Gottschalk v. Benson

LOCATION:Allegheny County District Court

DOCKET NO.: 71-485
DECIDED BY: Burger Court (1972-1975)

CITATION: 409 US 63 (1972)
ARGUED: Oct 16, 1972
DECIDED: Nov 20, 1972

Hugh B. Cox – for respondents
Richard B. Stone – for petitioner

Facts of the case

Engineers Gary Benson and Arthur Tabbot invented a faster and more efficient mathematical procedure for transforming the normal “decimal” type of numbers (base 10) into true “binary” numbers (base 2) which are simpler to process within computers. Their mathematical procedure was somewhat akin to long division, albeit with different steps. Their attorney argued before the patent examiner that the inventors were entitled to a broad patent covering any use of their new mathematical procedure, even use of it by a human using pencil and paper. The examiner rejected their invention. An appellate court overruled the examiner and ordered a patent to issue. The Commissioner of Patents then petitioned successfully to have the Supreme Court review this decision. Before the Supreme Court, the inventors’ attorney backed down from his earlier position and argued that the inventors were entitled to a patent covering all uses of their new mathematical procedure in computers, but not necessarily to its use by humans using pencil and paper. (The members of the Supreme Court at that time knew very little about computers.)


Is a computer program patentable? More specifically, is a mathematical procedure such as long division patentable?

Warren E. Burger:

We’ll hear this morning in 71-485, Gottschalk, Commissioner of Patents against Benson and Tabbot.

Mr. Stone, you may proceed whenever you’re ready.

Richard B. Stone:

Thank you, Mr. Chief Justice and may it please the Court.

This case which is here on in a somewhat unusual writ of certiorari to the United States Court of Customs and Patent Appeals raises the question whether respondents are in entitled to a patent on a method which they have devised for converting numerical information expressed in one form of mathematical language into another mathematical language.

Both of which languages are used extensively in general purpose digital computers.

Warren E. Burger:

Very well, must they not have the digital computer as part of this combination in order to make the whole thing meaningful?

The formula that process standing alone after a meaningful, is it?

Richard B. Stone:

No, our bet is precisely — our contention Mr. Chief Justice is that what respondents have claimed here is simply a set of steps to be carried out in a machine. They have tried to link the claim that they have made to the machinery a number of ways but our contention which I will shortly develop —

Warren E. Burger:

They’re interdependent, aren’t they?

Richard B. Stone:

Our contention is that the mathematical procedure which respondents are claiming.

The procedure for converting from one form of mathematical language into another is indeed entirely independent of the machinery.

Warren E. Burger:

Interdependent, I said.

Richard B. Stone:

No, independent is our — is precisely our claim and I will develop that shortly.

That is the basic thrust of our argument Mr. Chief Justice.

The underlying mathematical technology involved in respondents’ claim as explained in great length in our brief and in respondents’ brief and though there is some difference in emphasis, I think there is little if any significant difference between the Government and respondents with respect to the technological nature of the claim discovery.

Furthermore, though the technical background is set forth in our brief, I believe it would help place respondents’ claim in its proper context.

The technology necessary to an understanding of the legal issue in this case, is I think simply that it may appear and I will briefly describe here exactly what it is that respondents wish to patent.

A computer is a device which solves problems involving either numerical information or other kinds of data which can be broken down by logic into numerical form.

By far, the most common type of computer in operation today is the digital computer which its name implies, operates on information and data expressed the numerical digits.

The basic function of a computer is quite simple after a problem has been broken down into the mathematical steps necessary to solve that problem.

The computer computes the solution by actually doing the arithmetic, though it is an enormously elaborate and complicated and sophisticated device, the modern computer itself is really an extension in principle of the old adding machine or calculator.

Its utility lies in its ability to perform in minutes or even in seconds’ calculations which would require years to perform by hand.

Although the computer represents digits and numbers in physical forms such as for example by series of electrical pulses, the mathematical processes which the computer performs or the same which a human would perform except that they are expressed by means of the physical symbols built into the computer which uses electrical signals for example in a manner of similar to the way in which we use pencil and paper.

Thus, the machine is built with the capacity that carry out a wide variety of arithmetical calculations but though the machine is built to do the arithmetic it is told to do.

The machine can’t think.

It cannot solve a problem unless the operator breaks that problem down into a series of mathematical or logical steps for the computer to carry out.

This series of mathematical steps is the computers instructions or as it is popularly known in the trade of programming.

Though some computers are built to carry out one particular program, obviously, the greatest utilities in general purpose of computers which are built to perform a wide variety of programs requiring only that the problems be broken down into logical mathematical steps and translated to a language that is compatible with the internal physical characteristics of a computer.

This brings us to this precise subject of respondents’ claim.

In the great majority of general purpose of digital computers, the simplest and most convenient means of physically representing numbers is in switch light alternatives such as the presence or absence of an electrical signal of pulse for example analogous and perhaps more understandable terms to the on and off the light bulb.

Richard B. Stone:

For this reason, data used within the digital computer is ordinarily recorded not by means of our decimal number system but instead by means of the binary number system which expresses numbers and terms of only two characters 0 and 1 which can be easily correlated to the presence or absence of an electrical pulse the on or off the light bulb in which is based on powers of the number 2 rather than powers of the number 10 like our decimal number system.

