SIDNEY A. DIAMOND, COMMISSIONER OF PATENTS AND TRADEMARKS, Petitioner
v. JAMES R. DIEHR, II, AND THEODORE A. LUTTON, Respondents
1979 U.S. Briefs 1112
OCTOBER TERM, 1979
August 12, 1980
On Writ of Certiorari to the United States Court of Customs and Patent Appeals
BRIEF AMICUS CURIAE FOR CHEVRON RESEARCH COMPANY IN SUPPORT OF AFFIRMANCE
WILLIAM L. MATHIS, Attorney of Record, FREDERICK G. MICHAUD, JR., BURNS,
DOANE, SWECKER & MATHIS, George Mason Building, Washington & Prince Streets,
Alexandria, Virginia 22313,
Attorneys for Chevron Research, Company
RALPH L. FREELAND, JR., HAROLD D. MESSNER, Chevron Research Company, 525 Market
Street, San Francisco, California
View Table of Authorities
II. INTEREST OF THE AMICUS
Chevron Research Company is a wholly owned subsidiary of Standard Oil company of
California and owns the patent application involved in Diamond v.
Sherwood, No. 79-1941. The seismic exploration invention of this Chevron
patent application was held to be unpatentable subject matter under 35 U.S.C. §
101 by the Patent and Trademark Office (PTO). n1 This decision was reversed by
the United States Court of Customs and Patent Appeals (C.C.P.A.) and a petition
for a writ of certiorari was filed by the government. The petitioner requested
that consideration of the petition be held pending the decision in Diamond
v. Diehr and Lutton, No. 79-1112 on the basis of petitioner's argument
that the outcome of Diehr will control Sherwood. n2 The issues in
Diehr are related to those in Diamond v. Bradley and Franklin,
No. 79-855, which is to be argued at the same time.
n1 The invention relates to gathering and processing seismic data in which
events in the gathered field data (side by side wiggly lines) are identified
with exact locations of reflectors deep within the earth. Such a process and
variants thereof, provide a final pattern of reflectors that precisely indicate
likelihood of hydrocarbon deposits justifying the spending of millions of
dollars to drill a well. "[Standard Oil Company of California's] first well in
Hibernia. . . cost $ 33 million, the most Socal has ever spent on a single hole,
and it was drilled after a careful analysis of Mobil's seismic data. Notes
Alexander Massad, Mobil's executive vice president for exploration and
production: 'Obviously what their people saw [in the data] was something
different from what we saw.'" J. Quirt, "Socal Is Looking Homeward," Fortune,
March 10, 1980, at 73.
n2 In documents filed in Sherwood, Chevron has explained that the alleged
similarity is due to the petitioner's incorrect perception of the issues in the
Chevron's immediate interest lies in assisting the court toward correct
decisions on issues which may be of importance to Chevron's rights in
Sherwood. Also the decision sought by the government potentially would
impact adversely the validity of many already existing patents and diminish
future patent incentives in an important field of technology where Chevron has
long been active and where patents are significant. n3
n3 "Socal has raised its exploration and production budget again in 1980, and it
expects to spend somewhere between $ 1.7 billion and $ 2 billion." J. Quirt,
"Socal Is Looking Homeward," supra, at 67.
The issues of interest to Chevron relate to what kinds of inventins cannot,
because their subject matter is not embraced within the definition of 35 U.S.C.
§ 101, be patented even though they are novel and nonobvious. On these issues
Chevron supports affirmance of the holdings below to the effect that the subject
matter before the Court is within the statute. Chevron takes no position on the
novelty or nonobviousness of respondent's invention.
I. AUTHORITY TO FILE
This brief amicus curiae is presented by Chevron Research Company under
Rule 36.2 of this Court. Letters of consent from the parties are on file with
the Clerk of the Court.
III. SUMMARY OF THE ARGUMENT
It is fundamentally wrong to consider an invention unpatentable solely because a
programmed machine is used to carry out the invention and the program for the
machine represents an algorithm. Algorithms, as step-by-step processes for
obtaining a particular result, have never been treated as per se
unpatentable. Algorithms carried out by a programmed computer should not be
subject to such treatment.
The underlying rationale of the body of case law differentiating patentable from
unpatentable subject matter does not lie in the characterization of claimed
subject matter as algorithms, computer programs or the like. Two ideas are
involved. One is that, although newly discovered, things that have always
existed should not be removed from the public domain. The other is that abstract
or general ideas, as opposed to concrete solutions to technological problems,
fall outside the area of intended operation of the patent system. The granting
of patent protection on computer implemented inventions should not be foreclosed
on any other basis.
In the context of inventions that involve the use of programmed computers, the
C.C.P.A. has attempted to develop a reasoned approach to the initial inquiry
required by 35 U.S.C. § 101. Mindful of this Court's prior decisions, the
C.C.P.A. approach treats laws of nature, physical phenomena and abstract ideas
as though they were a familiar part of the prior art while treating the claimed
subject matter as a whole. Parts of claims are not ignored or set aside. Rather
the analysis takes into account the presence of the unpatentable principle or
idea in the claim through its relationship to other claimed structure or process
steps. The result is consistent with this Court's decisions; it rewards
appropriate inventions such as innovative applications of principles to
practical uses without removing the underlying principles from the public
A. Patentability of Computer Programs Is Not At Issue
The petitioner maintains that the respondents in Diehr, in Bradley
and in Sherwood are each attempting to patent a "computer program" and
that computer programs are unpatentable. In all of these cases the petitioner
has framed its questions in identical "computer program" terms.
This Court has never decided that computer programs are unpatentable. The
majority in Gottschalk v. Benson n4 specifically declined to do
so. Likewise, in Parker v. Flook, n5 the Court refused to so hold.
Chevron urges this Court not to decide whether computer programs per
se are or are not patentable, for such a decision is not required by the
facts of any of the cases now before the Court. The only common question
presented by the facts of Bradley and Diehr (and Sherwood)
is whether implementation of a process or machine invention by means of a
digital computer, in and of itself, renders the process or machine invention
unpatentable subject matter under 35 U.S.C. § 101. Chevron respectfully submits
that the answer to this question is, no, and that each case now before the Court
must be individually judged not by some unworkable standard based upon the
presence or absence of "algorithms", "equations", "programmed algorithms",
"computer programs" or the like, but by whether or not each patent application
seeks to remove from the public domain an unpatentable principle of nature,
physical phenomenon or abstract idea.
n4 409 U.S. 63 (1972).
n5 437 U.S. 584 (1978).
It is unfortunate that petitioner has made interchangeable use of words that
have different meanings. For example, "algorithm', "computer program",
"algorithm for a computer program" and "programmed algorithm" are used
synonomously in petitioner's briefs. n6 Yet these words are not synonomous.
n6 See, for example, p. 8 of petitioner's brief in Bradley where
the terms "computer program" and "algorithm" are seemingly used synonomously.
Further, on the bottom of p. 13 of the same brief the phrase "algorithm for a
computer program" is used. In its Diehr brief, the term "programmed
algorithm" is used at p. 11. A computer program is a series of
operational instructions in a language and form that are useable by a computer.
An algorithm, in general, is a logical series of steps that leads from
one point to a desired result. The algorithm in computer terms is not the
computer program; it is more akin to the outline for the operations which must
be performed by the computer to arrive at the desired result. It may be
expressed in words or mathematical formulae but is commonly expressed as a
combination of the two. The flowchart narrative and the computer program at
pages 17-18 of the Appendix to petitioner's brief in Diehr illustrate the
difference between a "computer program" and what might be termed the "algorithm"
underlying the program. Apparently, petitioner would use the term "mathematical
algorithm" to describe steps 33, 41, and 52 of the flowchart narrative.
Petitioner seems to use the word "equation", in its mathematical sense, as a
mathematical formula. In this sense, "equation" and "mathematical algorithm" may
One must be particularly careful to distinguish between a "computer program" and
what the computer does. The computer program is a series of instructions telling
the computer how to operate to produce a desired result. n7 The computer can
only manipulate binary signals (digits) so the operations may consist of
receiving and storing information in a register in the form of digits; receiving
and counting a series of digits and transferring stored digits from one register
to another on the appropriate count, or comparing the digits in two registers.
But in performing these operations, the computer may be amplifying a signal,
comparing an actual temperature with a desired temperature or performing a
variety of other process steps that are apparent neither from a program listing
of the steps the computer performs to complete each task nor from the digital
manipulations performed within the computer.
n7 In the Matter of An Application by IBM for the Revocation of Letters
Patent No. 1,352,742 in the name of Frederick Nymeyer, decided October 16,
1978 (Appendix A, p. 6a).
