Application of Charles D. Prater and James Wei

415 F.2d 1393 Application of Charles D. PRATER and James Wei. Patent Appeal No. 7987. United States Court of Customs and Patent Appeals. August 14, 1969. Woodcock, Phelan & Washburn, Virgil E. Woodcock, Philadelphia, Pa., attorneys of record, for appellants. James H. Littlepage, Washington, D. C., O. G. Hayes, New York City, D. Carl Richards, Dallas, Tex., Richard E. Kurtz, Philadelphia, Pa., James F. Powers, Jr., Dorchester, Mass., of counsel. W. Brown Morton, Jr., James W. Falk, Howard R. Popper, Washington, D. C., amicus curiae. D. D. Allegretti, Bair, Freeman & Molinare, Chicago, Ill., amicus curiae. Joseph Schimmel, Washington, D. C., for the Commissioner of Patents. Jere W. Sears, Washington, D. C., of counsel. Jacobs & Cohen, by Morton C. Jacobs, Philadelphia, Pa., Attorneys for Applied Data Research, Inc. and Association of Independent Software Companies, amici curiae, David Cohen, Philadelphia, Pa., of Counsel. Before WORLEY, Chief Judge, McGUIRE, Judge, sitting by designation, and RICH, ALMOND and BALDWIN, Judges. BALDWIN, Judge. — 1 — This appeal was originally argued1 before this court2 on May 9, 1968, and decided3 on November 20, 1968. On petition of the Commissioner of Patents, a rehearing was granted by this court4 on January 16, 1969, under the provisions of Rule 7 of this court. Briefs were filed,5 and the case was reargued6 before this court7 on March 3, 1969. This supersedes the decision and opinion of November 20, 1968, although various portions of the latter are repeated herein without specific reference thereto. — 2 — This appeal is from the Patent Office Board of Appeals decision affirming the examiner's rejection of claims 1, 6-10, 12, and 17-21, all of the claims in appellants' application,8 based solely upon considerations of law and statutory construction.9 No prior art references have been cited. THE INVENTION — 3 — The invention includes both a method10 and apparatus11 for the processing, or analysis, of conventionally obtained spectrographic data to produce a quantitative spectrographic analysis of a qualitatively-known mixture, for example a mixture of gases, by which the unknown component concentrations may be determined with minimum error. The raw spectrographic data is conventionally obtained in the form of a spectrogram, typically a continuous trace having a plurality of peaks. By conventional techniques, there may be derived for each peak an independent first order linear algebraic equation relating the peak height to the unknown component concentrations. Typically however, there are more peaks from which equations may be derived than are required to solve for the unknown component concentrations; that is, a set of equations, greater in number than the number of components, may be derived from the spectrogram.12 From that set, any subset of equations, equal in number to the number of components, may be selected13 to solve for the concentrations.14 The selected subset of equations may then be conventionally solved to produce a quantitative analysis. — 4 — Appellants have made a discovery which lies at the heart of their invention. Appellants have discovered: (1) that the different subsets of equations result in varying degrees of undesired "error amplification" in transforming the spectrographic data involving peak heights to the desired concentrations;15 (2) that there exists a certain relationship indicative of such error amplification; and (3) that that relationship is related to, and may be expressed in terms of, the determinants of the subsets of equations, the determinant of largest magnitude indicating the subset of equations involving least error amplification. "As far as this record shows, this discovery was new and unobvious."16 Prior to appellants' work, there was no systematic method for, or means of, selecting the subset of equations generating least error amplification. — 5 — Thus, based on their discovery, appellants have invented a method, within the otherwise conventional spectral analysis method, for selecting the optimum peaks providing that particular subset of equations least susceptible to error amplification. The essence of the method is finding the subset of equations having the largest determinant amongst all the possible subsets that might be chosen. Applicants have also disclosed in detail a machine17 for carrying out their invention. While the disclosed method and apparatus are described by reference to a special-purpose analog device, such device is visualized by appellants as but one device having the inherent capabilities to perform the desired functions; and it is disclosed that a general-purpose digital computer might also be used.18 — 6 — Claims 9 and 17 are typical of the method claims and read: — 7 — 9. In mass spectrographic analysis where, from a given sample of material there is generated a spectrum function having peaks therein spaced along a mass scale with respect to which the relationship between concentration, contribution factor of each of the m constituents of the mixture and the magnitude of each of the n peaks in said spectrum is represented by a set of m linear algebraic equations and where n is an integer greater than m, the method of selecting for analysis a set of m peaks least susceptible to error in concentration determination which comprises — 8 — dividing each said contributing factor for each peak by a normalizing function, — 9 — successively generating a determinant function for each said set of peaks, — 10 — successively generating output indications of the magnitudes of said determinant functions, and — 11 — selecting therefrom the determinant function of greatest magnitude for identification of said peaks least susceptible to error. [Emphasis added.] — 12 — 17. The method of determining with minimum error from the spectra of spectral analysis the concentration of the components of a mixture where the components are known and the concentration-determining peaks of the spectral analysis are present in number exceeding the number of said components, which comprises — 13 — generating physical representations