This two characters as I say are easily represented in the computer for example by the on and off of an electrical pulse.

Now, the numbers that we deal with in our daily lives ordinarily in decimal form.

Therefore, much of the data which we feed into computers has to be converted from the Decimal Number System to the binary number system.

And an immediate step of this conversion process is the binary code decimal — binary-coded decimal number system which is a combination of binary and decimal numbers and that it expresses the ordinary decimal digits in binary numerals and arranges them in the order of the decimal number system.

Thus, the number 53 for example would consist of the binary representation of 5 in the left hand place and the binary representation of 3 in the right hand place.

It is frequently desirable to convert binary-coded decimal numbers to pure binary form.

Respondents have discovered a mathematical relationship between binary-coded decimal numbers and pure binary numbers.

They have translated this mathematical relationship or theorem into a mathematical procedure for performing the conversion of binary-coded decimal numbers to binary numbers.

What they are claiming in their claims are set forth to pages 3 and 4 of our brief and a numerous other places in the record is an algorithm, which the dictionary defines as any procedure for solving a given type of mathematical problem.

The basic step in respondents’ algorithm as we show in the appendix to our reply brief in the appendix to our petition is the repeated multiplication in binary according to the rules of binary by the decimal number 10.

Their claim would, if granted give them a monopoly over any computer program based on this mathematical relationship which they have discovered.

Now, what they claim is not exactly a computer program itself, it is a generalized formulation for programs which solve the mathematical problem of converting from one form of numerical representation, binary-coded decimal to another form of a numerical representation that is pure binary.

Now, let us deal with precisely what it is that the patent office and its Board of Appeals found wrong with the respondents claim.

The Patent Law defines patentability essentially in two aspects.

First, the statute sets forth at 35 U.S.C. 101 what constitutes patentable subject matter that is what types of invention are patentable.

Then there are other provisions notably 35 U.S.C. 102 and 103 which set forth conditions for the patentability of otherwise qualified subject matter and which are essentially that the invention be new, that it be useful and that it be non-obvious.

The patent office did not deal here with the issue whether the algorithm claim by respondents is new, useful, and non-obvious because it found that respondents in any of it or claiming a patent or non-patentable subject matter.

Under 35 U.S.C. 101, a patentable invention must be a “process machine, manufacturer or composition of matter.”

It is fundamental axiom of the patent law, originating in the writings of Thomas Jefferson and repeated in perhaps the most consistent theme throughout the patent decisions of this Court that ideas including scientific principles or mathematical formulas, mental processes and other abstract intellectual concepts are not patentable, only machines, manufacturers, composition of matter or processes.

Now, how exactly thus respondents claim for an algorithm for converting binary-coded decimal numbers into binary numbers fit into this fundamental distinction between a non-patentable abstract mental procedure and a patentable process machine manufacture or composition of matter.

Clearly their mathematical procedure is not a machine, manufacturer or composition as those three are all tangible products or substances.

The only question is whether their claimed invention is a “process” within the meaning of 35 U.S.C. 101 and therefore the starting point of this inquiry is what is the distinction between an abstract principle or mental procedure and patentable process, the statute merely defines process at 35 U.S.C. 100 to include “process art or method” which brings in the old terminology historically used synonymous with the word “process.”

Obviously, there is no sharp line between these two concepts.

I need that every process and I suppose every tangible products such as a machine or manufactured article embodies and can be explained at some level in terms of abstract principles.

But in the case of machines or other tangible products the patent is not granted on the idea but on the tangible object itself.

So that the ideas themselves are not monopolized except in the limited tangible form in which they are embodied in the machine or manufactured product.

In what circumstances then may patents be granted on processes?

In other words, where have the decided cases drawn the line between ideas and patentable processes?

There are essentially two lines of cases which I believe cover virtually all process patents that have been granted.

Richard B. Stone:

First, the patent office has granted and this Court has validated patents on processes which involve the chemical or physical transformation of tangible substances such as for example processes for the vulcanization of rubber to conversion of patent to soap or the grinding of flour as it used in the several case of Cochrane and Deener.

These cases and many of these are cited at pages 67 of our reply brief.

Consistently emphasized, the court’s conception of a patentable process is a series of acts to be performed on a tangible substance to change of its physical properties in some way and that concept is expressly articulated in the leading decision of Cochrane and Deener at 94 U.S.

The rationalities decisions is that when a process deals with the transformation of a specific substances by specific physical acts, a patent on that process has a finiteness, a forcibility of scope and a tangible quality that distinguishes it from a patent on a pure idea.

The other line of cases which must be considered here in the context of process patents are those 19th century landmark cases in which process patents were granted to Mr. Morse connection with his invention of the telegraph and to Mr. Bell in connection with his invention of the telephone.

The inventions claimed in these cases were series of steps performed on electric currents to produce physical reactions, in one case to print letters at a distance and in one case to produce sound at a distance.

In our view of these cases are really analogous to cases such as Cochrane and Deener which involved the physical transformation of substances because the invented element in these cases was in the physical manipulation of electric signals, in our case by contrast electricity is really an arbitrary means of symbol by which numbers in mathematical operations are physically represented.

The invented element in respondents claim is not in the physical manipulation of electricity but rather in the mathematical steps which the computer carries out under respondents instructions with the physical symbols that are already built-in to the computer but even assuming that the Morse and telephone cases were analogous to our case and avoiding a metaphysical argument over the tangible or intangible nature of electric pulses.

