Diamond v. Diehr: A Computer-Run Industrial Process Is Patent-Eligible

Diamond v. Diehr, 450 U.S. 175 (1981), is the Supreme Court’s clearest early statement that a process does not lose patent eligibility simply because it incorporates a mathematical formula and a programmed computer. After Gottschalk v. Benson and Parker v. Flook had invalidated claims the Court viewed as attempts to patent algorithms, many assumed that any computer-implemented invention was suspect. In a closely divided 5–4 decision written by Justice Rehnquist, the Court held otherwise: a method for curing raw synthetic rubber that used the Arrhenius equation to calculate cure time, and a computer to signal when to open the press, was patent-eligible subject matter under 35 U.S.C. § 101 because the claim was directed to an industrial process transforming an article, not to the equation in the abstract.

At a glance

• Case: Diamond v. Diehr, 450 U.S. 175 (1981), Docket No. 79-1112
• Court: Supreme Court of the United States, on certiorari to the Court of Customs and Patent Appeals
• Decided: March 3, 1981; 5–4
• Opinion: Justice Rehnquist, for the Court (joined by Burger, C.J., and Stewart, White, and Powell, JJ.); Justice Stevens dissenting (joined by Brennan, Marshall, and Blackmun, JJ.)
• Subject matter: Patent eligibility under § 101 of a rubber-curing process implemented with a mathematical formula and a digital computer
• Holding: A claimed process is patent-eligible even though it employs a mathematical formula and a programmed computer, so long as the process as a whole performs a function the patent laws were designed to protect

The curing process and the problem it solved

The respondents, James Diehr and Theodore Lutton, sought a patent on a process for molding raw, uncured synthetic rubber into cured precision products such as tire treads and seals. It had long been known that the time required to cure rubber depends on the temperature inside the mold, and that this relationship can be expressed by a well-known equation called the Arrhenius equation. The practical difficulty was that the temperature inside a closed press could not be measured accurately during the cure, so molders had to rely on estimates and average cure times — resulting in products that were often overcured or undercured.

The claimed invention addressed that difficulty by continuously measuring the actual temperature inside the mold with a thermocouple, feeding those measurements into a computer, and having the computer repeatedly recalculate the cure time using the Arrhenius equation and the real-time temperature data. When the calculation indicated the cure was complete, the computer signaled a device to open the press automatically. The patent examiner rejected the claims as directed to nonstatutory subject matter under § 101, reasoning that the steps carried out by the computer were unpatentable and that the remaining steps were conventional. The Court of Customs and Patent Appeals reversed, and the Commissioner of Patents sought review.

The claim as a whole, not its parts

The Court began from the broad language of § 101, which extends patent eligibility to “any new and useful process, machine, manufacture, or composition of matter.” It reaffirmed the settled exceptions — “laws of nature, natural phenomena, and abstract ideas” are not patentable — and acknowledged that a mathematical formula, standing alone, falls within the excluded category as an abstract idea or “basic tool of scientific and technological work.” But the Court drew a sharp line between claiming a formula and claiming a process that uses one. Diehr and Lutton, it stressed, did “not seek to patent a mathematical formula.” Instead they sought “patent protection for a process of curing synthetic rubber,” an activity that “has been the subject of patent protection for years.” The presence of the Arrhenius equation as one step did not doom the claim.

Central to the analysis was the instruction to consider the claim “as a whole.” The Court rejected the government’s approach of dissecting the claim into old and new elements, ignoring the steps that recited the formula, and then asking whether what remained was novel. “In determining the eligibility of respondents’ claimed process for patent protection under § 101,” the Court wrote, “their claims must be considered as a whole. It is inappropriate to dissect the claims into old and new elements and then to ignore the presence of the old elements in the analysis.” The Court also insisted that § 101 eligibility is a distinct question from the novelty inquiry of § 102 and the nonobviousness inquiry of § 103: whether an invention is new or obvious “is of no relevance in determining whether the subject matter of a claim falls within the § 101 categories.”

