Maday Patent Law PLLC



[Posted on December 30, 2009]

In re Bilski: Altering the Landscape of Subject Matter Eligibility for Process Patents


In re Bilski1 is a recent case decided by the United States Court of Appeals for the Federal Circuit (CAFC)2, concerning the patentability of process claims, particularly business method claims. The CAFC affirmed the rejection of a claim to a method of hedging risks in commodities trading, ruling that the claim failed to disclose patent-eligible subject matter. In so doing, the court identified a single test to be used for the analysis as opposed to a myriad of tests that had been considered in previous cases, and arguably raised the bar of patent eligibility for methods or processes.

To be sure, the wisdom of granting patent protection for processes (particularly for business methods and certain software processes), and to what extent that protection should apply, has been perhaps the most highly contested area of contemporary patent law. After all, is a new method of conducting a business really an “invention”? Certainly not in the conventional sense, but the increasing importance of these method patents for certain business enterprises has been undeniable. Software patents largely driven by the technological emergence of the information age are also achieving a higher degree of notoriety. Bilski, therefore, has drawn a lot of attention in patent law circles.

The debate over defining the proper boundaries for patent-eligible subject matter has been fueled at least partially by Amazon’s now-infamous “single-click” patent, in which customers can purchase an item on-line by clicking only one button instead of electronically mimicking a “conventional” shopping experience by first adding the item to the customer’s shopping cart, and then proceeding to the checkout via a second button. Whether or not this is a “technological innovation” worthy of patent protection has been called into question. Additional concern has been expressed about the potential for patents such as this to stifle future innovation by precluding from others the use of basic building blocks adaptable to a multitude of possible circumstances.

In Bilski, the CAFC reviewed and discarded several tests that have previously been used to determine whether the subject matter of process claims is eligible for patent protection. Instead, the court adopted the so-called “machine or transformation” test as the single standard for determining the patent-eligibility of processes. This article will identify the present boundaries of patent eligibility, in part by reviewing the now-outdated tests that were identified and subsequently discarded by the court. The ultimate breadth of the Bilski decision is yet to be seen, and will surely develop in the form of future case law applying the present standard. Most importantly, the decision gives rise to critical considerations that must be taken into account when drafting process claims in patent applications. The likelihood of claim allowance may be enhanced by understanding the court’s reasoning for its decision.

Background & Holding

The subject matter requirements for patentability are set forth in 35 U.S.C. 101:

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...

Processes clearly fall within the realm of patent-eligible subject matter, along with three other categories of patentable subject matter (machines, manufactures, and compositions of matter). The Bilski court outlined how to determine whether a particular claim is a “new and useful process” under 101 or an unpatentable “fundamental principle” (meaning “laws of nature, natural phenomena, [] abstract ideas,” and “mental processes”) free for all to use and for which no party may acquire exclusivity. Ultimately, Bilski adopted the machine-or-transformation test:

A claimed process is patent-eligible under 101 if:

(1) it is tied to a particular machine or apparatus; or,
(2) it transforms a particular article into a different state or thing.

The meaning and effect of Bilski can be better understood through review of several tests that were rejected by the court, as listed below. The greatest impact will likely be realized by the court’s move away from the State Street test.

Outdated or Rejected Tests

  • Diehr/Benson pre-emption test3
  • Freeman-Walter-Abele (FWA) “physical elements” test4
  • State Street “useful, concrete, and tangible results” test5
  • Technological Arts test
  • Comiskey “physical steps” test6
  • AT&T “physical limitations” test7

Diehr and Benson both focused on whether a claim had merely made use of a fundamental principle to achieve a particular purpose, or whether it had, in effect, sought patent protection of that fundamental principle by pre-empting all practicable uses of it. In the earlier of the two cases, Benson attempted to patent a process for converting data in a binary-coded decimal format to a pure binary format in a computer. The court held that the claim pre-empted all use of the algorithm (a recognized fundamental principle) because the algorithm would almost always be employed on a computerized platform; therefore, allowing the claim would essentially be allowing a patent on the algorithm itself.

The claim at issue in Diehr made use of a mathematical equation (the Arrhenius equation) to calculate the curing time of rubber. Diehr held that this use of a mathematical formula in a process claim was patent-eligible subject matter even though a mathematical formula itself may not be patented. The Diehr claim made use of the equation for only a specific application without monopolizing its use; therefore, allowing the claim would not equate to allowing a patent on the mathematical equation, which others were free to use for alternative purposes.

FWA was a 2-part test that involved: (1) determining whether a claim recites an “algorithm” within the meaning of Benson, then (2) determining whether the algorithm is applied in any manner to physical elements or process steps. The court criticized the FWA test for dissecting claims into individual elements rather than analyzing the claim as a whole (a requirement that is well-understood for evaluating the patentability of claims). The court concluded that the FWA test was inadequate, indicating that it is possible for a claim failing this test to be eligible for patent protection.

State Street had been the prominent standard prior to Bilski, although technically the claim at issue was drawn to a machine rather than a process. State Street held that “the transformation of data, representing discrete dollar amounts, by a machine through a series of mathematical calculations into a final share price, constitutes a [patent-eligible invention] because it produces ‘a useful, concrete and tangible result.’” State Street used a machine to perform a transformation, but it was the “useful, concrete, and tangible result” language that achieved most of the attention from the holding. Under this reasoning, the scope of patent-eligible subject matter was wider than the Bilski holding, because money is concrete and therefore manipulation of financial commodities (representing money) could equate to a “useful, concrete, and tangible result.”