of the magnitudes of the coefficients of simultaneous linear equations defining the concentrations of said components as functions of the heights of said peaks of said spectral analysis, — 14 — generating from said physical representations of the magnitudes of said coefficients the magnitude of the determinant of a plurality of sets of said simultaneous equations, the number of equations of each of said sets being equal in number to the number of said components, — 15 — comparing said physical representations of the magnitudes of said determinants of said sets of equations for identification of the set of said equations whose determinant has the largest magnitude, and — 16 — generating physical representations of the concentration of each said component of said mixture from said physical representations of the magnitudes of said coefficients of said set of simultaneous equations having said determinant of largest magnitude and from said heights of said peaks included in said last-named set of equations. [Emphasis added.] Claim 10, the only machine claim, reads: — 17 — 10. In spectrographic analysis where, from a given mixture of m constituents, spectral functions having peaks therein are obtained and wherein the relationships between the concentrations of said constituents and the peaks in said function correspond to relationships in a set of linear simultaneous equations, the system for selecting from said functions the combination of m of the n peaks therein least susceptible to error in concentration determination, where n is an integer greater than m, which comprises — 18 — means for generating a scalar function representative of the determinant for a first set of said equations corresponding with a first set of peaks in number equal to m, — 19 — means for generating successive scalar functions each representative of a determinant for sets of equations corresponding with different sets of m of said peaks, and — 20 — means for determining that one of said scalar functions of greatest magnitude for identification of said combination of m of the n peaks least susceptible to error. [Emphasis added.] — 21 — THE EXAMINER'S REJECTION AND BOARD'S AFFIRMANCE PROCESS CLAIMS — 22 — In his Answer, the examiner restated the rejection of the process claims for failure to comply with 35 U.S.C. §§ 101, 102, and 112. He preliminarily reasoned: (1) that, if the invention is within a statutory class of patentable subject matter under 35 U.S.C. § 101, it must be as a "process"; (2) that the appealed process claims are readable upon a mental process with the only physical limitation being the generation of physical representations, which may be merely pencil markings on paper; and (3) that processes having only mental novelty are unpatentable because they are outside the statutory "process" class, citing In re Abrams, 188 F.2d 165, 38 CCPA 945 (1951). The Answer continued: — 23 — At this point the rejection can take alternative forms, depending upon an apparently unsettled point of law, i. e., whether a claim is unpatentable if it reads on non-statutory subject matter in addition to statutory subject matter. The first form is under 35 U.S.C. 101 and 102, and the second form is under 35 U.S.C. 101 and 112. Under 35 U.S.C. §§ 101 and 102 — 24 — The examiner argued that, if a claim is readable on subject matter outside the statutory classes, then the claim is unpatentable under 35 U.S.C. § 101, it being irrelevant that the claim can also be read on subject matter within the statutory classes. The examiner felt that each of the present process claims can be read on mental calculations with the appropriate mathematics and, as the only physical steps, writing on paper. The examiner held that, to the extent that those claims require anything physical, they are fully met under 35 U.S.C. § 102 by the process of making marks on paper. The examiner concluded that the only novel subject matter recited in the claims must reside in the calculations which are outside the statutory classes of 35 U.S.C. § 101 and which, therefore, may not patentably distinguish the claims over the process of marking on paper. Thus, the claims were rejected under 35 U.S.C. §§ 101 and 102. — 25 — In regard to the examiner's 101-102 rejection, the board held that the process claims "do not fall within the statutory definition of a process as that term has been interpreted by the Courts over the years," citing In re Yuan, 188 F.2d 377, 38 CCPA 967 (1951) and Cochrane v. Deener, 94 U.S. 780, 24 L.Ed. 139. The board further stated: — 26 — It * * * [is] beside the point that the solution of the mathematical problem can be done by machine. The claims have set forth nothing which cannot be performed purely as a mental exercise using appellants' discovery that the equations having the largest determinant are the ones to use. Under 35 U.S.C. §§ 101 and 112 — 27 — Under the 101-112 rejection, the examiner assumed that a claim is within the statutory classes of 35 U.S.C. § 101 if it is readable on subject matter within that section, even if it can also be read on subject matter outside that section. The examiner argued, however, that, if a claim covers or reads on subject matter outside the statutory classes, as well as subject matter within the statute, then the claim fails to particularly point out and distinctly claim the invention as required by 35 U.S.C. § 112. That is, said the examiner, a reasonable interpretation of the process claims makes them appear to cover subject matter outside the statutory classes of patentable subject matter under 35 U.S.C. § 101, and claims which read literally on unpatentable subject matter do not particularly point out and distinctly claim the subject matter which applicants regard as their invention as required by 35 U.S.C. § 112. — 28 — The board did not specifically comment in regard to the 101-112 rejection of the process claims, apparently affirming the examiner. See Rule 196(a), United States Patent Office Rules of Practice. APPARATUS CLAIM — 29 — The rejection of apparatus claim 10 was also in two forms. Under 35 U.S.C. §§ 101, 102, and 103 — 30 — The examiner