On July 31, 2020, in American Axle v. Neapco, the Federal Circuit split 6-6 on the question of whether to grant en banc review of a panel decision on patent eligibility seen by many as an unwarranted extension of the principle that abstract ideas, natural laws and phenomena are ineligible for patent protection into the realm of physical things. Previously, such rejections were largely focused largely on computer implemented inventions or comparisons of data, including comparisons used in diagnosis. In American Axle the invention was a method of manufacturing a shaft assembly and tuning it to attenuate vibrations. However, no specific way of tuning was set out in the claim or described in the specification and it was accepted that such tuning could involve an application of Hooke’s Law. Accordingly, a panel decision holding the claim to be ineligible as a simple application of a natural law to a particular technology without specifying the way in which the law was to be applied was left undisturbed.
In October 2019 a 2-1 majority of a panel of the Court of Appeals for the Federal Circuit held that a claim lacked patent eligibility as being directed to a natural law without reciting any inventive application of that natural law.
The panel decision focused on Claim 22 which was directed to a method for manufacturing a shaft assembly of a driveline system, which required providing a hollow shaft member; tuning a mass and a stiffness of at least one liner, and inserting the at least one liner into the shaft member; wherein the at least one liner is a tuned resistive absorber for attenuating shell mode vibrations and wherein the at least one liner is a tuned reactive absorber for attenuating bending mode vibrations.
The specification indicated target values for the tuning but apart from indicating a range of 5% – 30% for the ratio of the mass of the liners to the mass of the shaft member gave no guidance as to how tuning was to be carried out. The claim was therefore of a type often referred to in England as a “free beer” claim. It was agreed that Hooke’s law is a natural law that mathematically relates the mass and/or stiffness of an object to the frequency with which that object oscillates (vibrates). Here, both parties’ witnesses agreed that Hooke’s law undergirds the design of a liner so that it exhibits a desired damping frequency pursuant to the claimed invention. The patent owner argued that application of the law to the problem of damping a drive shaft assembly was too complicated to be regarded as a simple application of Hooke’s law and that it was inventive to apply the law to achieve the stated result. The majority (Dyk and Taranto JJ) disagreed; Moore J dissented and would have addressed the question of patentability as being one of lack of an enabling disclosure as to how to effect the tuning to achieve the desired result. Similar conclusions were reached on claim 1 which differed from claim 22 by simply requiring tuning at least one liner to attenuate at least two types of vibration transmitted through the shaft member without reference to mass or stiffness.
The decision caused a significant outcry from those who saw this decision as a dangerous extension of the Federal Circuit’s jurisprudence on patent eligibility into the realm of physical things.
The patent owner sought en banc rehearing. The active judges of the Federal Circuit split 6-6 on whether to grant this request with the result that it was denied, but the panel hearing the case originally issued a revised opinion in light of the issues raised in the request for rehearing and amici briefs that had been filed in connection with it.
In the amended opinion, the panel reaffirmed its view that claim 22 was a simple application of a natural law with the claim lacking any indication as to how the law was to be applied, but recognized that nothing in claim 1 invoked any natural law and remanded the case to the district court to consider whether this was simply an application of an abstract idea lacking any defined inventive step in its means of application.
Views of the Federal Circuit Judges on Whether to Grant En Banc Rehearing
Five separate opinions by different combinations of judges were issued arguing for and against an en banc rehearing. Much of their content argued as to the true meaning of the Supreme Court’s 1953 decision in O’Reilly v. Morse. Those agreeing with the panel decision saw claim 22 of American Axle’s patent as analogous to claim 8 of Morse’s patent which had been found to be invalid. That claim claimed the use of the motive power of the electric or galvanic current for marking or printing intelligible characters, signs, or letters at any distances and differed from the other claims of Morse’s patent in that it made no reference to the particular ways of doing this described in his specification. Those dissenting and supporting the request for an en banc rehearing saw the panel decision as creating a new “nothing more” test to add to the complexities of deciding on patent eligibility and noting that in O’Reilly v. Morse, the Supreme Court had not drawn the distinction between Morse’s claim 8 and his other claims that had been drawn by those concurring in denial of an en banc rehearing.
Perhaps the most useful practical comment in the discussion on en banc rehearing came from Judge Chen when he noted:
The lesson to patent drafters should now be clear: while not all functional claiming is the same, simply reciting a functional result at the point of novelty poses serious risks under section 101.
Questioning the patentability of claims that are functional at the point of novelty is not new. In 1938 in General Electric Company v. Wabash Appliance Corporation, the Supreme Court found a claim that defined a tungsten light bulb filament “made up mainly of a number of comparatively large grains of such size and contour as to prevent sagging and offsetting” to be invalid for being indefinite, the court condemning claims that use “conveniently functional language at the exact point of novelty.” The Supreme Court noted that it had previously pointed out the problem with this type of claim in Holland Furniture Co. v. Perkins Glue Co. where it was pointed out that such claims “would extend the monopoly beyond the invention.”