These cases granted claims only on those processes which were limited by reference to a specific apparatus in envious and this is the discussed at great length in our brief.

For this reason, Mr. Morse was denied a patent on his claim 8 — his claim 8 was not limited to operation through the machinery which he have described and Mr. Morse was going rather they patent on his first claim which was so limited.

And the court granted in Mr. Bells claims on processes carried out by means of the two specific apparatuses which he described.

For in this cases, as in that case is such as Cochrane and Deener on which patented processes for transforming physical and chemical substances, any monopoly that might have been granted on the underlying scientific principles was limited by reference to tangible apparatus.

And now finally, what is it that respondents’ claim here as a patentable process?

Respondents have discovered a relationship between binary-coded decimal numbers and pure binary numbers that enables them to convert from one numerical language to another by repeated multiplication according to the rules of binary multiplication by the decimal number 10 which is expressed in binary form as 1010.

The algorithm which they set forth in their claim merely translates into the language of computer programs a set of multiplications and additions by the binary equivalent of 10.

Though there is no time here to explain in detail why respondents’ discovery is no more than a procedure involving multiplication by binary 1010.

The Court need not except this conclusion on faith for it is fully develop in the appendix to our reply brief where we analyze respondents’ claims step by step.

The respondents have illustrated their claim at great length with complicated explanations and diagrams and charts.

They have not contradicted our analysis of the mathematical procedure that they claim nor had they denied that the algorithm described in their claim merely translates their mathematical discovery into the language of computer programming.

When the operator tells the computer to add, he uses words like shifting, masking and storing. These words do not change the mathematical theorem that respondents have discovered into a patentable process.

Any more than one changes 2+3 into a patentable process by calling it (Inaudible) or translating it into some other foreign language.

William H. Rehnquist:

Mr. Stone, don’t they at least still limit the scope of the patent to the computer field?

Richard B. Stone:

Mr. Justice Rehnquist, it is not entirely clear whether there claim is limited to operation in the computer field because the only words in that — in one of their claims they refer to shifting in a shift register and that is a word which is probably related at least to general purpose digital computers.

Their claim 13 has no such limiting reference.

I supposed if there are impracticality, their claim is likely to be carried out primarily into the digital computers because the computations are very elaborate.

But even if that is the case, we believe that the general purpose of digital computers is trivial limitation on the claim because it’s really no more than an extension of an adding machine or calculator.

It’s no more than the most advance device we have for calculating extensive mathematical calculations that are too — take up too much time to be done with pencil and paper.

But as a matter of fact the patent office felt that the respondents’ claims could be carried out by means of pencil and paper.

William H. Rehnquist:

Wouldn’t you agree with the question of Chief Justice that outside of the computer field there are simply no utility for that thing?

Richard B. Stone:

I would agree that outside of the computer field that there was no general utility known right now for this process other than carrying out calculations on a computer but that is such — data processing itself is such an abstract science and its really a — its the science of finding answers to mathematical problems and it’s an inherently, the information that is fed into a data processor is inherently non patentable in the most extreme sense in which we don’t allow patents on scientific and mathematical ideas.

Byron R. White:

Is the computer itself patentable?

Richard B. Stone:

Yes, the computer is certainly patentable and had respondents claimed a computer for carrying out this process which was new and useful and non-obvious, they certainly could have been given a patent but what they can’t get is a patent for —

Byron R. White:

Assuming they’re having done any computers until they invented one to make this conversion?

Richard B. Stone:

Well, I suppose they could have — if they have invented the computer, they could certainly have claimed the computer.

Byron R. White:

And you just think they haven’t claim the computer?

Richard B. Stone:

They haven’t claim the computer because they haven’t invented the computer —

Byron R. White:

Well, assume — assume if — if there haven’t been any computers and they invented the computer to carry out this process and that was patentable, why is that — why isn’t this against the background of the existing computers, why is this a new piece of an old machine?

Richard B. Stone:


Very simple, Mr. Justice White, it’s not a new use of an old machine because the computer is built to carry out all the calculations which this program calls for being carried out.

Byron R. White:

Assuming something it never did before although to put it that —

Richard B. Stone:

No more than an adding machine is doing something it never did before anymore when it adds a series of numbers that’s never added before.

Byron R. White:

But well, until this process was discovered, computers have never done this particular operation had it?

Until this came along, the computers never done this.

Richard B. Stone:

Well, Mr. Justice White the analogy which we used in our brief and I think that this is the appropriate analogy, is an old play of piano and which carries out, which plays songs when piano rolls are inserted into it.

We do not believe that the computer requires a new function every time it carries out new calculations that it is inherently built to perform anymore than a play of piano carries out a new function acquires a new use, every time a new piano rolls is inserted into it.

That is precisely the heart of our case.

The computer and the machine are one thing, many aspects of the computer and the machinery are patentable but the computer is built to carry out through physical symbols of electricity and other devices, all the calculations which this respondent has told that they carry out.

Byron R. White:

When a computer is programmed to give this answer it isn’t the same machine as it is when it isn’t program?

Richard B. Stone:

It’s precisely the same machine Mr. Justice White.

It’s precisely the same unit.

Byron R. White:

But why put in — why put in the instructions then?

