), and Polk Wagner (U Penn). These four were attorneys for, and signatories to, an amici brief in Bilski that argued for a technology-neutral applied/abstract approach: "Where an idea is claimed as applied, it is eligible for patentability, but if it is claimed merely in the abstract it is not." Their symposium piece moves away from this focus on abstraction: "We don't exclude inventions from patentability because the invention is too abstract. We refuse to patent certain claims when those claims reach too broadly and thereby threaten downstream innovation."

The authors argue that the machine-or-transformation test "threatens to effectively become mandatory," and this is bad because "[g]atekeeping approaches have proven unsatisfactory." (This builds on Risch's Everything is Patentable.) The thesis is that the abstract ideas doctrine is not meant to exclude certain subject matters; rather, it "prevent[s] inventors from claiming their ideas too broadly." (§ 112 also limits claim scope, but while § 112 is only concerned with disclosure, § 101 "is primarily concerned with removing obstructions to follow-on innovation.") The authors describe three historical cases that "illustrate how excluding abstract ideas limits claim scope"—O'Reilly v. Morse, The Telephone Cases, and Mackay Radio—and describe how their theory "is also largely consistent with the Court's more recent 'abstract idea' decisions," such as Gottschalk v. Benson, Diamond v. Diehr, Parker v. Flook, and Bilski. They also claim that their theory "provides the only reasonable explanation" of In re Abele.

Section III.B describes the new proposed § 101 test, a "contextual, common-law approach" that considers at least five factors:

1. "Is the claimed invention potentially generative of many kinds of new inventions?"
2. "Does the industry rely heavily on cumulative invention?"
3. "Is the technological field fast-moving?"
4. "Has the patentee disclosed a small number of embodiments but claimed a broad inventive principle?"
5. "Has the patentee made an important contribution relative to the prior art?"

Section III.C argues that this test should not be the first step in evaluating patentability; the authors "agree with a number of commentators that the right time to apply section 101 is as a backstop after all other validity doctrines have been examined." Finally, Section III.D shows how this test might be applied. The authors argue that patent claims at issue in LabCorp v. Metabolite (method for diagnosing a vitamin deficiency) and Prometheus v. Mayo (method for adjusting medication dose) satisfy their test because the technologies were not generative and the patents were unlikely to hinder future innovators. For software patents that could impede follow-on innovators, as in Ex parte Heuer, the question is whether the patent "disclosed sufficient embodiments—particularly with respect to the existing prior art—to justify the relatively broad language" of the claim. The authors also argue that even if patents on "exercising" or "every method of controlling ball motion using spin" are too broad, specific patent claims for "a novel method of exercising that helps relieve knee pain" or "a novel method of throwing a particular kind of curveball" should be patentable subject matter because they would not foreclose follow-on innovation (though they may fail as obvious, etc.).

So does this proposed test help move the ball forward? There was no shortage of ideas on how the Supreme Court should define patentable processes in Bilski (see PatentlyO's summaries of briefs here and here), and articles of this form ("the current test is unworkable, and here's my better test") often fail, but this piece seems to satisfy my basic patent criteria for a valuable article. I haven't seen this idea that § 101 should be about claim scope, and I thought the examples of how this test would work in practice were very useful. Is their test really workable? Part of the appeal of the formalistic machine-or-transformation test is its relative ease of application, but I think this claim-scope test does a decent job of following Peter Lee's suggestion of making patent standards "enabled," such as by providing examples. And I think this piece could do more to emphasize the potential efficiency gains of the claim-scope approach, as it would help reduce the dynamic inefficiencies caused by patents.