In the context of machine inventions, the Patents Appeal Tribunal of the British
Courts of Justice has likened the computer program to "a cam shaped according to
certain formulae so that, when fixed into a machine, it controls the latter in a
certain way" (Appendix A, p. 8a). A computer with a specific program in it is
described as "a piece of machinery which, by reason of the fact that certain
perforated cards are placed in it, is constrained to function in a certain way
and produce a certain result which is desired" (Appendix A, p. 9a). As such, the
British Patents appeal Tribunal views the computer when loaded with a particular
program as "a manner of manufacture" which never existed before. n8
n8 This case and others referred to in it were decided under the Patents Act of
the United Kingdom, Patents Act 1949, Chapter 87, Paragraph 101. Recent revision
of this Act may or may not have changed the test of patentability under British
law (Appendix A, p. 3a), since the Patents Act 1977, Chapter 37, Paragraph 1
declares that a "program for a computer" is not an invention for the purposes of
the Act "only to the extent that a patent or application for a patent relates to
the [program for a computer] as such." (emphasis added)
In reaching this conclusion, the British Tribunal recognized that the computer
per se was old and that the ideas are not patentable (Appendix A, p. 5a).
But it was also recognized that there is a difference between a "computer
program" and a "programmed computer", the former term being something akin to a
writing and the latter term being a machine having a fixed mode of operation. In
Diehr and Bradley, however, the petitioner has failed to
distinguish between programs, programmed machines and what the programmed
machine does (as opposed to how it operates). As a result, the petitioner is
suggesting that the Court must decide the issue of "patentability of computer
programs". This is a fundamental misconception. The real issue to be addressed
is whether in Diehr and Bradley (and eventually Sherwood)
the applicants for patent are claiming an abstract idea, a principle of nature
or a naturally occurring physical phenomenon, for these are the only categories
of subject matter this Court has ever declared unpatentable.
B. An Invention Does Not Become Non-Statutory Just Because A Digital Computer
May Be Used To Carry Out The Invention
An invention is itself a product of the mind, distinguishable as such from the
particular equipment used to practice the invention.
Implementation of a given invention usually can take a variety of specific
forms, all of which depend upon the state of technology at the time of
implementation. Machine implemented processes for curing rubber and apparatus
for controlling the internal operations of machines have always been patentable
subject matter. They should be no less patentable when a computer is involved
than when the work is done by some other means. The focus in determining
patentability should be on what is being done by the machine and not on how it
is being done. The C.C.P.A. recognized this when it stated: "The statutory
nature of a process or method [or machine] does not depend, under § 101, on the
means used to carry it out." n9
n9 In re Diehr, 602 F.2d 982, 987-988 (C.C.P.A. 1979).
It must be recognized that the programmable digital computer is merely a new
tool for doing what was formerly done with specific mechanical or electrical
hardware. The steps or machine functions performed by the programmed
computer are no different from those performed by hardware; they only appear to
be different because of the nature of digital computers. To use a programmed
computer, the functions to be performed must be described in the language of
mathematical operations carried out on data expressed as digits, since the
computer ".. . is characterized by its ability to process data, usually by
executing mathematical operations on the data at high speeds". n10 Thus, the
function of attenuating a signal which, in hardware, is accomplished by a
fixed arrangement of circuit components (e.g., an attenuator), may be carried
out on a digital computer by dividing a number representing signal magnitude by
another number representing the desired attenuation. Similarly, the function of
amplifying a signal, accomplished in hardware by a fixed arrangement of circuit
components (an amplifier), may be accomplished in a digital computer by
multiplying the signal (converted to a number so the computer can handle it) by
the desired amplification factor (another number). Thus, the operations
performed within the computer can be characterized as "mathematical", but to
characterize the functions performed as mere mathematical exercises is ".. .
misleading because it ignores what the computer is doing, concentrating instead
on how it is being done". n11 Signal processing (e.g., attenuating, amplifying,
filtering, etc.) is no less a technological endeavor when expressed and
performed with digital circuits on a discrete, mathematical, level than when
accomplished by fixed "analog" circuits.
n10 In re Walter, 618 F.2d 758, 764 (C.C.P.A. March 27, 1980).
n11 Id. at 769.
The petitioner apparently has never questioned the thousands of patents that
have been issued by the examiners in the Patent Office on circuits or
processes carried out by circuits that perform signal attenuation or
amplifying functions. However, when these very functions are performed by a
digital computer under instructions of a program that necessarily expresses
these functions in a different form, the petitioner would take these inventions
out of the technological arts and characterize them as unpatentable
"algorithms", or as "computer programs".
The Sherwood case illustrates quite clearly the danger of this biased
treatment of computer implemented inventions. Geophysicists have long used
seismic techniques to locate oil bearing formations beneath the ground. Patent
protection has always been available for inventions in this field. n12 It is
only recently as the use of digital computers has become more widespread in this
field that the patentability of such subject matter has been questioned.
n12 We are lodging with the Clerk of the Court copies of a number of patents in
the field and the PTO class definitions referred to herein.
In the early seismic signal processing patents, fixed arrangements of circuit
components (analog systems) are disclosed to perform the claimed methods. In the
later patents, digital processing techniques are used. The availability of such
techniques seems to have lead to the realization that there frequently is little
reason to use fixed circuitry to process seismic signals because the programmed
digital computer is faster, less expensive, and more versatile. The process
steps applied to this art (sonogramming, inverse sonogramming, etc.) are no more
"mathematical" when done by the computer than when done by hardware, yet
there is no room for a reasoned approach if everything done by the computer is
to be ignored as an unpatentable "algorithm" as in petitioner's test.
The Patent Office classification system officially embraces classes of subject
matter relating to seismic signal processing, both with and without computer
involvement. Class 367 -- "Communication, Electrical: Acoustic Wave Systems and
Devices" -- contains thousands of patents on seismic exploration and contains a
subclass (21) entitled "Signal Processing" (for processing signals
representative of acoustic waves to identify information in the signals
pertaining to underground formations). Class 364 -- "Electrical Computers and
Data Processing Systems" -- contains numerous patents in which programmed
computers are used in processes including, in subclass 420, processes wherein
the data processing system or calculating computer is designed for or used in
the study of the earth and/or related sciences. n13 In the official
classification definitions there is even a recognition that the nature of a
"functional system or element [of a claim]" does not change merely because it
"is simulated by a programmed data processor or calculator. . . . " n14
n13 U.S. Patent and Trademark Office, Classification Definitions, Classes 367
(Sept. 1979) and 364 (June 1977). There is even a subclass of computer
applications (subclass 400) defined as "Subject matter under the class
definition wherein the data processing system. . . is designed for or utilized
in a particular art device system, process, or environment. . . . "
n14 Id. See the suggestion that when searching for prior art one should
look for functional equivalency of computer elements and:
E. Search the class and subclass providing for a functional system or element in
addition to this subclass when the functional system or element is simulated by
a programmed data processor or calculator. . . . " Class Definitions, Class 364,
June 1977, p. 364-3.
The Patent Office examiners are prepared and able to examine applications on
these inventions. Such applications have long been examined and allowed. It
clearly would be wrong to exclude them now merely because the inventions are
carried out in practice through the use of digital computers. A more rational
approach that preserves subject matter always considered patentable must be
C. Petitioner Seeks A Rule Of Law That Would Preclude Patentability Of
Inventions Implemented Through The Use Of Programmed Digital Data Processing
What the Patent and Trademark Office seeks in Diehr is a decision
spelling out how the lower courts must analyze a particular type of invention in
determining whether it is of a character such that if new and nonobvious, it
might properly be patented.
Consideration of the Patent and Trademark Office brief in Diehr along
with its statements in Bradley and in Sherwood provides a fairly
clear picture of the rule it wishes the Court to embrace. This would require:
(1) identifying the component of the claimed invention that in practice would
involve the programming and use of digital data processing equipment; n15
(2) setting aside (i.e., ignoring) the component identified in (1) and examining
the remaining portions of the claimed invention in light of the prior art to
determine whether any inventive contribution is represented by these portions
without reference to the component identified in (1).
If (2) did not reveal an inventive contribution, the Patent and Trademark Office
would conclude that the claimed invention was of a nature such that a patent
could not properly be granted even if the claimed invention as a whole were new
n15 The term "computer" is used popularly with reference to such equipment, but
this suggests a connection with mathematics which does not in fact exist. The
data processed need not be mathematical data although the manner of processing
the data in a computer involves the use of numbers.
Petitioner purports to derive the proposed rule from the decisions of this Court
in Benson and Flook.