Richard B. Stone:

I don’t understand what the connection between those two propositions is.

It’s the same machine because it is inherently able to carry out all of these calculations.

That is exactly what it is built and designed to do.

It’s the same machine even though it’s playing different music or carrying out different calculations exactly the same way that an adding machine is the same machine when it says 3+3 is when it says 3+2.

Byron R. White:

Well, you certainly do reduce the programs doing nothing.

Richard B. Stone:

I don’t reduce —

Byron R. White:

They don’t add a thing to a computer.

Richard B. Stone:

I don’t reduce into nothing, the math — the underlying mathematical discoveries that are involved with them maybe extremely ingenious and extremely important and extremely useful but they don’t add anything to a computer, they are mathematical calculation which a computer is already there to carry out.

Indeed, if respondent had invented the Pythagorean Theorem, and the Pythagorean Theorem were complicated enough so that it could only be carried out on a computer, I don’t think he would say that respondent was entitled to a patent on Pythagorean Theorem merely because it was likely to be carried out by computer or that the computer required a new function in carrying out the Pythagorean Theorem that it doesn’t have when it say computing the area of the circle.

Warren E. Burger:

Mr. Stone, when scientists and physicians and researchers have put together a known substances which perform a totally new function that is it will open up or close up particular valve in the heart for example.

May be a crude example but that type of thing, is that patentable?

Richard B. Stone:

That may be patentable if it meets all the necessary qualifications Mr. Chief Justice.

That is of course —

Warren E. Burger:

Well, now isn’t — what are they giving the patent on, the substances or the intellectual concept?

Richard B. Stone:

They’re giving a patent as — assuming that they asked — that it’s white — the device itself that they asked for patent on, they are given a patent on the device.

Warren E. Burger:

Well, the device is the sum total of the end substance that is the matter on which producers does result.

These are all known substances, aren’t they?

Richard B. Stone:

Yes, and they are cases if all they have discovered is that certain substances in the combination produces certain result.

It maybe that they don’t get a patent because a number of cases such as the Funk Brothers case decide 333 U.S. and the Armour case at 396 F.2d Armour versus Richardson-Merrell have made the point quite forcefully that the mere discovery of a better combination is useful is not patentable in itself unless they are some element of invention including that combination together or some synergistic relationship between the substances that makes for an invention other than the pure invention of the mathematical or scientific principle.

But our point in this is that the only element of invention in respondents’ claim, the only element of invention in this algorithm is the mathematical invention and the mathematical theorem itself.

There is no element of invention or whatsoever that takes place subsequent to the discovery of this mathematical process and therefore it is not unusual arts and not a patentable subject matter.

We have touched upon these subjects already in the questioning but the respondents basically described their mathematical procedure as a machine process and do a number of other and take another steps to describe their claims and ways that make it look like more than pure mathematics.

One which we’ve already touch upon and questioning by Mr. Justice White is that they imply that it’s a machine process rather than a mathematical process because it is in effect creates a new use for the machine.

As I’ve said these instructions which respondent has devised really contribute nothing and to the operation of this machine because the computer was always able to carry out these steps before the programs of the respondents have invented was ever devised.

Respondents have never denied this indeed they admitted in the court below when they stated in their brief that all of the recited single manipulations are conventional operations which performed by conventional electronic apparatus.

They also alleged the steps which they set forth are not strictly mathematical but a machine steps because they carried out by means of electricity in the machine.

As we have said, the invented element in respondents claim is entirely mathematical and has nothing to do with the electricity.

Respondents have not invented the technique for carrying out their calculations on the machine.

That was done by inventor of the digital computer.

Indeed, virtually everything that respondents claim as inventible other than mathematics and as patentable other than mathematics in this lawsuit was invented by the inventor of the digital computer and not by respondents.

Electricity as I’ve said is merely that writing medium was the computer uses to perform the mathematical steps that respondents have devised and no more imports patentability to respondents theorem to translate it into a computer algorithm of program anymore that it would import patent ability to the equation “A equals pi R squared” to translate it into a computer algorithm for determining the area of the circle.

Respondents also attempt to make a patentable process out of their theorem by referring repeatedly in their brief to the especial private exchange telephone system called the PBX which includes the digital computer apparatus that respondents alleged they — algorithm is especially designed to accommodate.

They seem to imply that they are claiming the patent on this mathematical procedure only as is used in this PBX device if indeed they’re not implying that their claim is in some way analogous to a claim on the PBX itself.

The fact is that their claim makes no reference or whatsoever to the PBX device or indeed to any other device and it brought a sense that can be read perhaps to refer cooperation on a digital computer but we do not believe that that limitation would give respondents invention and tangibility anymore than no limitation at all because if they have a monopoly on this procedure is carried out in a digital computer, they have an effect a monopoly on the theorem itself.

The thrust of our argument is that there is no invention in this process other than a mathematical invention.

Their invention adds nothing to the digital computer except in the sense that it gives the digital computer another operation to carry out which the digital computer is built to carry out inherently and which really amounts to know more than the normal inherent operation of that digital computer.

For this reason, we believe the decision reason below should be reversed.

I would save the remaining time for rebuttal.

Warren E. Burger:

Thank you Mr. Stone.

Warren E. Burger:

Mr. Cox.

Hugh B. Cox:

Mr. Chief Justice, and may it please the Court.

