By: Robert R. Sachs
In my posting on the differences between abstract ideas and abstractions, I stated that “The problem is not abstraction: the problem is inclination of the academy and the judiciary to throw up their hands in defeat when dealing with these issues. See Mark A. Lemley et al., Life After Bilski, 63 STAN. L. REV. 1315, 1316 (2011) (“Put simply, the problem is that no one understands what makes an idea ‘abstract’ . . . .”).
Michael Risch, one of the co-authors of Life After Bilski (LAB), correctly pointed out that I read this statement too broadly, that they were referring to “observers,” rather than themselves, and that they did propose various tests for abstract ideas. Michael is, of course, correct. I should have been more careful in my phrasing.
However, having now re-read LAB, I remain of the view that the approach therein does not tackle the question of What is an abstract idea? itself, independent of any reliance on patent law requirements and policy considerations. Instead, LAB proposes a functional approach to the question of patent eligibility by employing various legal, policy and value considerations of claim scope, preemption, contribution to innovation and so forth, as factors. In short, I would argue that LAB elides the question entirely and instead attempts to answer the ultimate question: What claims are patent ineligible?
First, LAB recasts the issue of abstract ideas as a placeholder for policy considerations:
That is, the abstract ideas exception operates where a patent claim is “too broad” in the sense that it encroaches upon society’s right to unfettered access to scientific truths, fundamental principles, and the like; these properly belong in the commons upon which future innovations can be built, “free to all men and reserved exclusively to none.” LAB at 1329.
Understood in this way, the abstract ideas doctrine is not about finding a conceptual category of inventions that is entitled to no protection at all, nor about determining the quality of the disclosure. Instead, it is about encouraging cumulative innovation and furthering societal norms regarding access to knowledge by preventing patentees from claiming broad ownership over fields of exploration rather than specific applications of those fields. LAB at 1329 (emphasis added).
To say that the abstract idea doctrine is not about finding a category that is excluded from patent eligibility is to say that we need not answer the question of What is an abstract idea?, but instead we should answer a different question: What approaches encourage innovation and further societal norms? To answer this question, LAB uses the mechanism of claim scope:
Focusing on claim scope provides a principled way to determine whether a particular claim, rather than a particular subject matter, is unpatentable. Specifically, our approach focuses on the application of the idea to determine the proper level of abstraction that merits protection. As claims become broader—and necessarily more general and abstract—they become more indefinite and difficult to understand, and more likely to ensnare future inventions embodying the inventive principle. At their limit, they claim everything and contribute nothing. The practical effect of allowing such breadth is to make later improvements more costly or even impossible. To be sure, identifying overclaiming is not straightforward. LAB at 1338.
In my view, this unties the horse from the cart: to decide claim scope without directly addressing whether a patent claims an abstract idea is to decide patent eligibility without addressing the meaning of the claim language. Further, the statement that broader claims are necessarily more abstract conflates the use of abstractions in claim drafting with abstract intellectual ideas, the very distinction that I draw in my article. In any event, LAB then sets forth a multi-factor test for over-claiming:
In sum, we believe at least five factors are critical to a proper scope-based determination for patentable subject matter eligibility under § 101:
- Is the claimed invention potentially generative of many kinds of new inventions?
- Does the industry rely heavily on cumulative invention?
- Is the technological field fast-moving?
- Has the patentee disclosed a small number of embodiments but claimed a broad inventive principle?
- Has the patentee made an important contribution relative to the prior art?
No one factor should dominate; we advocate a contextual, common-law approach. Courts and scholars are likely to develop other factors as our approach is applied over time. LAB at 1341.
These are all interesting questions in the, forgive me, abstract. But in my view they have little, if anything, to do with patent eligibility. My primary difficulty with most of these factors is that they are questions that simply cannot be answered in typical patent litigation. Whether an invention is “generative” or an “important contribution” cannot be determined without the long lens of history—for some inventions, it takes many, many years before their value is appreciated or their impacts on society felt. If a patentee sues immediately upon issuance of its patent, does the defendant get to stay litigation for 20 or 30 years to see how generative the invention is, and if too generative (whatever that means), have the claims declared patent ineligible? Obviously not. Thus, while these questions are interesting from an “eye of God” perspective of patents, they are not susceptible to practical application in the courtroom.
More generally, I take issue with LAB’s approach. LAB takes the judicial exceptions to § 101—laws of nature, natural phenomena and abstract ideas—as markers or signs for representing a vast complex of considerations. In this regard, LAB’s methodology is no different in kind from those of the courts and most other scholars: they differ primarily as to which considerations matter and how they are applied in a given case. This approach is textually and contextually at odds with the statute: Congress could have, but (wisely) did not include any such considerations in § 101, or even suggest them in the legislative history. The Supreme Court has told us “more than once” that “courts should not read into the patent laws limitations and conditions which the legislature has not expressed.” Diamond v. Chakrabarty, 447 U. S. 303, 308 (1980). We should likewise not read into the patent laws policy considerations that have similarly not been expressed by Congress.
In contrast, my approach is to start with the language of the exceptions themselves and attempt to define what these words mean: What is a law of nature? What is a natural phenomenon? What is an abstract idea? To the extent that exclusions from patentability are to be “applied narrowly” (Classen Immunotherapies, Inc. v. Biogen IDEC), then starting with definitions of these terms would seem a better approach than importing essentially open-ended policy considerations that can almost always be argued either way, and which, in any event, are generally beyond the fact-finding capabilities of a district court.
Most fundamentally, I presume that when the Supreme Court enunciated these exclusions to § 101, they—like Horton the Elephant—meant what they said and said what they meant. Now, you could say that’s an unreasonable thing to believe, but I would hope not. After all, the business of the Court is transacted in the currency of language, for they have no other means of exchange. If we cannot take the Court’s statements for what they say, if we must read between the lines, then we do not have a system of jurisprudence, we have nothing more than legal tasseography.