The American Axle case brings into focus two important issues in the whole patent eligibility debate:
1) What is the purpose of the patent eligibility test; and
2) Why does it matter if something is unpatentable whether it fails the eligibility test or the novelty test, the obviousness test or the enablement test?
The Purpose of the Patent Eligibility Test
As to the first, the extensive discussion of O’Reilly v. Morse in American Axle, may represent an attempt to get back to basics. Before a specific requirement that to be patentable what was claimed had to be non-obvious was introduced into the statute in 1952, the courts had to decide whether what was claimed was “an invention” that merited patent protection in return for the inventor having disclosed something new and useful to the public. The addition of 35 U.S.C. 103 to the statute in 1952 was seen by many as being intended to end the debate as to what constituted an “invention.” The fall-out from the Supreme Court’s decisions in Alice v. CLS Bank and Mayo v. Prometheus has shown that the issue is still with us.
Part of the problem seems to be that many attempts to address the question conflate two separate issues:
1) what types of invention should be patentable and
2) if the invention is of a type that should be patentable, how do we make sure that the claim does not reward the inventor with exclusivity extending beyond what he, she or (soon perhaps) it has invented.
The first of these is essentially a policy issue and conventional wisdom is that policy should be set by the legislature. Other jurisdictions, such as the European Patent Convention and Japan have done this by creating laundry lists of inventions that are not patentable. The result has been that to meet the needs of industry the courts or other tribunals have had to “interpret” items on the list sometimes resulting in allowing or upholding claims that seem to be prohibited by the language of the legislation (see for example the treatment of medical and diagnostic methods by the European and Japanese Patent Offices). Technology is moving too fast for a statutory laundry list approach to work well. However, a broad statement of policy, for example requiring some technical content consistent with Article 27 of the TRIPS Agreement may be useful, particularly if combined with delegated authority for the USPTO to make rules as to what falls within this definition as was done with respect to the provisions in the America Invents Act relating to Covered Business Method Review. Such rule-making would, however, need to be subject to significant public input to ensure the desirability of patent protection for various types of innovation was properly considered, otherwise we return to the big question at the start of my career, namely should the Fosbury Flop that caused a sensation in high jump at the 1968 Olympics have been patentable?
The second issue is related to, but distinct from, the first, namely if the innovation is of the type for which patent protection should be granted, what makes it an “invention.” The last time that I am aware of when an attempt was made to do this on an almost blank sheet of paper was when the initial examination guidelines were written for the European Patent Office. These gave three examples of how inventions might be made:
1) The formulation of an idea or of a problem to be solved (the solution being obvious once the problem is clearly stated).
2) The devising of a solution to a known problem.
3) The arrival at an insight into the cause of an observed phenomenon (the practical use of this phenomenon then being obvious).
The first of these was omitted when the Guidelines were substantially revised in 2012, but the other two remain at G-VII-9. The dropping of the first of these seems to have been the result of no case decided by the Boards of Appeal of the EPO ever having been found to be a convincing fit to the test (the only example ever given was the unconvincing one of T2/83 where an invention was found in solving the problem of deterioration of a chemical over time when it had not previously been known that it deteriorated).
Current U.S. practice would have a problem with the third of the EPO’s original list in that such an invention would be likely to be looked at as providing only an obvious application of a natural phenomenon or, as the dissenters in American Axle would say, nothing more than the discovery of the cause of the phenomenon. A valid question arises as to whether such insights should be entitled to patent protection, and if so how broad should such protection be. In a way, this was the question in American Axle. Had it not been known that the vibration problem was caused by failure to establish the right mass and stiffness in the liner, would the insight that the problem could be solved by “tuning” these two factors have been sufficient to justify the scope of claim 22 or would it still have been necessary to limit to a specific way of carrying out the tuning? Would it matter if the specification gave clear details as to how to carry out the tuning? These are valid questions that need consideration in deciding how to solve the present patent eligibility mess.
Why Does It Matter Which Ground of Invalidity is Used?
On the second topic noted above as to why it matters when a patent claim is invalid whether it be for lack of eligibility, indefiniteness or lack of enablement, the answer lies in U.S. civil litigation procedure. Much of the cost of litigation in the United States comes from the use of discovery procedures to uncover facts that are relevant to the case. Such discovery and associated costs can be avoided if the case can be decided by a judge on a summary judgment motion. However, this is only possible if either the issue to be decided is either a matter of law or there are no disputed factual issues. Although as noted in Berkheimer v. HP Inc. and Aatrix Software v. Green Shades Software there can be factual issues that need to be resolved before a decision can be made on questions of patent eligibility, determination of such eligibility is primarily a matter of law and can often be dealt with in a summary judgment proceeding. On the other hand, issues such as whether a claim is indefinite, whether the specification provides enough enabling disclosure or whether the claim is obvious can rarely be decided without a factual inquiry. If one can succeed in invalidating a claim on the basis that the claim lacks patent eligibility then that is typically less costly than having to prove that the claim is indefinite, lacks an enabling disclosure, or is obvious.
We still have some way to go before the patent eligibility mess is resolved, but an effort to get back to the basic questions could be a hopeful sign that the Federal Circuit will at last be willing to go where others have feared to tread.