In our view, the question is added here is whether a method for converting electrical signals in an electrical data processing machine that signals that represent binary-coded decimal numbers into pure binary numbers is a patentable invention under Section 101 of the statute.

The only issue seems to be an issue of subject matter.

The claims were not rejected in the patent office on any grounds of utility, novelty or business.

Those points weren’t raised or decided in the court below as I understood in argument for counsel for the Commissioner this morning, he is not raising in those points in this Court.

So that I submit the question of subject matter has to be considered on the assumption that these claims are useful, noble, and non-obvious.

And I should like to begin by saying a word about the background with the nature of this invention.

The respondents here have made the invention because they were working on what was essentially a machine problem and they saw the problem by inventing a machine method.

They were working on a PBX and I think the PBX as I assume all, everyone knows is a switching device that connects a group of telephones one to the other and also connects those telephones with outside telephones.

It’s used in a hotel for example, connect all phones in the hotel in all zones and outside phones to use the same way in an office.

Now, one of the elements of this switching device was a control unit which was a kind of digital computer.

It received electrical impulses around the telephone making the call by electrical impulses that identified the telephone that was being called and then it connected the two telephones.

Now, to carry out to this function it was necessary in that control unit to convert binary-coded decimal numbers (BCD) numbers into pure binary numbers.

What the respondents are working on this problem there were numbers of quite a number of known ways of converting BCD numbers into binary numbers.

But all of these ways were unsatisfactory simply because of the physical limitations of the computer was which the respondents were working.

Some of the message required storage over in that computer of elaborate cables of conversion of values and then circuits to do what is called multi-digit adding.

Some of them required others special circuits to perform special functions.

Now the computer at least that the respondents are working with didn’t have storage capacity.

It didn’t have multi-digit adding circuits.

It didn’t have these others special circuits.

So their problem was to find a method of converting BCD numbers into binary numbers that was to be consistent and feasible and efficient within the terms of the physical limitations of this computer that they’re working with.

And they discovered such a method.

Now, we’ve described the method in our brief and the appendix to our brief, briefly simply will say this it consists of processing a BCD number in a series of repetitive cycles and do this to the electrical impulses in the machine and each cycle and electrical impulse is detected, discovered by another electrical impulse, its canceled and then two electrical impulses are inserted in pre-designated places in the original sequence of impulses that you are converting.

And when you have done that for each signal in the BCD, series of signals under this process, you then obtain a series of signals which have a representational value that is equivalent in binary to the representational value in BCD with which you stock.

Now, in discovering and inventing this method the respondents detected that there was a mathematical relationship that would be useful and I am emphasize that because the important thing here is the recognition that this relationship, this mathematical relationship could be useful in devising a machine method that would solve their problem and by the result of that recognition may devised this simple method of carrying out this operation and it had great advantages of course for them that made it possible for this computer they were working with to do something that it couldn’t do before.

And it has advantages for any digital computer, even though it has more elaborate equipment in the equipment in that computer with which they were working, it reduces a number of signals in the amount of circuitry you have to use and that reduces a possibility of errors through circuit malfunction which sometimes happens.

It means that you don’t have to use a lot of storage space in the computer even if you have which is often a convenience if necessary.

Now, as I’ve said and it sorted, it has advantages, substantial, synergistic advantages in any kind of a digital computer.

Now, as I have said, they were able to do to make devised this method because they recognized that this mathematical relationship could be used to advantage in a machine in converting this BCD rapper signals its impulses and the impulses representing pure binary numbers.

Hugh B. Cox:

But the fact in claim is on the machine method and not on the mathematics.

I want to emphasize that point because it’s a confusion that seems to me to run through the argument of the petitioner here.

There are two claims in this case.

They were both rejected by the patent office on the ground and the sole ground that they represented mental or they we don’t have to claim mental processors and mathematical steps.

The court below have reversed holding that they were claims on the machine — on the machine method that we submit to the court below was right.

We’ve discussed the language of those can’t — claims of great lengths in our brief.

I don’t propose that we tread that discussion here, I will simply say this.

But I think they’re really isn’t any room for argument about claim 8.

It’s stated in terms of a machine method no one would describe anything else in the terms of used in that claim and the Government, the petitioner doesn’t really argue that there’s any doubt that that claim is confined to a method to be practice on an electrical data processing machine computer.

They make some arguments about claim 13, we think those arguments are wrong.

We think when the claim is read as a whole and in the context it too like claim 8, simply claims a method that is to be practice on an electrical data processing machine.

But whether there is merit in those arguments are not they can’t — they don’t apply to claim 8 because they depend upon the language of claim 13, claim 8 stands on an independent basis.

Now, I should like now to examine the reasons of the petitioner here argues the same that this machine method is not covering, not patentable subject matter under Section 101 of the statute.

I don’t really think that petitioner is relying in any of degree on the conventional Patent Law Doctrine about mental steps.

He said this morning that the computer doesn’t think and I supposed there’s no about that when this method is practiced, it is a practice for manipulating electrical signals and when the steps describe in the claims are taken signals are detected, blanked out, transferred, inserted in other places.

It is a mechanical operation.

The rule against patenting mental steps is simply a rule that you can’t patent something, that goes on the line and the respondents here are not attempted to do that.

These claims are confined of the steps that take place in a machine.

