US business method patents – the Bilski case

12 April 2010

Hilary Pearson

State Street

Since the 1998 decision of the US Court of Appeals for the Federal Circuit (CAFC) in State Street Bank & Trust Co v Signature Financial Group Inc [1998] 47 USPQ 2d 1956 (Fed Cir) (State Street), which held that there was no statutory basis for the previous case law exclusion of business methods from patentability, the US has been regarded as having a very generous approach to the patentability of business methods.  Although State Street required use of a computer, methods not tied to any kind of "technological" means of implementation have been held to be patentable provided they meet the standard tests of novelty and non-obviousness.  The US statutory basis for patentability is section 101 of the patent statute (35 USC), which provides:

"Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title."

It should be noted that there are no express statutory exclusions from patentability in the US such as are contained in Article 52 of the European Patent Convention.  There are exclusions created by case law, such as abstract ideas, laws of nature and natural phenomena, but the scope of section 101 is always open to Supreme Court interpretation so exclusions can change.  The general principle to be applied to the determination of the scope of patentability was said by the Supreme Court in Diamond v Chackrabarty 447 US 303 (1980) to be that Congress had intended patentable subject matter to "include anything under the sun that is made by man."   One example is computer programs: in early cases these were treated as unpatentable algorithms but this approach changed with the Supreme Court decision in Diamond v Diehr 450 US 175 (1981) (Diehr).  In that case, which involved use of a computer controlled process for the curing of rubber using a reiterative algorithm, the Supreme Court drew a distinction between those claims that sought to pre-empt the use of a fundamental principle (such as a mathematical algorithm), and claims that sought only to foreclose others from using a particular application of that fundamental principle.

Bilski – CAFC

The wide scope for patenting business methods that prevailed after State Street was severely narrowed in 2008 by the en banc decision of the CAFC in In re Bilski 545 F. 3d 943 (Fed. Cir. 2008) (Bilski).  The patent application, which related to a method of hedging commodities transactions which did not include any use of a computer or other machine, was refused by the US Patent and Trademark Office on the basis that no physical transformation was involved.  The applicant appealed to the CAFC which, after hearing oral argument with a normal three judge panel, announced that there would be an en banc review of the standards for determining whether a claimed process is patent-eligible subject matter.  By a majority of eight to one it rejected the appeal.

Defining a patentable process

This invention could only come into the category of new and useful processes under section 101.  However, the meaning of the term "process" in the statute had to be interpreted in accordance with Supreme Court law, which had held it was narrower than the ordinary dictionary definition and excluded certain fundamental principles, namely laws of nature, natural phenomena and abstract ideas.  In order to distinguish between such unpatentable principles and patentable subject matter, the Supreme Court test in Diehr was to examine the claims to see whether they pre-empted all uses of such fundamental principles (unpatentable) or covered only specific, particular uses of such principles (potentially patentable).  While this principle was itself clear, its application to specific claims was not usually straightforward.  The Diehr court relied on the earlier Supreme Court decision in Gottschalk v Benson, 409 U.S. 63 (1972) (“Benson”), which set out a definitive test to determine whether a process claim was tailored narrowly enough to encompass only a particular application of a fundamental principle, rather than to pre-empt the principle itself. Under this test, a claimed process was patent-eligible if one of the following two conditions was met:

The process claim was tied to a particular machine or apparatus, or

The process claim transformed a particular article into a different state or thing.

Diehr clearly showed that reliance on this Supreme Court ‘machine-or-transformation’ test as the applicable test for section 101 analyses of process claims was sound.  The CAFC recognised that future scientific and technological developments might present difficult challenges to use of the 'machine-or-transformation test', so the Supreme Court might ultimately decide to alter or even set aside this test to accommodate emerging technologies. However, at present the ‘machine-or-transformation’ test was the governing test for determining patent eligibility of a process under section 101.

The claimed process did not transform any article to a different state or thing. Purported transformations or manipulations simply of public or private legal obligations or relationships, business risks, or other such abstractions, could not satisfy the ‘transformation’ part of the test because they were not, and did not represent, physical objects or substances.  As the process also failed to meet the machine-implementation part of the test it was not patentable subject matter.