1. Mathermatical Expressions of Principles of Nature or Scientific Truths are
the Only Algorithms Which Should Be Treated as Being In The Prior Art Under
In 1890, William C. Robinson in his treatise Robinson on Patents
described the nature of the inventive act and exhaustively explored the bases
for the prohibition against patenting what we now call principles of nature,
ideas and physical phenomena. n16 Robinson recognized that everything in the
material world operates in accordance with natural forces (principles of
nature). He stated:
"In one sense, the word "principle" denotes the physical force employed by an
invention. The other appellations given to this force are very numerous, and
most of them are wholly inappropriate. It has been called "an elementary truth,"
"a principle of science," "a property of matter," "a law of nature," the "root
and ground of science;" but the idea which underlies these phrases is
sufficiently apparent, and is neither less nor more than that of some natural
power or energy, which operates with uniformity under given circumstances, and
may thus be contemplated as obedient to law. (footnote omitted). A principle,
in this sense, is a necessary factor in every means which produces physical
effects, whether such means be natural or artificial, and it is generally
this which makes the chief impression on the senses of the observer; but it is
in itself no true invention, nor can it be protected by a patent." § 135, pp.
193-5 (emphasis added).
n16 Robinson on Patents, Vol. 1, Section V, §§ 133-143 (1890).
The central thrust of the Benson and Flook decisions was a
confirmation, in the context of inventions involving computer technology, that
"laws of nature, physical phenomena, and abstract ideas. . . [are] not
patentable." n17 As the Court explained recently in Diamond v.
".. . Einsten could not patent his celebrated law that E (gravity. Such
discoveries are 'manifestations of. . . nature, free to all men and reserved
exclusively to none.'" n18
n17 Diamond v. Chakrabarty, 48 U.S.L.W. 4714, 4715 (June 16,
n18 Id. at 4715-16.
Although the Court in Flook "carefully scrutinized the claim at issue to
determine whether it was precluded from patent protection under "the principles
underlying the prohibition against patents for "ideas" or phenomena of nature',"
n19 the petitioner's brief here makes no reference to abstract ideas or laws of
nature and does not deal at all with the principles which foreclose patents in
these areas. Instead, petitioner has chosen, unjustifiably, to view Benson
and Flook as applicable to everything that might be termed an "algorithm"
or a "programmed algorithm", provided only that the algorithm be used in digital
data processing equipment.
n19 Id. at 4717.
In each of Benson and Flook the Court used the term "algorithm" to
refer to something specifically described as a "mathematical formula", n20 and
"algorithm" was said to mean "[a] procedure for solving a given type of
mathematical problem." However, the term is given quite a different meaning in
the petitioner's brief here without a clear explanation of either the fact of
the shift or the reason for it.
n20 Note, for example, the parenthetical usage in the Flook opinion of
"or mathematical formula" to qualify "algorithm": "Reasoning that an algorithm,
or mathematical formula, is like a law of nature, Benson applied the
established rule that a law of nature cannot be the subject of a patent." 437
U.S. at 589.
The invention at issue in Diehr does involve a mathematical formula, the
Arrhenius equation. But the Arrhenius equation is not the "algorithm" the Patent
and Trademark Office wants to have "set aside" (Brief for Petitioner at 7)
before a determination is made as to whether the claimed invention is of a type
which is ineligible for patent protection even if it is new and non-obvious. The
"algorithm" which the Patent and Trademark Office wants to set aside is the
series of method steps by which the patent applicant in Diehr sought to
operate a mold through a specified way of applying the principle expressed in
the Arrhenius equation.
This important distinction can be clarified perhaps by dwelling for a moment on
the technological aspects of the Diehr invention. The patent applicant
did not claim to have discovered the Arrhenius equation or that it could be used
in controlling a molding machine. Rather, the invention proceeded from a premise
that the relationship expressed in the Arrhenius equation could not be
applied directly to a molding process because real world molds did not in
practice operate at a fixed temperature throughout a curing cycle. How then to
obtain uniformly cured products in spite of the practical impossibility of
duplicating the ideal conditions under which cure time could be computed in
accordance with the known relationships governing rate of cure? The Diehr
answer was a method in which one concurrently (1) kept track of the time elapsed
after the closing of a mold and (2) made repeated measurements of mold
temperature external of the molded product itself and individual cure time
computations based on those measured temperatures. Each computed cure time was
in turn compared with the then existing elapsed time until finally there was a
match between elapsed time and the current computed cure time, and then the mold
was opened so the cured article could be removed.
It is this method, not the Arrhenius equation, that petitioner's brief calls a
"programmed algorithm [which] is respondents' only contribution." n21 Thus,
petitioner's real premise as to non-statutory subject matter relates not to "law
of nature" or "mathematical formula" considerations, but rather to the "fixed
step-by-step procedure for accomplishing a given result" n22 which the Diehr
patent application says should be carried out in practice through the use of
digital data processing equipment.
n21 Brief for Petitioner at 11.
n22 Brief for Petitioner at 6, n.12, Bradley (defining "algorithm").
Brief for Petitioner in Diehr, at p. 7, n. 4, refers the reader to the
Bradley brief for this definition.
Any doubt on this point is removed by reference to Sherwood, which
petitioner equates to Diehr. Although the invention there obviously is
far removed from a law of nature or a mathematical formula, petitioner again
calls a process an "algorithm" and contends that it is not statutory subject
matter for this reason. The preferred way of practicing the Sherwood
process is through the use of digital data processing equipment but the patent
application also describes in detail a system of electrically driven mechanical
devices (cams, gears, wheels, etc.) for carrying out the invention. The claims
cover both modes of implementation. n23
n23 The Sherwood claims at one time were indicated by the Patent and
Trademark Office to be allowable, apparently in momentary recognition that the
law provides no justification for a distinguising between electromechanical
implementation and digital data processor implementation of an invention. The
Patent and Trademark Office still does "not. . . suggest that 'a step-by-step
procedure for obtaining a given result' [citing source for this definition of
algorithm] necessarily is unpatentable as an algorithm when no computer program
is involved. In other contexts, a description of a patentable process could be
similarly characterized." Brief for Petitioner at 12, n. 18, Bradley.
Since a single invention cannot be both statutory and non-statutory subject
matter, the Patent and Trademark Office now refuses patents on inventions that
would not be challenged except for the fact that digital data processing
equipment is the best mode for commercializing them. That which would be viewed
as a patentable process if practiced only by electromechanical devices is
considered by the Patent and Trademark Office as an unpatentable algorithm if
digital data processing equipment is preferred.
The Sherwood invention is a new seismic signal processing system used in
locating oil and gas deposits. Within the field referred to as geophysical
prospecting, n24 seismic signal processing is an established technology in which
inventions have been made and patents have been granted for over fifty years.
n25 Its practical importance in these times of scarce oil is clearly
demonstrated by a comparison of Figures 7 and 8 from the patent application. n26
Figure 7 may be viewed as the starting point for the invention, a conventional
seismic time section, while Figure 8 represents the new product created by the
inventor's distinctive way of combining the many wiggly traces of Figure 7 and
shifting ("migrating") portions of the signals to new locations. This new
product shows, within the circle added to Figure 8, a "fault" location not
detectable at all from the pre-invention Figure 7 material. Locating faults is
of significance in finding oil and gas. n27
n24 Technology Assessment & Forecast, Eighth Report, December 1977,
Government Printing Office Stock No. 003-004-00550-5 [hereinafter cited as
Forecast ], prepared by the Patent and Trademark Office, outlines the state
of this technology with particular reference to the many patents that have been
granted and emphasizes its practical importance: "[G]eophysical exploration. . .
is an area of wide spread interest and impact. OVer $ 1 Billion was spent in
worldwide geophysical hydrocarbon exploration in 1976 alone; reflecting the
dependency of most nations on oil and gas (the demand for which is sharply
increasing as the supply is decreasing) to supply the major part of their energy
needs." Id. at 45 (footnote omitted). A copy of this Patent and Trademark
Office forecast is being lodged with the Clerk of this Court.
n25 See Forecast, supra, at 85-97, for a discussion of this subject
matter by Patent Examiner Moskowitz. It was Examiner Moskowitz who rejected the
Sherwood invention as nonstatutory because it "has as its best mode
application by programmed digital computer", professing a belief that "[t]the
policy basis for. . . [this Court's decision in Benson ]. . . precludes
patentability where. . . the invention comprises a program and algorithm
utilized in conjunction with a digital computer." Petition at 39a, Sherwood.
The disabling effect of this view of Benson on patent incentives in this
important technology is attested by Examiner Moskowitz, reporting that "[t]he
Benson decision resulted in a dramatic decrease in the number of
applications directed toward geophysical computer programs, algorithms and
processing formulae". (Forecast at p. 91, emphasis added). Hence, under
the PAtent and Trademark Office view of Benson, crucially important
technology which had been subjected to the stimulating influence of the patent
system now has been withdrawn from the reach of that system.
n26 These drawings are reproduced in Appendix B. Petition at 1a- 63a,
Sherwood, contains a reproduction of the claims and portions of the Sherwood
n27 Forecast, supra, at 86 mentions faults as one of several "structural
features which are associated with accumulations of gas and oil."
