# Gottschalk v. Benson: Why a Pure Algorithm Cannot Be Patented

> The Supreme Court held that a method for converting binary-coded decimal numerals into pure binary is an unpatentable abstract idea because a patent would preempt the formula itself.

Topic: Patents  |  Author: Lidiia Levitska  |  Source: Intellectual Property Law (outsideipcounsel.com)
Canonical: https://outsideipcounsel.com/blog/gottschalk-v-benson-algorithm-eligibility/


*Gottschalk v. Benson*, 409 U.S. 63 (1972), is the Supreme Court's first sustained encounter with the patent eligibility of computer-implemented inventions, and it established a principle that still governs: a claim to a pure mathematical algorithm is not a patentable "process" under 35 U.S.C. § 101. The applicants sought to patent a method for converting numbers from binary-coded decimal form into pure binary form — a fundamental operation in digital computing. Writing for a unanimous Court, Justice Douglas held the method unpatentable because it was, in substance, an abstract idea, and because a patent on it would "wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself."

## At a glance

- **Case:** *Gottschalk v. Benson*, 409 U.S. 63 (1972), Docket No. 71-485
- **Court:** Supreme Court of the United States, on certiorari to the Court of Customs and Patent Appeals
- **Decided:** November 20, 1972; unanimous (6–0, with Justices Stewart, Blackmun, and Powell not participating)
- **Opinion:** Justice Douglas, for the Court
- **Subject matter:** Patent eligibility under § 101 of a method for converting binary-coded decimal numerals into pure binary numerals
- **Holding:** An algorithm that has no practical application except in a general-purpose computer is an unpatentable abstract idea, because a patent would preempt the mathematical formula itself

## Numbers, computers, and the claimed method

Modern digital computers represent numbers in binary — sequences of ones and zeros. Human operators, however, often supply data in binary-coded decimal (BCD) form, in which each decimal digit is separately encoded as a group of binary digits. To perform arithmetic efficiently, a computer must convert BCD input into pure binary numerals. Gary Benson and Arthur Tabbot devised a method for making that conversion and sought a patent on it. Their claims described a series of steps — involving a "reentrant shift register" in one claim and a purely numerical procedure in another — for carrying out the conversion.

The Patent Office examiner rejected the claims, but the Court of Customs and Patent Appeals reversed and held the method patentable. The Commissioner of Patents, Robert Gottschalk, sought review in the Supreme Court. The claims were notably broad. They were not limited to any particular use, machine, or technology; one claim was not even tied to any apparatus at all. The conversion procedure could be carried out mentally or with pencil and paper, though its practical value lay in programming a general-purpose digital computer.

## An algorithm is an abstract idea

The Court situated the claims within the long-recognized exceptions to patent eligibility. "Phenomena of nature, though just discovered, mental processes, and abstract intellectual concepts are not patentable," the Court explained, "as they are the basic tools of scientific and technological work." A mathematical formula belongs to that category. The Court defined the term at the heart of the case: an "algorithm" is a "procedure for solving a given type of mathematical problem," and the claimed conversion method was exactly that — a generalized formulation of a mathematical procedure, expressed as an algorithm.

The decisive difficulty was preemption. The claims were "so abstract and sweeping as to cover both known and unknown uses of the BCD to pure binary conversion." Because the algorithm had no substantial practical application except in connection with a digital computer, a patent covering the method would in effect confer a monopoly over the formula itself, reaching every use of the conversion in any computer program. "The mathematical formula involved here has no substantial practical application except in connection with a digital computer," the Court observed, "which means that if the judgment below is affirmed, the patent would wholly pre-empt the mathematical formula and in practical effect would be a patent on the algorithm itself." Granting such a patent would remove a basic tool of technological work from the public domain, which § 101 does not allow.

## A narrow ground and a call for Congress

Although the holding was firm, the Court was careful to cabin it. It expressly disclaimed any broader ruling that computer programs are categorically unpatentable. "It is said that the decision precludes a patent for any program servicing a computer," the Court wrote. "We do not so hold." The Court likewise refused to hold "that no process patent could ever qualify if it did not meet the requirements of our prior precedents" tying processes to the transformation of physical articles, recognizing that transformation of an article to a different state or thing had been the "clue" to patentable processes but not necessarily the only test. The opinion emphasized that considerations of this magnitude — the patentability of software in a rapidly developing field — were better suited to congressional study and action than to judicial pronouncement, and it invited Congress to address the subject.

The result was a decision that closed the door on patenting an algorithm in the abstract while leaving open how much practical application would be enough to render a computer-related process eligible. That unresolved question drove the next generation of § 101 litigation, including *Parker v. Flook* and *Diamond v. Diehr*.

## Open questions

*Benson* announced a preemption principle but did not supply a workable rule for the many claims that fall between a bare algorithm and a fully implemented industrial process. How much application must an inventor add to a mathematical formula before the claim escapes the abstract-idea exclusion? The Court's suggestion that transformation of an article was merely a "clue" rather than a requirement left the boundaries of a patentable "process" uncertain. Six years later *Flook* held that adding a specific field of use and conventional post-solution steps was not enough, and *Diehr* held that embedding a formula in a genuinely transformative industrial process was. The modern *Mayo/Alice* framework — asking whether a claim adds an "inventive concept" beyond the ineligible concept — descends directly from the preemption worry first articulated in *Benson*.

## Implications

- **A bare algorithm is ineligible.** A claim to a mathematical procedure, untethered to any specific and substantial application, is an unpatentable abstract idea under § 101.
- **Preemption is the touchstone.** The key question is whether a patent would foreclose all practical uses of a formula; a claim that would monopolize the algorithm itself fails.
- **Software is not categorically barred.** *Benson* did not hold that computer programs can never be patented; it left room for claims that do more than recite the algorithm.
- **Draft toward a concrete application.** Practitioners tie mathematical methods to specific, practical implementations to avoid the preemption objection that sank the claims in *Benson*.

## Frequently asked questions

**What did Gottschalk v. Benson decide?** The Court held that an algorithm for converting binary-coded decimal numerals into pure binary numerals is not a patentable "process" under 35 U.S.C. § 101. The method was an abstract mathematical formula, and granting a patent would in practical effect preempt all uses of the formula itself, which the patent laws do not permit.

**What is the preemption concern the Court identified?** Because the claimed conversion method had no substantial practical application except in connection with a general-purpose digital computer, a patent on it would wholly cover the mathematical formula. That would grant a monopoly over an abstract idea and a basic tool of scientific work, foreclosing others from using the algorithm in any setting.

**Did Benson hold that all software is unpatentable?** No. The Court expressly declined to hold that computer programs can never be patented and disclaimed any intent to freeze process patents to old technology. It decided only that the particular claims before it, which amounted to the algorithm itself, were ineligible.

## Authorities and sources

- *Gottschalk v. Benson*, 409 U.S. 63 (1972), Docket No. 71-485 (decided November 20, 1972). [Justia](https://supreme.justia.com/cases/federal/us/409/63/); [Cornell Legal Information Institute](https://www.law.cornell.edu/supremecourt/text/409/63).
- Oral argument and case summary via [Oyez](https://www.oyez.org/cases/1972/71-485).
- The unanimous vote, Douglas authorship, non-participating Justices, and BCD-to-binary facts corroborated by [Wikipedia: Gottschalk v. Benson](https://en.wikipedia.org/wiki/Gottschalk_v._Benson).