The “technological arts” test was rejected on the grounds that the terms “technological arts” and “technology” are ambiguous and ever-changing. Additionally, the court had rejected a categorical exclusion of business method claims, thereby purporting to place business methods on equal footing with all method claims.

The Comiskey “physical steps” test was - as indicated by the Bilski court - not really a test at all but might be perceived as such and thus needed to be dismissed for clarification. In rejecting the claim at issue (purportedly by applying the machine-or-transformation test), the Comiskey court noted that the claim recited a “mental process” lacking any significant “physical steps.”

By rejecting the AT&T “physical limitations” test, the court affirmed that patentability is not barred merely due to the inclusion in a claim of fundamental principles such as mathematical equations or algorithms, so long as the claim meets the requirements of the machine-or-transformation test.

Synthesis & Analysis

Bilski identified a couple of considerations in addition to defining the general test to be applied.

Pertaining to the second prong of the test, the transformation must be something more than “extra-solution activity” (sometimes also referred to as “post-solution activity,” although presumably the transformation could be made to occur at any stage of the process). The court recognized the ability of skilled claim draftspersons to insert perfunctory language into almost any claim for the purpose of including a trivial transformation without sufficiently narrowing the scope of the claim. The transformation, rather, “must be central to the purpose of the claimed process.”

Even less clear is the first prong of the test - that is - when a claim is sufficiently “tied to a machine” to be eligible for patent protection. Because the Bilski claim did not limit any process step to any specific machine or apparatus, the court did not evaluate this prong in detail, and rather indicated that the contours would have to be determined through future case law. Importantly, the court declined to comment on the circumstances under which recitation of a computer is adequate for tying a process claim into a “machine.” In today’s information age (and particularly with business method and computer software claims), the use of computers in claimed processes is sure to take on a important role in future development and associated claim drafting.

In its reasoning, the court routinely reverted to a Diehr/Benson-type pre-emption analysis to support its promulgation of the machine-or-transformation test. If this is the central consideration of patentability (as it appears to be), perhaps the court should have adopted Deihr/Benson as the proper test to apply. Such an approach would have been a “purer” replication of Constitutional provisions and Congressional intent, but in the eyes of the court, may have fallen short for practical considerations for failing to clearly delineate the legal boundaries of patentability. Although it may be argued that Bilski created more confusion than clarification.

Judge Rader dissented from the majority opinion in Bilski, and criticized the majority for not “sa[ying] in a single sentence: ‘Because Bilski claims merely an abstract idea, this court affirms the Board's rejection.’” He further criticized the majority for instead introducing a myriad of “unanswerable questions.”8

While the court made it clear that the sole test that may be relied upon to determine the patentability of process claims is the machine-or-transformation test, the actual boundaries of the test are presently open to interpretation. As often occurs in the legal field, particularly when new tests are promulgated or singled out, the precise application of the court’s language to particular sets of facts are necessarily indeterminate until ruled upon by the judiciary. Presently, claims must be drafted with the machine-or-transformation test in mind with the understanding that the validity of some claims by may influenced by future court precedent.

The Bottom Line

Let’s review the bottom line of Bilski. A process claim may receive patent protection if one of two conditions apply.

  1. The claim is tied to a particular machine or apparatus.
    • The conditions under which a computer will suffice as a “machine” are presently unclear.
  2. The claim transforms a particular article into a different state or thing.
    • The transformation must be more than mere “extra-solution activity,” that is, it must be “central to the purpose of the claimed process."

Finally, the court’s reasoning behind the machine-or-transformation test may help shed light on the ambiguous boundaries of its applicability. As Judge Rader recognized, the court arguably relied on a Diehr/Benson-type analysis to determine whether the claim pre-empted a fundamental principle. At the end of the day, this consideration may be employed to fine-tune the applicability of the machine-or-transformation test. Even if not explicitly recognized by the Bilski majority, the concern over whether a claim effectively pre-empts a fundamental principle or merely makes use of it to perform a specific function is likely to remain a central aspect of subject matter patent-eligibility.


1 545 F.3d 943 (Fed. Cir. 2008).

2 The CAFC is the court charged with handling all patent-related appeals from the United States Patent and Trademark Office.

3 Diamond v. Diehr, 450 U.S. 175 (1981); Gottschalk v. Benson, 409 U.S. 63 (1972).

4 In re Freeman, 573 F.2d 1237 (CCPA 1978); In re Walter, 618 F.2d 758 (CCPA 1980); In re Abele, 684 F.2d 902 (CCPA 1982).

5 State St. Bank & Trust Co. v. Signature Fin. Group, 149 F.3d 1368 (1998).

6 In re Comiskey, 499 F.3d 1365 (Fed. Cir. 2007).

7 AT&T Corp. v. Excel Commc’ns, Inc., 172 F.3d 1352 (Fed. Cir. 1998).

8 What form or amount of “transformation” suffices? When is a “representative” of a physical object sufficiently linked to that object to satisfy the transformation test? (E.g., Does only vital sign data taken directly from a patient qualify, or can population data derived in part from statistics and extrapolation be used?) What link to a machine is sufficient to invoke the “or machine” prong? Are the “specific” machines of Benson required, or can a general purpose computer qualify? What constitutes “extra-solution activity?” If a process may meet eligibility muster as a “machine,” why does the Act “require” a machine link for a “process” to show eligibility?

Copyright 2009-2011 Maday Patent Law, PLLC