Promoting inventions of this type promotes the progress of the useful arts as
intended in the Constitution. "The patent laws promote this progress by
offering. . . an incentive for. . . inventiveness and research efforts."
Chakrabarty, supra, at 4715, where the Court adopted the expression from
Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470, 480 (1974) that "[t]he
productive effort thereby fostered will have a positive effect on society
through the introduction of new products and processes of manufacture into the
economy, and the emanations by way of increased employment and better lives for
Chakrabarty also noted as to § 101 of the statute: "The subject matter
provisions of the patent law have been cast in broad terms to fulfill the
constitutional and statutory goal of promoting "the Progress of Science and the
useful Arts' with all that means for the social and economic benefits envisioned
by Jefferson." n28 Recognizing that courts "should not read into the patent laws
limitations and conditions which the Legislature has not expressed", n29 the
Court found the micro-organism invention before it to be patentable subject
matter: "[The] claim is not to a hitherto unknown natural phenomenon, but to. .
. a product of human ingenuity. . . . " n30
n28 48 U.S.L.W. at 4717.
n29 U.S. v. Dubilier Condenser corp., 289 U.S. 178, 199 (1933).
n30 Chakrabarty, 48 U.S.L.W. at 4716.
Of course, the seismic signal processing system of Sherwood also is the
result of human ingenuity rather than a physical phenomenon that had "always
existed" n31 but only recently been discovered by man. The seismic traces
themselves are brought into being by human action and they are processed in a
manner devised by man to product a new technological product (Figure 8) having
qualities never present before.
n31 Flook, 437 U.S. at 593, n.15: "The underlying notion [why § 101 was
not intended to make the discovery of natural phenomena eligible for patent
grants] is that a scientific principle. . . reveals a relationship that has
This is not to say, however, that no mathematical equations are even involved in
the practice of the Sherwood invention. Algebraic equations that deal
with elemental phenomena (e.g., speed, reflection angles, etc.) of sound wave
movements in the earth are used in connection with some aspects of the
invention. Formulae such as these necessarily underlie engineering work of many
types. They represent fundamental relationships only, however, and do not tell
how the traces from many geophones can be combined and portions of the signals
migrated to produce a depth section product revealing fault formations and the
like. The Sherwood invention rests not upon a relationship that always
has existed in nature, but, rather, upon the human application of a variety of
operations combined in a particular manner to solve a technological problem. It
simply is not relevant to patentability to inquire as to whether man, in
addition to devising the process itself, also has devised a mathematical
description of it.
Nevertheless, in Sherwood as in Diehr and Bradley the
petitioner argues in "algorithm" terminology that suggests a close relationship
with mathematics. The "computer-implemented algorithm" n32 is not a mathematical
formula at all but is rather an elaborate step-by-step procedure by which the
seismic traces are processed. It is, in short, that entire sequence of
operations for which digital data processing equipment would be used in a
preferred way of practicing the invention commercially.
n32 Petition at 7, Sherwood.
In neither Diehr nor Sherwood does the government give any
explanation for its attempted application of Benson and Flook
language, offered by the Court in the context of essentially mathematical
subject matteR, to different kinds of algorithms. However, the real position of
the government is exposed in Bradley where the PTO "did not. . . direct
attention to any specific formula. . . or. . . mathematical calculations. . . "
n38 but instead contended that everything done by a computer
(digital data processing equipment) is necessarily a mathematical operation. n34
n33 In re Bradley and Franklin, 600 F.2d 807, 811 (C.C.P.A. 1979).
The government repeats this contention in its brief to this Court in Bradley,
and erroneously suggests at p. 14, n.21, that the C.C. P.A. subscribed to its
view in later deciding In re Sherwood, 613 F.2d 809 (1980). However, the
passage quoted was written in a wholly different context. Incredible as it may
seem, the PTO before the C.C.P.A. contended not only that the invention was a
computer program (and therefore unpatentable subject matter in its view),
but also that the patent application failed to satisfy the "best mode"
requirement of 35 U.S.C. § 112 because it did not disclose a computer program
for the preferred implementation of the invention. The passage quoted by
petitioner actually dealt with the best mode contention. The C.C.P.A. was
explaining its view that the Sherwood disclosure of an electromechanical system
for implementing the invention was so thorough that a person skilled in the art
would have no trouble programming digital data processing equipment to
accomplish similar operations if desired. The court was not discussing § 101 or
what subject matter is eligible for patenting or what operations are
This contention is without force. That mathematical expressions may be used to
describe and characterize the internal operations of digital data processing
equipment is meaningless. Words also can be used to describe the same events,
and equations can be used to describe all sorts of things. As the C.C.P.A.
"[The PTO] analysis confuses what the computer does with how it is
done. It is of course true that a modern digital computer manipulates data,
usually in binary form, by performing mathematical operations, such as addition,
subtraction, multiplication, division, or bit shifting, on the data. But this is
only how the computer does what it does. Of importance is the
significance of the data and their manipulation in the real world, i.e., what
the computer is doing. It may represent the solution of the Pythagorean theorem,
or a complex vector equation describing the behavior of a rocket in flight, in
which case the computer is performing a mathematical algorithm and solving an
equation. This is what was involved in Benson and Flook. On the
other hand, it may be that the data and the manipulations performed thereon by
the computer, when viewed on the human level, represent the contents of a page
of the Milwaukee telephone directory, or the text of a court opinion retrieved
by a computerized law service. Such information is utterly devoid of
mathematical significance. Thus, the board's analysis does nothing but provide a
quick and automatic negative answer to the § 101 question simply because a
computer program is involved." (emphasis in original). n35
n35 In re Bradley and Franklin, 600 F.2d at 811-12.
Thus, it will be seen that the subject matter petitioner seeks to have removed
from the reach of the patent system is not limited to any of the categories
which the Court has heretofore considered to be excluded. There is no
justification for concluding that every "step-by-step procedure for obtaining a
given result" (petitioner's meaning for algorithm) is a mathematical
formula, or a law of nature, or a physical phenomenon, or an abstract idea
whether in a programmed computer context or some other. Such a conclusion "would
have the effect of totally reading the word 'process' out of § 101, since any
process is a step-by-step procedure to arrive at a given result." n36 Nor is
there any reason for concluding that such subject matter is outside the
categories of patentable subject matter defined broadly by Congress to "include
anything under the sun that is made by man." n37
n36 In re Walter, 618 F.2d 758, 764-65, n.4 (C.C.P.A. March 27, 1980).
n37 S. Rep. No. 1979, 82d Cong., 2d Sess., 5 (1952).
The C.C.P.A. has attempted to develop an appropriate definition of the claim
"component" which Flook dictates be treated as being in the prior art.
The definition arrived at appears to be based on the C.C.P.A.'s observation
"[T]he common thread running through prior decisions regarding statutory subject
matter is that a principle of nature or scientific truth (including any
mathematical algorithm which expresses such a principle or truth) is not the
kind of discovery or invention which the patent laws were designed to protect."
[citing, inter alia, Benson ]. n38
n38 Walter, 618 F.2d at 765.
Thus, the mathematical expressions of principles of nature or scientific truths
are claim components to be treated as being in the prior art. This should not
mean that works of man or manipulations of man-made phenomena, though
expressable as mathematical formulae, are to be treated as being in the prior
n39 Thus, for example, in In re Sherwood, only mathematical expressions
of geophysical principles are treated as algorithms under the Walter
2. Ignoring Subject Matter Recited In A Claim Is Not Treating That Subject
Matter As Though It Were A Familiar Part Of The Prior Art
The petition filed by the government in Sherwood, No. 79-1941, contains a
succinct passage that displays the PTO view of how a patent claim containing a
reference to non- statutory subject matter is to be examined. Assuming that a
"computer-implemented algorithm" is non-statutory subject matter, the petition
describes as proper a PTO action that simply ignored the subject matter:
"[T]he Board properly applied the test laid down in Parker v. Flook:
it set aside the computer-implemented algorithm, found everything else in the
claims to be well-known, and concluded that the claims did not recite patentable
[statutory] subject matter." (Petition at 7, Sherwood).
Similar language is used by the petitioner in describing what the PTO did in
n40 "[The PTO] identified and put aside the computer implemented algorithm in
the claims and considered whether what remains is old in the art. . . . It found
that everything in Diehr's claim except that algorithm was 'conventional'. . . .
It concluded that Diehr was attempting to patent an unpatentable algorithm
embodied in a computer program. . . . " (Petition at 8, Diehr, No.