I think what —

Byron R. White:

Mr. Cox, if the claims work or do machine, would these claims be in trouble?

Hugh B. Cox:

I think they might ask.

I think they will be in trouble.

Byron R. White:


Hugh B. Cox:

Well, I think they — I say — I have taken your word in troubles because I think that you would raised in a serious questions.

I think there is a rule and probably a sensible rule that you cannot get a patent on something that goes on simply in the mind.

Byron R. White:

Well, if these applicants had simply stated in their claims this method, this relationship, this way of converting, you would agree it would not be patentable?

Hugh B. Cox:

I think there would be a serious question there because I would doubt whether they have would have made the specific application over other than a useful arts.

See there’s a very old doctrine that I am going to say something about that moment about —

Byron R. White:

Well, it may still be very useful.

Hugh B. Cox:

It would.

Byron R. White:

I mean if the man could —

Hugh B. Cox:

Yes, but the trouble with it Mr. Justice White I think would be in position rather of arguing on the other side of this but the trouble would be that make specific application of the — of the thing of useful arts of claim may be so broad as to preclude any use of the formula told by even for purposes of a speculation or thought or something of that kind and I think that is one of the basis use the old rule that you can’t — you can’t patent a natural law as such or a principle as such question.

Respondents here haven’t tried to do that.

But I think there would be if they had — if this patent claims simply read, “We patent this method of converting numbers,” I think that it would be an arguable question whether they will be patentable.

But it isn’t here that question isn’t here.

Byron R. White:

Well if the, I would suppose it is — if there weren’t such a principle, the Government wouldn’t have any case at all?

Hugh B. Cox:

They wouldn’t have any case at all —

Byron R. White:

So the question is here I suppose?

Hugh B. Cox:

Well, the question isn’t here because respondents here haven’t made that kind of a claim, Mr. Justice White.

They have only claimed the use of —

Byron R. White:

Yes, but the Government is claiming you are.

Hugh B. Cox:

Well, they’re claiming we are but the fact here is that the claims —

Byron R. White:

What’s this case is all about?

Hugh B. Cox:

The claims simply do not extend that far and furthermore the Government doesn’t really want to win on the ground but the claims go too far what they want is a rule that exclude all this programs and we’ve any trouble about these claims.

Claims could easily be drawn.

They would put it beyond the doubt of even 17th century’s advances that these claims didn’t extend to anything except the use of this method in the machine.

Byron R. White:

What do you — what — do you claim this is a process, I take it?

Hugh B. Cox:

It’s process in the machine process.

Byron R. White:

Under what part of the definition of process in the statute do you think is called?

Hugh B. Cox:

Well, there isn’t any definition beyond the word.

It’s — but it is you could say it’s a new use about.

It’s a process for a new use of an old machine possibly although I really think that this point is developed both in our brief and the brief of others that when a thing like this is used in a computer, you so change the function of the computer that really become a new machine.

But I would say that this is process because it achieves by a result, a useful result by a series of machine steps that had not there before have been achieved in this manner.

Byron R. White:

Well, couldn’t you get a — then as you — as you put it you could get a patent on the machine once you put the program in there and you get a patent on that machine when it’s set up that way.

Hugh B. Cox:

Well, I am not sure of that and I think you could because the machine, you change the circuit – you change the circuitry in the sense temporarily.

Of course, if you had a problem why had Circuit did this?

I think you could get a patent on.

And in fact, the patent has been issued on by —

Byron R. White:

But its instantly, if the machine hasn’t been change then argue your dealing is using the machine to carry out a mental process, if it is in the machine, why would you need to process patent?

Hugh B. Cox:

Well, you need to process patent because the very — the very purpose of programs and the utility programs sends the inventiveness in programs lies in the fact that it makes possible to use a machine without changing the circuitry.

Hugh B. Cox:

You can change the machine without changing the circuitry by changing the way it functions it performs.

And the point is when you have a method, a process that does that.

It seems to me to fall squarely within the definitions as a process that they’ve been laid down by the decisions of this Court.

Section 101 simply speaks, uses the word “process” without trying to define it but there is a definition and I think in 100 which says that tries to give a little content.

But the — I think this might be inappropriate place for me to say something about the piano roll analogy which was brought into the argument a little while ago because I think the — precisely on this point that the piano roll analogy breaks down the — when the program does make a difference in a technological functioning of the machine.

It rearranges the way the signals operate, the way the circuits operate, so that there is a difference of that kind and now that the piano roll produces a different song but that’s the only difference between one piano roll or another, and the real, of course that kind of value is not protected by the patent law, it’s necessary value.

So that the — I think the piano roll is more comparable to the numbers of data that’s fed into a computer because the piano roll doesn’t actually change the function of what to play the piano does produces a different tune which is not subject to patent — protection at all.

But I want to emphasize at this point because I think there’s a confusion that’s involved that this patent claims do not attempt to claim the mathematics as such.

Anybody who wants to practice this method of mathematics but not the method because the method is series of machines steps but anyone who wants to use the mathematics to convert numbers by pen or pencil or by some other kind of a machine calculating machine or a machine that isn’t an electrical data processing machine can do so without infringing these claims.

The claims don’t cover that.

All these claims cover is this specific series of steps carried on within an electrical data processing machine.

Byron R. White:

Why didn’t any claim 13 say so?

