So Exactly What Can Be Patented?

October 12, 2009 Published Work
Connecticut Law Tribune, Vol. 35, No. 41

Another milestone was reached Sept. 25, in Bilski v. Kappos, likely the most important patent law case in more than 20 years. On that day, the U.S. Solicitor General, representing the Patent and Trademark Office (PTO), filed a brief in the U.S. Supreme Court stating that the Court of Appeals for the Federal Circuit should be affirmed in refusing to grant Bernard L. Bilski a patent.

At the heart of the issue looms the larger question as to what kinds of processes should be "patentable subject matter" for an applicant and examined for patentability by the PTO. A particularly thorny question is whether software and non-automated business methods should enjoy the same protection as more traditional, "industrial" processes, such as processes for making soap or vulcanized rubber.

The statute relating to patentable subject matter, 35 USC § 101, broadly states that "whoever invents or discovers any new and useful process" is eligible to have that process examined by the PTO to see if it is patentable. For the last decade, software and business methods have been considered patentable if they provided a "useful, concrete and tangible result." Last year, in an en banc decision in Bilski, the Federal Circuit discarded that test in favor of a "machine-or-transformation" test.

The Federal Circuit unearthed the machine-or-transformation test from dicta in Supreme Court decisions dating back three decades, mandating that in order to be patentable subject matter, a process must be "tied to a particular machine or apparatus" or transform "a particular article into a different state or thing." The Federal Circuit majority's opinion in the en banc decision observed that this test is "the only applicable test" for whether a process is patentable.

Bilski's claims related to a method for hedging risk in commodities transactions, including steps of "identifying market participants" and "initiating transactions." However, the claims did not recite the use of any machine, such as a computer, to affect the claimed method. The Federal Circuit found Bilski's claims unpatentable for failure to meet the machine-or-transformation test.

The Supreme Court granted certiorari in the Bilski case on June 1, 2009. Subsequently, a total of almost 60 briefs have been filed, including those from the parties, from amici for each side, and from amici for neither side.
Bilski's brief, and amicus briefs supporting his position, argued that the machine-or-transformation test is unduly narrow and rigid; does not reflect the realities of information-age technology; and "disrupts settled expectations" of those who relied on the test for patentable subject matter that the Federal Circuit has applied to all patent applications during the last decade. Several briefs stated that by pursuing a bright-line test involving a machine, the Federal Circuit had not followed Supreme Court precedent, and was imposing an outmoded "physicality" requirement. Numerous briefs emphasized the breadth of patentable subject matter, quoting the last Supreme Court case to analyze the statute: "Anything under the sun that is made by man" should be eligible for patenting. Amici representing companies involved in software, communications, medical devices and medical diagnostics pointed out the difficulties of tying process innovations in their industries to machines.

‘Enemy Of Innovation'

Bilski's amici concluded that the machine-or-transformation test is inadequate, and that a clear, predictable and more flexible test is needed. One brief prominently quoted a dissenting opinion in the Federal Circuit case: "Uncertainty is the enemy of innovation." There was no consensus or consistent conclusion, however, regarding what test for patentable subject matter should be used instead.

The various briefs arguing that Bilski's claims were patent-eligible did not, in general, conclude that Bilski should actually receive a patent. Other patent statutes (35 USC §§ 102, 103 and 112) require that even if a claim to an invention recites patentable subject matter, it must still be new and not obvious, claim the invention in clear and definite language, and be clearly supported by a written description of the invention.

The Solicitor General's brief, urging the Supreme Court to affirm the Federal Circuit, discusses at length the meaning of the terms "process" and "art" (as processes were called in the early days of development of patent law). The brief quotes dictionaries from the 18th and early 19th centuries to support a position that processes must refer to technology or industry.

The brief also quotes 19th- and early 20th-century patent cases holding that a patentable process must involve technology, thus requiring a machine or transformation of matter. The Solicitor General further opines that Bilski's claims should not be patentable, since "business-related methods untethered to technology" are not "processes" under the statute (or "useful arts" in the language of the Constitution). However, the brief stops short of a full endorsement of the machine-or-transformation test as the sole test for patentability.

The Solicitor General's brief draws a fine distinction between claims to software and claims to "modes of organizing purely human activity," arguing that Bilski's business method falls within the latter category. Similarly, the brief distinguishes medical diagnostic techniques from business methods, stating that the former involve some transformation of matter and suggesting that the latter do not. The brief concludes that the Supreme Court could decide that business methods (such as risk management techniques and legal methods) are unpatentable, without affecting the patentability of software or medical diagnostic techniques.

As an aside, the Solicitor General's brief identifies dicta in a prior Federal Circuit decision to the effect that a computer, when programmed to perform a process, may thus become "sufficiently particular" so that the process (expressed in the executing program code) could be patentable. However, the brief goes on to argue that, even if it were tied to a computer, Bilski's method would still be unpatentable, since the use of a computer is not "central to the purpose of the method."

The Solicitor General also argues that Bilski's method should be unpatentable because Bilski's claims seek to monopolize the abstract idea of hedging risk. The Supreme Court has consistently held that claims to an abstract idea are not patentable.

Drawing Distinctions

Amicus briefs supporting the government's position were filed the week ending Oct. 2. Most of those briefs agree with Bilski in rejecting the Federal Circuit's stance that the machine-or-transformation test should be the only test for patentability of a process. Not surprisingly, advocates of free software state that all software patents should be eliminated, arguing that such patents have had "perverse economic effects." Other briefs draw a distinction between software and business methods such as Bilski's, arguing that a process requiring the use of a machine, as software requires a computer to be useful, should be patentable. Several briefs argue that business methods should not be patentable since they are economically harmful.

There seems to be a consensus, on both sides, that the machine-or-transformation test set forth by the Federal Circuit is "overly difficult to implement in practice" and should not be applied rigidly. Most amici, on both sides, are in favor of retaining patentability of software-based inventions, at least in some form; perhaps, for example, by requiring the software process to be executing on a machine, as opposed to simply reciting the coded steps. Although there is general agreement that a clear, predictable and flexible test is needed, a return to the former "useful-concrete-tangible result" test is considered unlikely.

Oral arguments are scheduled for Nov. 9. The entire patent community, nay the entire legal community, will be watching and listening.