Page 7 of petitioner's brief in Diehr reveals the supposed support in
Flook for this PTO position to be the passage:
"Respondent's process is unpatentable under § 101 not because it contains a
mathematical algorithm as one component, but because once that algorithm is
assumed to be within the prior art, the application, considered as a whole,
contains no patentable invention. Even though a phenomenon of nature or
mathematical formula may be well known, an inventive application of the
principle may be patented. Conversely, the discovery of such a phenomenon cannot
support a patent unless there is some other inventive concept in its
application." 437 U.S. at 594).
Of course, we do not agree that what the petitioner labels a "computer
implemented algorithm" in Sherwood, is at all like what the Court in
Flook considered to be part of the prior art. The Court was talking about a
principle or phenomenon of nature (expressed by way of a mathematical formula),
and, the Court stated elsewhere in Flook the belief "that a scientific
principle. . . reveals a relationship that has always existed." n41 Thus,
the Court was treating as prior art only that which had always existed; it was
not licensing a trip into a fairyland where new subject matter can be treated as
old when in fact it never existed before.
n41 437 U.S. at 593, n.15 (emphasis added).
However, the difficulties associated with treating the new as old are
side-stepped by petitioner. Instead of considering subject matter as part of the
prior art, the only suggestion which could possibly be derived from Flook,
petitioner proposes to ignore the subject matter altogether. According to
petitioner, the claim limitations dealing with the computer-implemented
algorithm are to be "set aside" n42 or "put aside" n43 or
"separated from the rest of the claim". n44 After the limitations pertaining to
the computer-implemented algorithm have been excised from the text, what remains
of the claim must then be analyzed. n45
n42 Petition at 7, Sherwood; see p. 22, supra.
n43 Petition at 8, Diehr; see note 40, supra.
n44 Petition at 8, Bradley, No. 79-855.
Not even prior art elements are treated so harshly. Old elements are found in
nearly all patent claims, and there is no authority for ignoring them. Some,
perhaps all, inventions contain no new elements, but are rather new
combinations of old elements cooperating with one another to produce a new
effect of some kind. Petitioner's drastic proposal that one read out of patent
claims all those portions which would be implemented by digital computers
clearly cannot be tied to the teachings of Flook. Although treating
phenomena of nature as in the prior art, Flook not only failed to suggest
that limitations in patent claims to such phenomena be ignored but specifically
endorsed the principle that patent claims should be "considered as a whole." n46
n46 Flook, 437 U.S. at 594, n.16. See also the patent claim in
Mackay Radio and Telegraph Co. v. Radio Corp. of America, 306 U.S. 86
(1939), cited with approval in Flook.
Mindful of this Court's concern for treating claimed subject matter as a whole
even where mathematical algorithms are present, the C.C.P.A. has attempted to
develop an analysis "to establish the relationship between the algorithm and the
physical steps or elements of the claim." n47 This analysis includes evaluation
of the claim (including the mathematical algorithm) as a whole, and determining
"the mathematical algorithm is implemented in a specific manner to define
structural relationships between the physical elements of the claim. . . or to
refine or limit claim steps. . . the claim being otherwise statutory. . . . "
n47 Walter, 618 F.2d at 767.
n48 Id. Thus, for example, in In re Sherwood, mathematical
expressions of geophysical principles were not ignored but rather were
considered in relationship to the recited, stepwise manipulation of man-made
seismic signal data which was employed to produce another man-made physical
thing -- an improved map of subterranean formations such as Figure 8 of Appendix
This developing analytical format is consistent with Flook and
appropriately encourages progress in the useful arts by rewarding the innovative
applications of principles of nature or scientific truths to practical
technological environments, without sanctioning the removal of the underlying
principles or truths from the public domain.
D. Specific Processes For Dealing With Practical Technological Problems Are
Patentable Subject Matter Whether Or Not Digital Data Processing Equipment Is
Used In Their Practice
The fault petitioner finds in the Diehr and Sherwood inventions is
not related to constitutional and statutory provisions. The technologies of
rubber molding and seismic prospecting clearly are within the useful arts. n49
It is equally clear that there is nothing in 35 U.S.C. § 101 from which one
might conclude that Congress intended to exclude the processing of rubber or the
processing of seismic traces from the reach of the incentive system provided by
the patent laws. n50
n49 U.S. Const., Arl. 1, Sec. 8, Cl. 8.
n50 The terms of § 101 make "any new and useful process. . . [and]. . . any new
and useful improvement thereof" eligible for the grant of a patent "subject to
the conditions" of novelty ( § 102) and nonobviousness ( § 103), and the other
requirements of the statute. And, 35 U.S.C. § 100 says "process" means not only
a process but also a new use of a known process.
The government attack springs from terminological confusion that irrationally
equates all algorithms with phenomena of nature (and the other
terms, such as laws of nature, abstract ideas, and principles,
used interchangeably with it). Mathematical formulae get mixed up in the
middle, presumably because some mathematical formulae describe prenomena
of nature and because some algorithms are for solving mathematical
formulae. Digital computers get tacked on because of assumed relations with
The Chakrabarty n51 explanations of Flook point to a path through
the maze. The limits of § 101 are geared to "laws of nature, physical phenomena,
and abstract ideas" because these are "manifestations of. . . nature,
free to all men. . . . " n52 Whether a particular patent claim is directed to
nonstatutory subject matter is to be determined by scrutinizing it in light of
the principles which proscribe patents for abstract ideas and phenomena of
nature. n53 It would violate these principles if it were something that "has
always existed" n54 or if it were so general as to be outside the "useful arts."
n55 Other subject matter would pass the § 101 test and would then be evaluated
for novelty and non-obviousness under §§ 102 and 103 of the statute. n56
n51 Diamond v. Chakrabarty, supra note 17.
n52 Id. at 4715. Flook, it will be recalled, had itself explained
that a scientific principle reveals a relationship that "has always existed"
(437 U.S. at 593, n.15) and cannot therefore be new.
n53 Chakrabarty, 48 U.S.L.W. at 4717.
n54 Flook, 437 U.S. at 593, n.15.
n55 U.S. Const., Art. 1, Sec. 8, Cl. 8; 35 U.S.C. § 101.
n56 An early portion of the opinion of the C.C.P.A. in In re Bergy, 596
F.2d 952, 960 (1979), presents an accurate explanation of the relationship
between Sections 101, 102 and 103 of the Patent Law and the types of evaluations
appropriate to each of them. An endorsement by the Court of this explanation
would be a major contribution toward elimination of confusion in the law.
Under this approach the intent of Congress evidenced by the broad language of
the statute would be honored and the incentives of the patent system would be
restored in areas which the Patent Office now seeks to block off without
authorization from Congress.
The approach also would be consistent with the prior holdings of the Court. The
Court's concern in this area always has focused on laws of nature. Mathematical
formulae (and computations concerning them) have never been considered
objectionable outside this context of naturally occurring phenomena that have
always existed. n57 Patent claims with formulae or computation steps are not
precluded as long as they represent concrete solutions to technological problems
rather than abstract expressions of principles of nature or scientific truths.
n57 See Mackay, supra, note 46.
The decision of the court below is correct and should be affirmed.
WILLIAM L. MATHIS, FREDERICK G. MICHAUD, JR., BURNS, DOANE, SWECKER & MATHIS,
George Mason Building, Washington & Prince Streets, Alexandria, Virginia 22313
Attorneys for Chevron Research Company
Of Counsel: RALPH L. FREELAND, JR., HAROLD D. MESSNER, Chevron Research
Company, 525 Market Street, San Francisco, California
In the Matter of an Application by IBM for the Revocation of Letters Patent No.
1,352,742 in the Name of Frederick Nymeyer
PATENTS APPEAL TRIBUNAL
Royal Courts of Justice,
Monday, 16th October, 1978
MR. JUSTICE GRAHAM
MR. JUSTICE WHITFORD
In the Matter of The Patents Act, 1949,
In the Matter of AN APPLICATION by INTERNATIONAL BUSINESS MACHINES CORPORATION
for the Revocation of Letters Patent No. 1,352,742 in the name of FREDERICK
(On Appeal from the Decision of the Superintending Examiner (Mr. A. L. Pheasey)
dated the 19th April, 1977)
(Transcript of the Shorthand Notes of Marten Meredith & Co. Ltd., 36-38
Whitefriars Street, London EC4Y 8BJ.)
MR. T. A. BLANCO-WHITE, Q.C. and MR. P. PRESCOTT, instructed by Mr. Lawrence
Perry (IBM), appeared as Counsel for the Appellants (Applicants for Revocation).
MR. G. D. EVERINGTON, Q.C. and MR. N. PUMFREY, instructed Messrs. Stanley,
Popplewell, Francis & Ross, appeared as Counsel for the Respondent (Patentee).