Hugh B. Cox:

Well, we think it did say so, if may it please.

Byron R. White:

Well, it didn’t use those words.

Hugh B. Cox:

Well —

Byron R. White:

It would have been aptly easy to do so.

Hugh B. Cox:

If that is — Mr. Justice White, the respondents here had consistently taken the position and patent over all the way through that that’s what claim 13 meant.

And in the patent office they offered and tried to amend claim 13 to put in the words “electrical signals” to make it absolutely clear and patent office examiner wouldn’t allow them to do it because as far as I understand what he said, he said it wouldn’t make any difference.

You see, the real point in this case is the point that I think that was made this morning and its point that has no sanction as submitted decision, it’s the notion.

But you can’t get a patent here on these machine steps because the machine by mechanical means is doing something that the human mind could also do by using the same mathematics.

That’s what the argument comes to and I submit there’s no justification for it and it’s really inconsistent with the whole 150 years of development under patent laws because the essence of most inventions is substituting — or many inventions is substituting mechanical effort for human effort.

Of course, a mental by mental process you can do this conversion.

What the respondents here have done is to invent a — device a method we are doing it in a machine which enhances the operations.

The machine and enables the machine to do something more efficiently than what could be done simply by other methods.

Now, to speak of this to merely because the conversion has certain representational value represents symbols that means something for a given conventional value in a man’s mind does not seem to be suggested if there’s anything, any reason why it can’t be patented.

After all, communication depends upon the same kind of convention and when you talk over the telephone and communication is sensible only because the words “by convention” have certain meanings to the person who listens to you and that’s true in this case as to the number.

They are assigned conventionally to certain symbolic significance.

But that doesn’t — it seem to me have anything to do with the question whether the machine method for making these computation is patentable.

It isn’t a mental process.

The claims don’t cover the mathematics as such they don’t monopolize the mathematics, it simply claim on a machine method.

Hugh B. Cox:

Now, the argument I think that this is an attempt to patent a principal or a long nature simply misconceives the origins and the nature of that rule because when you read the cases, and it’s an old rule you see that those cases really stand for this.

There are cases that hold in the first place a man cannot get a patent on a principal in the general and the abstract and simply he must make a specific application of principle to the useful art and furthermore, he cannot get a patent on some application that he hasn’t invented.

Now, I submit that these claims here satisfy both tests.

The respondents discovered this mathematical relationship what is more important to determine that would be useful in device and machine method and the only thing that they are claiming is a particular machine method that they invented.

They are not claiming the principle except that it is applied by these steps and machine so they do not run or follow of either aspect of the rule against patenting a general principle.

There is a — I submit a kind of strained and artificial quality about all these arguments that are made against patent ability of these claims.

They used the conventional terms to patent law and the conventional concepts but when you look at the argument say do not those applications don’t simply don’t apply take the reliance upon the telephone cases.

Now, there’s one thing that is perfectly clear about the telephone case is that that case did not hold that a method patent had to be confined to use on a particular kind of apparatus.

What that case held was that if you had a solid method claim — method claim it extended to the use of the method on any apparatus.

That was used to practice the claim and in fact in that case, the court upheld that claim and held that it was infringed by the use of an apparatus of practice to method although it was an apparatus that Mr. Bell had not described in this patent as far as you can tell from the record and never even dreamed of.

The limitations that petitioners says must be put on a method claim of this kind are simply not required by the cases and indeed in certain respects their argument is puzzled because in their brief at least they seem to be more cautious in oral argument.

They suggested it may be this patent would be all right if it had been confined to use in a PBX.

That is really strange suggestion because the subject matter is the same whether it’s use in PBX or whether it’s use in all computers.

It’s no more or no less than mathematical axiom or a principle in the one case than it is on the other.

The point is that this is useful in all digital computers and was properly claimed in those terms not to claimed in any terms of any other machine would claim in terms of electrical data processing machine.

Now, all of these arguments really seem to me to be as I say I think they are restrained arguments because they are directed to an end.

They are examples though realizing trying to do the work of analysis what the petitioner wants and what the amici wants who were supporting him is a broad rule of law that will exclude from the Patent Act and the protection of Patent Act all computer processes and program.

And indeed, they presented that case in those terms in the petition for certiorari and I suppose that’s why the case is here because this Court will ordinarily not take cases on a certiorari simply or normal run affairs simply to look at the circumstances affecting a particular patent.

And I — therefore I think I am obliged to say something about the policy arguments if they advance to support this broad result.

Now, the first thing I am going to say and I’m just I am going to pass over it because we’ve developed it in our brief.

We think, it’s inconsistent with the both the (Inaudible) patent law, letter of the patent law, and 150 years of decision to exclude the whole field of technology in this way because the patent law has constantly been interpreted and applied as extending the new field of technology as they come along if the inventions meet the standard of the Act.

Beyond that, I submit that these reasons, these policy reasons which they advanced with a broad rule of law that they while will not withstand examination what they’re viewed in relation to the Patent Act itself.

The petitioner or the petitioner patent says it will burdensome and inconvenient if he has to search classify and examine the patents in this field and I submit that it’s hardly a good reason for a denying patentability to a whole area of patents.

Of course, it will involve some additional work any new technology does.

But the standards of Patent Act should not be bent or departed from for that reason and using for difficulties in any even or exaggerated for the reason set forth in our brief.

