Function of a machine doctrine

Explicit statutory authority for "processes" as patentable subject matter first appeared in the Patent Act of 1952. While processes had been accorded patent protection prior to 1952 by judicial interpretation of the "useful arts" category, this occurred only after prolonged, evolutionary development in the last half of the nineteenth century."' In the course of that evolution, the Supreme Court, on several occasions, granted patents on apparatus while denying claims to the general operation of the same apparatus. The following dicta in Corning v. Burden is frequently cited: "It is well settled that a man cannot have a patent for the function or abstract effect of a machine, but only for the machine which produces it."

Although process patents were subsequently upheld in several cases, the language of Corning was echoed in Risdon Locomotive Works v. Medart in which the Court denied coverage for "a process which involved nothing more than the operation of a piece of mechanism, or, in other words, for the function of a machine."

Soon thereafter, the functional predecessor of the C.C.P.A. declared that "a process, which amounts to no more than the mere function of a machine, is not patentable" unless "the process may be performed by hand or by another mechanism than that exhibited, although perhaps not with equal efficacy." This doctrine prevailed in the C.C.P.A. until the late 1960s.

Early writers on program related patents recognized that claims to processes might be vulnerable to the "function of a machine" objection. It was commonly felt that the disclosure of alternative hardware configurations to carry out the same algorithm could avoid the problem, although there were doubts that such alternatives could be disclosed adequately without a crushing burden of hardware detail. The possibility that the doctrine might be avoided by disclosure of alternative pencil and paper practice was also proposed, but not pressed in later discussions, since pencil and paper practice of computational steps is clearly "mental" activity and directly raises the “mental steps" doctrine.

The 1968 case of In re Tarczy Hornoch, involving claims to an analog pulse sorting device, terminated the "function of a machine" doctrine, In Tarczy Hornoch the C.C.P.A, found that the doctrine was supported neither by actual Supreme Court holdings nor by policy considerations. Foes of program patents raised the "function of a machine" objection again in Benson, asking the Supreme Court to overturn the Tarczy Hornoch holding on the ground that it flew in the face of prior Supreme Court rulings. The Court made no mention of either Tarczy Hornoch or the "function of a machine" doctrine in its opinion.