MR. JUSTICE GRAHAM: Mr. Justice Whitford will deliver the judgment of the
MR. JUSTICE WHITFORD: This is an Applicants' Appeal on an Application to revoke
Letters Patent No. 1,352,742 standing in the name of Frederick Nymeyer. The
applicants, International Business Machines Corporation (IBM) have a
considerable business in the field of equipment used in data processing. This
includes apparatus of a type commonly referred to as computers. The title of the
patent is "Improvements relating to Data Processing". Claims 1 to 7 are directed
to a data handling system, claims 8 to 14 to a method of using a data handling
system, claims 15 to 22 to a method of programming a data handling system, claim
23 to a computer programmed in accordance with the method claims, claim 24 to
programming means in physical form suitable to programme a computer by any one
of the method claims while the remaining claims are for methods of processing
electric signals and the signals when processed.
The sole ground of objection argued on appeal was that the subject of the
principal claims is not an invention within the meaning of the Patents Act 1949.
The case was listed in the Patents Court but when it was called on a preliminary
question was raised as to whether we ought not to be sitting as the Patents
Appeal Tribunal. The hearing of appeals from the Comptroller under the
continuing provisions of the 1949 Act is dealt with in Schedule 4 of the
transitional provisions to the Patents Act 1977. Section 11(3) of Schedule 4
provides that where an appeal has been instituted before the Tribunal before the
appointed day and the hearing of the appeal has begun but has not been completed
before that day, the appeal (and any further appeal) shall be continued and
disposed of under the old law. 11(4) provides where such an appeal has been so
instituted but the hearing of it has not begun before the appointed day, it
shall be transferred by virtue of this sub-paragraph to the Patents Court on
that day and the appeal (and any further appeal) shall be prosecuted under the
old law, but as if references in the 1949 Act and rules made under it to the
Appeal Tribunal were references to the Patents Court. This appeal was instituted
before and was called on before the Patents Appeal Tribunal before the appointed
day. It was then adjourned to be heard before the Patents Judges sitting in
banc. We are of the opinion that under the transitional provisions the
appeal should be treated as having begun before the Patents Appeal Tribunal but
as, in any event, it was agreed on all sides that the old law applies the
question is of no importance so far as the issues of substance are concerned.
As was pointed out to us at an early stage, the test as to whether an invention
is patentable may, on the face of it, have been radically changed by the 1977
Act. We however have to consider the position under the old law. Our decision is
a decision under the provisions of the 1949 Act. We venture no opinion as to
what the conclusion under the 1977 Act might be. Nor do we think it appropriate
to take into consideration in reaching our conclusion any consideration of the
possible effects of the 1977 Act or the probable attitude of the European Patent
Office to claims in the field which we have to consider in the light of the
Guidelines for Examiners in the European Office touching computer programmes
which have already been issued and to which we were briefly referred.
The question as to whether all the more far reaching claims should be denied
protection has been simplified by the very reasonable concession made on both
sides that the matter can be tested on the first claim in the Patent under
The decision under appeal sets out on pages 1 to 3 the nature of the data
processing with which the specification is concerned so clearly that we shall
not attempt to improve upon it. We agree that it is plain that this patent
derives from an idea Mr. Nymeyer conceived as to the best way of determining a
selling price for, for example, a stock or share namely that this could best be
done by an automatic computation involving a comparison of buying orders and
selling orders so as to arrive at a price having the lowest buy order price
equal to or greater than the nearest sell order price.
Such an operation could in theory be done without the need for any automatic
aids but in practice needs to be automatically computed. Mr. Nymeyer envisaged
the making of the comparison and the determination of the price with the aid of
a computer and so arrived at a system involving in the language of claim 1:
1. "A data handling system suitable for establishing prices for a given kind of
fungible goods as hereinbefore defined in an auction market as hereinbefore
defined, comprising data storing means suitable for storing data representing
buy orders for the goods and representing sell orders for the goods, order
entering means suitable for entering data representative of individual buy and
sell orders including price information in said data storage means, comparing
means which in use read out and compare the prices of buy and sell orders from
the data storage means, by pairs chosen by progressing sequentially and
simultaneously through a descending sequence of the buy orders and an ascending
sequence of the sell orders by price, and means arranged to be controlled by
said comparing means so as to select the price at which the goods are to be sold
in dependence on the price or prices of the last compatible pair of orders which
are matched and so as to select each paid of said buy and sell orders to be
executed as a sale at a transaction price".
The question is whether this is an invention within the definition to be found
in section 101 of the Act which, together with section 6 of the statute or
monopolies, is to be found set out at page 5 of the Decision. The question is
whether the data handling system of claim 1 is a manner of new manufacture.
These words have over the years been considered and reconsidered in a great many
authorities. Counsel for IBM of course accepted that a data handling system,
being an apparatus having components which carry out the functions required by
claim 1 is a manner of manufacture. His fundamental objection was that in the
view of IBM claim 1 covers, inter alia, a standard computer and accordingly in
its present form cannot be said to be directed to a manner of new manufacture.
We were not taken through the body of the specification but counsel for IBM told
us, and counsel for Mr. Nymeyer did not dissent, that the specification does
envisage the use of a special comparator as part of the mechanism of the
computer and it is conceded that if claim 1 or any other claim were limited to a
computer embodying such a special comparator the objection upon the ground
advanced would not be open. He said also, however, and counsel for Mr. Nymeyer
again concurred, that at least one of the standard computers made and sold by
IBM would be capable of storing the relevant data, making the necessary
comparison and selecting the appropriate price if it were programmed to do this.
Now IBM do not on this application dispute the novelty of Mr. Nymeyer's concept
but they say, and in our view rightly, that his idea of selecting a price by
making a comparison of buying orders in descending order and selling orders in
ascending order is not a manner of manufacture. They go further, however, and
say that even on the assumption that the idea is new and not obvious it does not
become a manner of manufacture if you do no more than call in the aid of an
ordinary computer to do the work for you.
It was not suggested on Mr. Nymeyer's side that his idea could be the subject of
patent protection. The authorities to which we were referred show clearly that
you cannot get protection for a mere idea and we need not refer to them.
If claim 1 on its true construction protects Mr. Nymeyer's business scheme
however carried out it must be bad. If it can be so construed as to cover no
more than a standard IBM computer it must be bad. We do not doubt that the
standard IBM computer it capable of storing the relevant data (buying orders and
selling orders). We do not doubt that the standard computer is capable of making
the relevant comparisons. Indeed the specification of the patent seems to
acknowledge that comparisons of this kind are a standard operation (see page 16
lines 85-89). We do not doubt that the standard computer can make the relevant
selection but it is agreed that it can only carry out these operations when it
is programmed to do this. The claim only covers a system in which there are
means arranged to be controlled by the comparing means so as to select the
price. If a novel comparator is used in the computer to exercise the necessary
control no objection can be taken. If there is no special comparator then in our
view the claim on its true construction still only covers a computer when it is
programmed to produce the required result and does not cover a standard computer
as such. As we understand it, a programme is a series of instructions which are
fed to the other equipment in the data processing system to produce a desired
result. It is possible to have a computer in which a number of different
programmes are stored to be used as and when required. Alternatively a
particular programme may be fed into a computer from outside. These programmes,
written in a language the computer can understand, have taken different physical
forms. These forms have included punched cards, punched tape, printed directions
which can be optically scanned and now, more commonly, magnetic information
stored on tape, wire or drums. It is not suggested that before Mr. Nymeyer came
along there ever had been a computer programmed to carry out the operations
required in conducting his system. The Superintending Examiner has found, and
this we entirely accept, that once Mr. Nymeyer conceived his idea any competent
computer programmer could have told him how to programme a standard computer so
as to produce the desired result. As matters stand, however, until Mr. Nymeyer
came along there was no reason to suppose that anyone would have thought of
writing the appropriate programme and building it into a computer or otherwise
putting it into a physical form suitable for use with a standard computer.
A computer programmed to carry out Mr. Nymeyer's system must we think be
considered as being an apparatus having novel characteristics but IBM say that
in as much as it can be said that by the introduction of this new programme a
change may have been brought about this is not a change in manufacture. Counsel
for IBM put it in this way that such "newness" as there may be is purely
intellectual. This cannot in our view be right. There must be different holes in
the card or different magnetic patterns on the tape, or some other automatic
control imposed to ensure that the computer carries out the particular operation
We must turn now to certain recent authorities touching the question of
patentability of improvements relating to data processing. In Slee & Harris's
Application (1966 R.P.C. 195) a claim to a method of operating a computer
with a programme involving iterations characterised in that succeeding
iterations were initiated before completion of the previous iteration was
rejected, but a claim to a computer programmed to produce this result was
accepted as was a claim to means for controlling a computer (programming means)
to produce this result. The decision was a decision of the Patent Office. In it
the Superintending Examiner observes, in relation to the claim to a computer
programmed to produce the result, "It is directed to a machine which has been
set into such a condition that it can proceed to solve a particular programme by
undertaking a series of specified steps. It may be regarded as a machine which
has been temporarily modified. No objection should in my view be raised by the
Comptroller against such a claim on the ground that the invention concerned is
not within the statutory definition". He further observes on the same page, page
198 of the Report, in relation to the claim to programming means, "The means
claimed is more than a record of a programme, it is an integer which physically
co-operates with a computer to control the latter to operate in a certain way.