Transformation and the limits of the holding

What made the claim eligible was that it described a process that “transform[s] or reduce[s] an article to a different state or thing” — raw, uncured rubber into a cured, molded product. The computer and the formula were tools deployed within a physical, industrial process, not the object of the patent. The Court took care to distinguish its recent decisions. In Gottschalk v. Benson and Parker v. Flook, the claims had, in the Court’s reading, sought to preempt the use of a mathematical algorithm itself, with the surrounding steps amounting to insignificant activity. Here, by contrast, the applicants sought “only to foreclose from others the use of that equation in conjunction with all of the other steps in their claimed process.”

The Court also cautioned against easy evasions running in the opposite direction. It warned that an applicant cannot obtain a patent on a formula or principle by adding “insignificant postsolution activity,” reaffirming the concern voiced in Flook. The distinction the Court drew — between a claim that recites a formula as part of a genuine, transformative process and a claim that dresses up an algorithm with token steps — became the governing framework for computer-related eligibility until the Court revisited the field decades later in Bilski, Mayo, and Alice.

Open questions

The 5–4 division reflected genuine disagreement about how to police the line the majority drew. Justice Stevens, who had written Flook, dissented and argued that the majority’s approach invited applicants to obtain patents on computer programs merely by embedding them in otherwise conventional processes, undermining the exclusion of algorithms. The tension between Diehr’s “claim as a whole” instruction and Flook’s directive to treat the algorithm as prior art was never fully reconciled, and it resurfaced directly in the modern § 101 cases. The two-step framework of Mayo and Alice — which asks whether a claim is directed to an ineligible concept and, if so, whether the remaining elements supply an inventive concept — can be read as an attempt to harmonize the competing impulses of Diehr and Flook, and courts continue to cite Diehr for the proposition that improvements to a technological process remain eligible.

Implications

• A formula does not defeat eligibility. Reciting a mathematical equation or a computer program as part of a claim does not, by itself, render the claim patent-ineligible under § 101.
• Assess the claim as a whole. Eligibility is judged by looking at the entire claimed process, not by stripping out the steps that recite a formula and examining only what is left.
• Transformation matters. A process that transforms an article into a different state or thing — like curing raw rubber into a finished product — sits comfortably within § 101.
• Eligibility is separate from novelty. Whether a claim is new or nonobvious is governed by §§ 102 and 103 and is irrelevant to the threshold § 101 question of patent-eligible subject matter.

Frequently asked questions

What did Diamond v. Diehr hold? The Court held that a process for curing synthetic rubber that uses the Arrhenius equation and a programmed computer to control the cure is patent-eligible subject matter under 35 U.S.C. § 101. A claim is not rendered ineligible simply because it includes a mathematical formula or computer program, so long as the process as a whole performs a function the patent laws were designed to protect.

How does Diehr differ from Gottschalk v. Benson and Parker v. Flook? In Benson and Flook the claims were, in substance, to a mathematical algorithm itself, with no meaningful application. In Diehr the formula was merely one step in an otherwise conventional industrial process that transformed raw rubber into a cured product, so the claim was directed to an eligible process rather than to the equation.

What is the “claim as a whole” principle? Diehr instructs that patent eligibility must be assessed by looking at the claimed invention as a whole, not by dissecting it into old and new elements and disregarding the parts that recite a mathematical formula. A process does not become unpatentable merely because one of its steps uses a formula or computer.

Authorities and sources

• Diamond v. Diehr, 450 U.S. 175 (1981), Docket No. 79-1112 (decided March 3, 1981). Justia; Cornell Legal Information Institute.
• Official U.S. Reports slip opinion via the Library of Congress.
• Oral argument and case summary via Oyez.
• The 5–4 vote, Rehnquist authorship, Stevens dissent, and rubber-curing facts corroborated by Wikipedia: Diamond v. Diehr.