But that brings me to the other saying that I find myself obliged to say something about which is the economic reality that underlies this contest over whether these process should be patentable.

We examined the briefs in this case.

The brief amici, you will see that the patentability of this program is opposed by the group of hardware manufacturers.

They are large companies whose business is primarily the manufacturer of the computers although they also manufacture programs called software.

The patentability of the program is supported by the companies which are engaged in the business of software or making device and getting computer programs and process.

Hugh B. Cox:

These companies are for the most as compared to the hardware industry is a concentrated industry.

There are three or four companies in it and one of them IBM has by far the largest share of the market.

The software industry on the other hand is a divert — is a defused and diversified industry with a lot of companies in it most of them are small.

They are precisely the kind of enterprises that need the protection of patents in order to stimulate research, invention, to raise capital and protect those time they spent on developing these things of which.

And they are the source from which you might reasonably expect invention would come because the hardware manufacturers are far more interested in their machinery and sometimes these programs cut down the use of the machinery and they make simpler machine revisable.

They are far more interested in their machineries than they are in the program.

So the patentability of these computer possessors and programs, I submit, will be consistent with the policy of that law because it will provide protection and stimulation for invention from the various sources which needs the protection and from which invention may reasonably be expected to come.

So, I submit in conclusion that whether you look at this case in terms of the conventional standards of the Patent Act as they’ve been developed in 150 years of decision or whether you go beyond that and look at the terms of the policy argument that have been presented here, either view of the matter the decision of the court below should be affirmed.

Warren E. Burger:

Thank you Mr. Cox.

Mr. Stone.

Richard B. Stone:

Thank you Mr. Chief Justice.

I think the essence of the conflict in this case is reflected here in this oral argument as intensively as it can be.

And that is that respondents have attempted repeatedly throughout their briefs and in their oral argument to its current effect that the invention which they have come up with and which they claim in this case as an invention purely mathematical in nature.

They have used repeatedly essentially two methods to try to relate this invention to machinery on which doubtless there are many possible and permissible inappropriate patents.

First, they have as they have given that brief I refer it extensively to this PBX telecom switching system which they alleged that this patent was very specifically designed to accommodate because of certain problems and limitations of that PBX machinery itself.

Our response to this is quite clearly and simply that their claim makes no mention or whatsoever of the PBX device.

It has nothing to do as far as we can determine from the PBX device it granted it would cover the use of this mathematical procedure and any type of digital computer and would certainly not be limited to the PBX.

Even if it were limited to the PBX, the patent office would perhaps hear arguments as to why there was some special synergistic relationship between this program and the PBX device and if respondents could prove some synergistic relationship that would import an element of invention that went beyond the invention of their mathematical theorem that the patent office might be presented with a different case.

We are highly doubtful, however that respondents could prove such a synergistic connection between this algorithm and the PBX device.

In any event they have not claimed any limitation whatsoever to the PBX device and we are therefore unconcerned with the underlying motive for their invention and for their discovery of this particular theorem.

William H. Rehnquist:

Well, Mr. Stone you say that the application of the Government’s theory here would not then require outright rejection of the validity of the patent claim if it had been limited to the PBX.

There would be at least be a further factual —

Richard B. Stone:

There would at least be a – Yes, Mr. Justice Rehnquist, the patent office would certainly have heard the issue whether there was some invented element beyond the pure mathematical invention or some synergistic relationship between the claimed machine and the algorithm which would import patentability.

But that simply is not present in this case.

In addition, respondents have repeatedly used the word “machine process” which I reiterate brings up the implication that what the mathematical steps which they have asked this machine to carry out and instructed this machine to carry out import a new function of some sort of this machine and are related in some way to an invention on the machinery itself there.

And one of the amicus briefs there is given a list of — a description of a number of important theoretical discoveries involving mathematical steps used for example to predict whether which are so complicated and so long to carry out that they can only practically be done on what is today known as the digital computer are most complex calculating device.

If respondents are entitled to a patent on this mathematical procedure merely because it most likely to be carried out in a digital computer than all of those important mathematical discoveries, as well as other important mathematical discoveries which are likely because of their — of inherent length of the mathematical calculations to be carried out on a digital computer would be monopolized and taken out of the whelm of a common usage even though they have nothing inherently to do with machinery except that they are likely to be performed on machinery that has already been invented and designed to carry out this mathematical steps.

Warren E. Burger:

Mr. Stone, your time is up but did I understand you correctly to say that if these have been limited to the PBX or a single purpose machine of some kind, then the Government’s position would be different?

Richard B. Stone:

No, Mr. Chief Justice.

I have not said that.

Richard B. Stone:

All I have said is that if it had been limited to the PBX, then it would have been appropriate perhaps to make an inquiry as to whether there was some special relationship between respondents discovering in the PBX which would import an inventive element other than the mere mathematical discovery to their claim and perhaps warranted patentability.

I always say that this might have been a different case but that’s not the case before us.

And finally the competitive, the problems of the competitive status of the software industry which respondents have alluded to are indeed important problems which of course I think it is a matter of public record, the Government is quite concerned with.

Our view is simply, that the solution to these competitive problems is not to grant undeserved monopolies on mathematical principles.

Warren E. Burger:

Thank you Mr. Stone.

Thank you Mr. Cox.

The case is submitted.