The means can, therefore, be likened to a cam shaped according to certain
formulae so that, when fixed into a machine, it controls the latter in a certain
Badger Company Incorporated's Application (1970 R.P.C. 36) was a case in
the Patents Appeal Tribunal. The claim under consideration was directed to a
method of producing a drawing with the aid of a computer involving a series of
steps as set out in the decision of Lloyd-Jacob, J. at page 39, line 33.
Lloyd-Jacob, J. held that the claim as originally framed was not directed to a
manner of new manufacture but suggested that it could be re-drafted in the form
of a claim to a process for conditioning the operation of a computer or a
computer of known type arranged to produce the solution and certain claims as
finally accepted are to be found set at page 41 of the report.
In the same volume in Gever's Application reported at page 91 the
Tribunal allowed a claim to:
"A data processing apparatus for recording the registration of word marks by
means of mechanographic supports arranged in a filing zone, the said word marks
being transcribed by replacing certain letters or combinations of letters when
occurring therein by other letters or combinations of letters" -- that is to
say, the letter Y by the letter I, the letters CI by the letters SI and any
double letters by a single letter -- "the said mechanographic supports being
filed alphabetically in order of the initial letters of the said word marks or
in order of the initial letters or a vowel or consonant sequences occurring in
In his judgment Graham, J. observes at page 95:
"As I understand it, what is claimed is a piece of machinery which, by reason of
the fact that certain perforated cards are placed in it, is constrained to
function in a certain way and produce a certain result which it desired. This is
at any rate partly achieved by the fact that the cards in question are
perforated in accordance with the information which it is desired to process,
and furthermore that such information is modified in the way specified in the
claim in that certain letters are altered on transcription. In other words, as
Mr. Hudson stated in propounding his step (1), it is fair to say that the claim
calls for apparatus which is programmed in a certain way so as to produce the
Having considered inter alia Slee & Harris and Badger Graham, J.
observes on page 98:
"In the light of these cases it seems quite impossible to say that the proposed
claim here or a claim along these lines to data processing apparatus, by reason
of the presence in it of certain perforated cards giving operating instructions,
constrained to function in a particular way and produce a particular result, is
not a manner of manufacture. Whether such a claim is fairly based on the
disclosure in the specification, whether it is new and whether it is obvious are
of course entirely different questions which will no doubt have to be decided at
some time, but the only matter before the Appeal Tribunal today is the question
of whether or not the claim as proposed to be amended is open to the objection
that it is not a manner of manufacture under section 101. To my mind, as I have
already stated, there can only be one answer to the question, namely, that it is
a manner of manufacture and I so hold.
"The second point which arises on the proposed claims is whether a punched card
having holes in it embodying information according to the claimed rules of
transcription can also be properly claimed as a manner of manufacture. To my
mind such a card is no different from the "means" in Slee & Harris's
Application or from the cam therein referred to at page 198, line 45. It is
to my mind clear that a newly invented lathe, the tool of which is constrained
by a cam to take a certain path in cutting, for example, a gear wheel, is a
manner of manufacture. It is equally clear that such a cam could be made the
subject of a subsidiary claim in a patent for the lathe. As sold, or as it lies
on the table, of course such a cam could only have a potential usefulness, but
it seems to be that it would be quite right that a claim should be able to be
made to such a cam because it is in the circumstances envisaged novel and
inventive and produces a useful result when it is placed in the machine for
which it is designed. It is exactly analogous to, for example, a new type of
carburettor which is necessary for the motor car for which it is intended. It
seems to me impossible to argue successfully that such a carburettor or such a
cam is not a manner of manufacture within the section.
"I cannot see that a punched card is in any different position. If the card was
in fact shaped so that it could be used in place of a metal cam its contour
being followed by some sensing device, it seems there could be little argument.
How can it make any difference that the card is punched instead of being
contoured? It is to my mind properly to be regarded as a manner of manufacture
since it is shaped in a particular way, namely, by having holes in it at certain
specified positions which ensure that it is capable when placed in its machine
of controlling that machine in accordance with pre-determined directions. In
this respect it seems to me to differ from a card which is merely printed and
which is intended to convey information to the human eye or mind but which is
not intended to be ancillary to some machine being specially shaped or
constructed for that purpose".
In Burroughs Corporation (Perkins') Application 1974 R.P.C. 147 the
Patents Appeal Tribunal sitting in banc accepted claims in these terms:
"8. A method of transmitting data over a communication link between a central
processor and designatable ones of a plurality of addressable remote data
terminals coupled by said communication link to said central processor, said
method comprising the steps of generating a sequence of inquiry messages at said
central data processors, transmitting said inquiry messages from said central
processor to said remote data terminals via said communication link in a
multidrop mode, storing terminal address data at each remote terminal, said
terminal address data designating the address of the remote data terminal at
which it is stored, temporarily storing received inquiry messages at each of
said data terminals receiving said inquiry messages, withdrawing from storage at
each remote data terminal receiving an inquiry message the terminal address data
designating the address of that data terminal, comparing said terminal address
data with an address portion of said inquiry message at each remote data
terminal, receiving said inquiry message to determine whether such received
message is addressed to the terminal receiving such inquiry message enabling the
addressed data terminal to respond to said inquiry message, and selectively
changing the terminal address data stored at at least one of the said remote
data terminals to selectively inhibit that data terminal from recognising an
inquiry message having an address portion corresponding to the original terminal
address data of said data terminal before said terminal address data was
"9. A method as claimed in claim 8 which includes the steps of recognising the
failure of an addressed data terminal to respond to a poll-select type inquiry
message addressed to it, and changing the poll-select type inquiry message
whenever an addressed data terminal fails to respond to an inquiry message of
one type addressed to it.
"10. A method as claimed in claim 8, which includes the step of transmitting a
control character from a data terminal to said central processor whenever said
data terminal receives an inquiry message addressed to it and its normal
terminal address data has been modified.
At this stage it had become the Office practice to allow claims to computers
programmed in a particular way and also to computer programmes embodied in such
physical forms as punched cards or tape. The objection to these particular
claims was that, being in form directed to a method of transmitting data or for
controlling a computer, as opposed to claims for a method of programming a
computer, the product to which they must be considered as being directed could
be said to be no more than intellectual information.
We do not propose to review again the authorities referred to in the judgment in
Burroughs' Application many of which were referred to on the present
hearing. We remain of the opinion which is to be found expressed on page 160.
"If a claim, whatever words are used, namely, whether the claim is, for example
for a method of transmitting data. . . ,' 'A method of controlling a system of
computers' or 'a method of operating or porgramming a computer. . . ', is
clearly directed to a method involving the use of apparatus modified or
programmed to operate in a new way, as the present claims are, it should be
We proceed upon the basis that the only thing that was novel in connection with
the present application was Mr. Nymeyer's concept of the way in which a price
could be fixed, but what he seeks to claim as a manner of new manufacture is a
method involving operating or controlling a computer in which, so far as the
contested claims are concerned, the computer is programmed in a particular way
or programmes in physical form to control a computer so that it will operate in
accordance with his method. The method is embodied in the programme and in the
apparatus in physical form and in our view the Superintending Examiner was right
in concluding that the claims should be allowed to proceed. It certainly cannot
be said of any of these claims that under the old law they are clearly not
directed to a manner of new manufacture.
We were additionally referred to two cases in the U.S. Supreme Court. In the
first of these Gottschalk v. Benson 1972 409 U.S. page 63 Mr.
Justice Douglas, who gave the opinion of the Court, observes at page 71:
"It is conceded that one may not patent an idea. But in practical effect that
would be the result if the formula for converting BCD numerals to pure binary
numerals were patented in this case. The mathematical formula involved here has
no substantial practical application except in connection with a digital
computer, 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.
It may be that the patent laws should be extended to cover these programs, a
policy matter on which we are not competent to speak."
In the case before us there is no attempt to patent a formula or its equivalent.
The claims are directed to apparatus arranged to operate in a particular way.