Alternative tests

The majority opinion also examined the alternative tests for patentable subject matter that had been put forward by previous cases, and found them all to be inadequate to properly separate patentable from unpatentable subject matter.  These were:

1. Freeman-Walter-Abele

This test was named after three decisions on computer-related patent applications of the Court of Customs and Patent Appeals (CCPA) (the predecessor court to the CAFC), which formulated and then refined the test. In its final form, this test had two steps:

Determining whether the claim recited an "algorithm" within the meaning of Benson, then

Determining whether that algorithm was applied in any manner to physical elements or process steps.

2. State Street

This case used the language "useful, concrete, and tangible result". A process tied to a particular machine, or transforming or reducing a particular article into a different state or thing, would generally produce a "concrete" and "tangible" result as those terms were used in prior decisions. But while looking for "a useful, concrete and tangible result" might provide useful indications of whether a claim was drawn to a fundamental principle or a practical application of such a principle, that inquiry was insufficient to determine whether a claim was patentable under section 101.

3. Technological arts

The patent office had in some cases, including this one, rejected applications as not being directed to the ‘technological arts’.  This test was unclear because the meanings of the terms ‘technological arts’ and ‘technology’ were both ambiguous and ever-changing. No such test had ever been explicitly adopted by the Supreme Court, the CAFC or the CCPA.

4. Physical steps

The proper inquiry under section 101 was not whether the process claim recited sufficient "physical steps", but whether the claim met the machine-or-transformation test. A claim that recited "physical steps" but neither recited a particular machine or apparatus, nor transformed any article into a different state or thing, was not patentable subject matter. On the other hand, a claim that lacked any "physical steps" but was still tied to a machine or achieved an eligible transformation met the requirements under section 101.

Other opinions

One of the majority would have expressly overruled State Street, as he believed the exclusion for business methods was correct and its removal had caused many problems.  The minority of two both argued that applying the ‘machine-or-transformation’ test would exclude kinds of inventions that applied today's electronic technologies, as well as other processes that handled data and information in novel ways which had long been regarded as patent-eligible and essential to the development of the present knowledge economy, and stated that the test adopted by the majority could hold back these advances and could preclude patent protection for future new technologies.

Bilski – Supreme Court

In June 2009 the Supreme Court granted a petition for certiorari (leave to appeal) to Bilski.  A large number of amicus briefs on the merits were filed: 18 in support of the petitioner, including briefs on behalf of Borland, Novartis and the AIPPI, 22 in support of the respondent, including from the American Bar Association, Microsoft, Philips and Symantec, Bloomberg, RedHat, the Business Software Alliance and several industry bodies, and 25 supporting neither party, among them briefs from IBM, Yahoo and the AIPLA.

Oral argument was heard on 9 November 2009.  It is interesting that one of the justices who took the most active part in the argument was the new member of the court, Justice Sotomayor.  When in private practice she did trade mark work, and heard a lot of intellectual property cases as a judge before her appointment to the Supreme Court; none of the other justices has a similar IP background.  It is never easy to predict the final outcome of an appeal to the Supreme Court, but the justices who took part in the discussion all seemed to be very sceptical as to whether pure business methods should be patentable.  It should also be noted that recent Supreme Court patent decisions have not been in favour of expansive interpretations of the patent laws.  However, the justices also seemed to be concerned not to lay down any test for patentability which could exclude inventions of a type that there was general agreement should be patentable, such as medical devices.  The court must deliver its opinion before the end of the legal term, September 2010, but there is currently no indication as to when the judgment will be given.

Comparison with EPO

It is interesting to compare the CAFC with the questions relating to the patentability of computer programs referred to the EPO Enlarged Board by the EPO President. Question 2 relates to the inclusion of a requirement for a computer (machine), while question 3 asks whether a feature must cause a physical effect in the real world (transformation) in order to have technical character.  Of course, the CAFC rejected the ‘technological arts’ test, which has some relationship to the EPO requirement for a ‘technical effect’.