In Parker v. Flook 98 S.Ct., page 2522 the Court was divided as to
the extent to which a novel feature in the form of a mathematical formula or
algorithm could be protected. Mr. Justice Stevens observes, in giving the
majority decision of the Supreme Court, that the Supreme Court was considering
section 101 of the U.S. Patent Code and it is plain it was not considering the
United Kingdom Act of 1949. He specifically observes: "Respondent's process is
unpatentable under section 101 not because it contains a mathematical algorithm
as one component, but because once that algorithm is assumed to be within the
prior art the application considered as a whole contains no patentable
invention. Even though a phenomenon of nature or mathematical formula may be
well known, an inventive application of the principle may be patented.
Conversely the discovery of such a phenomenon cannot support a patent unless
there is some other inventive concept in its application".
We agree with the Superintending Examiner that under the Act of 1949 and the
decided cases the law is that an inventive concept, if novel, can be patented to
the extent that claims can be framed directed to an embodiment of the concept in
some apparatus or process of manufacture.
The appeal must stand dismissed.
MR. EVERINGTON: My lords, I ask for costs. By tradition, going back to, at any
rate, the last century, and the practice of the law officers, the costs awarded
are not on an indemnity basis but a contribution towards the costs. The only
thing I would say in this case is that the patentee here is an individual and
this is, as my learned friend has said, quite clearly a test case by a large
corporation and in those circumstances I do ask your lordships to make a
MR. PRESCOTT: First of all, to deal with the question of costs, it is my
understanding that our respective clients agreed beforehand that if IBM won they
would not ask for costs at all. We are astonished, however, to hear that they
are asking for indemnity costs in the event of us losing. I would simply observe
that this is an ordinary case in the Patent Appeal Tribunal and your lordship
should be invited to proceed in the usual way in making a contribution.
MR. JUSTICE WHITFORD: I did not understand Mr. Everington to ask for indemnity
costs. I thought he asked for a generous contribution.
MR. PRESCOTT: My lord, it is a case which is no more difficult or involved, in
fact less so because of the sparsity of the evidence required. It is a case
which has not taken up much time either here or below.
MR. JUSTICE GRAHAM: What happened in the court below?
MR. EVERINGTON: [*] 150, my lord.
MR. PRESCOTT: We would ask for the normal sort of scale in these case and that
it should not be treated differently.
MR. JUSTICE GRAHAM: We think the right order is [*] 250 costs.
MR. PRESCOTT: I have an application which I am instructed to make for leave to
appeal to the Court of Appeal. May I make two points in connection with that.
First of all, this is quite a different kind of case from the previous computer
programming cases which have been before this tribunal.
MR. JUSTICE GRAHAM: Why do you say that?
MR. PRESCOTT: We are here not concerned, as I think all the other cases were,
with matters of form, in other words, whether it makes any difference to frame
or claim a machine when programming as opposed to a programme. I think we
conceded at the outset that we did not care how it was put, a machine when
programmed or a card full of punched holes, it is all the same. The point of
substance we make is this: there is, with respect, no attempt here to
distinguish between two kinds of "newness" in punched cards, to take your
lordship's illustration. Your lordships made the point that if you have a
punched card which has holes in it and that pattern of holes is novel, then you
have got a novel article of manufacture and a manner of new manufacture. My
lords, there must be a distinction between a punched card with holes in its
which is new which is intended to feed a musical box and play a new tune and
whose only novelty lies or derives from the novelty of the tune, and a punched
card which might serve, for example, to control a machine to produce a new kind
of shaft or other article of manufacture.
MR. JUSTICE GRAHAM: You say there must be a distinction between those two?
MR. PRESCOTT: My lord, I submit so; otherwise the effect of your lordships
decision would be this, that all you have to do is to come before the Patent
Office and say, "I have a new tune which I thought up yesterday. I want to make
it suitable for a musical Box. Here is my punched card suitable for fitting into
a standard musical box. Nothing like this punched card has ever been seen
before, it is new, non-obvious, an article of manufacture and therefore I demand
to patent it". It goes further than that, because by parity of reasoning we must
extend the protection to have any good to the actual grooves in a gramophone
MR. JUSTICE WHITFORD: When you mention a musical box it is a box which is
intended to function in the same old way.
MR. PRESCOTT: In the same old way but in a totally new manner; so perhaps now we
are going to what we consider the distinction is. The point is there and I will
not elaborate it here, because after all your lordships have given a long and
reasoned decision. I do submit it is a point of first class capital importance
and leave should be granted.
MR. JUSTICE GRAHAM: The trouble is that this is under the old law. It is a pity
in a way that we are not acting under the new law. No doubt you will in the next
MR. PRESCOTT: The new law may well have effect for the next 20 years.
MR. JUSTICE GRAHAM: Yes, I know, but this is the old law.
MR. PRESCOTT: I am sorry, the old law may have effect for the next 20 years. It
may proceed some way into the future.
MR. JUSTICE GRAHAM: What do you say, Mr. Everington?
MR. EVERINGTON: As your lordship points out it is only a question under the old
law. The old law was tolerably clear and established by the cases your lordships
referred to. In my submission this really does not carry the matter any further
in this case and simply reaffirms what has already been settled as the old law.
MR. JUSTICE WHITFORD: None of it has ever gone to the court of appeal. It is
going to have effect probably to a number of cases.
MR. EVERINGTON: Yes, my lord.
MR. JUSTICE WHITFORD: So on those, if they ever come before the courts, the
decision of the Appeals Tribunal will not in any sense be conclusive certainly.
MR. EVERINGTON: That is right, my lord.
MR. JUSTICE GRAHAM: I think we can give you leave to appeal.
MR. EVERINGTON: May I just say this on the question of costs here. As I pointed
out my client is an individual, and the costs in the Court of Appeal will be the
ordinary costs awarded by the Court of Appeal. My lord, as this is a test case,
and the case of a large company taking an individual up before the Court of
Appeal, I would ask your lordship to consider whether leave should not be
conditional upon the applicants paying the costs in any event.
MR. PRESCOTT: I understood your lordships to have given leave.
MR. JUSTICE GRAHAM: That argument had not been put up. The order is not
MR. PRESCOTT: There are two aspects on this. As I said earlier IBM had already
decided even if it won not to ask for their costs here because we are an
international corporation of very great stature and do not want to be seen to be
pushing people around. It is one thing to say, if one likes, out of a sense of
generosity one should not do that, but it is another thing to be treated as a
matter of law --
MR. JUSTICE GRAHAM: It is not a question of a matter of law but what is the
right thing to do in the circumstances.
MR. PRESCOTT: There is no reason why we, in order to protect what we conceive to
be our rights, that we should be treated any different from any other litigant.
MR. JUSTICE WHITFORD: All right you get leave and nothing is said as to costs,
and maybe Mr. Nymeyer decides he cannot appear before the Court of Appeal
because of the burden of costs. I think it would be unfortunate in a case of
this kind which may vitally affect the interests of IBM in wholly different
cases that it should be argued before the Court of Appeal without the assistance
of counsel on the other side.
MR. PRESCOTT: Your lordship, if I may say so, is absolutely right. I wonder if
your lordships would consider making a recommendation to suitable quarters in
such a case that it might be a case for an amicus to assist the court in the
event of that situation transpiring. One must have regard to the distinction as
to what might happen in practice. I dare say in practice we would be horrified
if he threatened to drop the case because of lack of funds, but in reality as we
are here we ask not to be treated differently to anybody else.
MR. JUSTICE GRAHAM: I think we will adjourn for five minutes.
(Their lordships left the bench.)
MR. JUSTICE GRAHAM: The order that we make is as follows: we do give leave to
appeal and it will be appreciated we can not in any way tie the hands of the
Court of Appeal as to what they may think is the right course to take. This is
obviously an important case from the point of view of IBM and we recognize that
probably the first opportunity they have had of going to the Court of Appeal in
order to get the matter fully tested, but we do hope that they will consider
their position as a result of today with a view to making it clear to Mr.
Nymeyer that in the event of their prosecuting the matter in the Court of Appeal
they will not ask for an order as to costs. In any event, if Mr. Nymeyer feels
he is unable to take part in the proceedings in the Court of Appeal we think
that IBM should make it clear to the Court of Appeal what the position is and
ask them whether or not they would like to make an order that they have an
amicus curie to assist them in the Court of Appeal. We appreciate that this is a
matter which ought to be argued fully on both sides in order to arrive at a
proper decision. Is that clear?
MR. PRESCOTT: That is clear. Although we do not have definite instructions,
therefore I cannot definitely commit my clients, it would not surprise me in the
slightest if your lordships' comments fall on receptive ground.
MR. JUSTICE GRAHAM: That is what we rather expected you might say from people of
MR. PRESCOTT: If your lordship pleases.
FIGURES 7 AND 8 FROM THE SHERWOOD PATENT APPLICATION