noted that, in addition to reading on appellants' disclosed analog embodiment, claim 10 may also be read on a properly programmed general-purpose digital computer. He argued that, having the principle of mathematics, which is unpatentable under 35 U.S.C. § 101 even though discovered by appellants, it would be obvious under 35 U.S.C. § 103 to program a general-purpose digital computer which is old under 35 U.S.C. § 102, such a computer having been known and in public use more than a year before appellants filed. Since, in the examiner's view, a properly programmed digital computer is thus not patentable to appellants, claim 10 was rejected under 35 U.S.C. §§ 101, 102, and 103. — 31 — The board viewed claim 10 as generally an apparatus counterpart of claim 17 and thus considered claim 10 unpatentable, citing In re Yuan, supra. That is, so held the board, the sole novelty in claim 10 resides in the mathematical computations which are themselves non-statutory subject matter under 35 U.S.C. § 101. Furthermore, the board felt that the physical "means" could be, and are anticipated under 35 U.S.C. § 102 by, such commonplace means as pencil, paper, and ruler, since the essential novelty of the claim is in the mathematical calculations. — 32 — Moreover, the board found that the essence of the rejection of claim 10 is that it defines nothing more than a general-purpose digital computer programmed to perform the required mathematical operations and that the programming would be obvious under 35 U.S.C. § 103. The board agreed that such a machine would be obvious. Under 35 U.S.C. §§ 101, 102, 103, and 112 — 33 — Here, the examiner argued that claim 10 does not particularly point out and distinctly claim appellants' invention as required by 35 U.S.C. § 112 since, in addition to reading "on a device (the analog embodiment) to which they are entitled to coverage," (emphasis added) the claim also covers or reads on a properly programmed general-purpose digital computer on which appellants are not entitled to coverage for the reasons set forth above. — 34 — The board did not specifically address itself to the 101-102-103-112 rejection of claim 10, again apparently affirming the examiner. THE SOLICITOR'S POSITION — 35 — The solicitor, in his original brief, stated that: — 36 — the Patent Office contends that all of the method claims read on a mathematician's use of pencil and paper, and that the "means" of apparatus claim 10 may correspond to those implements along with a ruler. — 37 — * * * Both the examiner and the board were concerned over the breadth of the claims,19 while conceding that appellants' discovery was apparently new and unobvious. [Emphasis added.] APPELLANTS' POSITION — 38 — In their original brief under the heading "The Facts Support Patentability," appellants contended that: — 39 — (1) They have made a discovery which has advanced the useful arts, one which the Board found to be "new and unobvious." — 40 — (2) They have developed a process for utilizing their discovery. — 41 — (3) They have invented a machine which makes possible the utilization of their discovery without human intervention. — 42 — (4) They have disclosed that their process, evolved from their discovery, may be practiced by a properly programmed general-purpose computer without human intervention. — 43 — (5) In the absence of prior art — and no references are relied upon by the Examiner or by the Board — all tests of patentability have been met. — 44 — Appellants further contended that the decisions relied upon below, namely, In re Abrams, supra, In re Yuan, supra, and Cochrane v. Deener, supra, are not here applicable. Appellants have not argued against the soundness of the results reached in those cases or of the so-called "mental steps" doctrine (i. e., the non-patentability of mental processes) as it may be based upon the Constitution,20 statutes, or case law, as they interpret them; rather they seek to distinguish the present case from the earlier cases and the "mental steps" doctrine, as interpreted by appellants. — 45 — Appellants point out that in both Abrams and Yuan, as contrasted with the present case, there was not disclosed a machine or apparatus for carrying out the invention automatically and without human intervention. Appellants argue that, where no such completely automated machine or apparatus has been disclosed, it is quite reasonable that certain steps of a process might be considered purely mental in nature. On the other hand, appellants urge that their disclosure of apparatus for carrying out the invention without human intervention precludes the process or any part thereof from being considered purely mental. In summary, they think that neither Abrams nor Yuan supports the proposition that a claim, limited to a process performed by apparatus by con- struing the claim in the light of the specification, is unpatentable simply because the process might conceivably also be carried out within the human mind without the apparatus. Thus, as we view appellants' position, they do not regard the appealed claims, when read in the light of the specification, as covering a purely mental process or a mental process coupled with pencil and paper markings, nor do they seek such claim coverage.21 — 46 — Insofar as Cochrane v. Deener is concerned, appellants urge that that case represented a reaffirmation or an extension, as opposed to limitation, of the subject matter which may be patented, Appellants feel that Cochrane v. Deener merely made it clear that a broad area of invention may be protected by process claims, not limited to any special arrangement of machinery, and that the decision did not set forth the metes and bounds of subject matter for which process claims may be granted. — 47 — As to the apparatus claim, appellants argue that it matters not whether the claim reads on a general-purpose digital computer, properly programmed to perform their invention. They urge that they are entitled to such protection inasmuch as such a properly programmed general-purpose digital computer was neither old nor obvious at the time of appellants' invention. OPINION PROCESS CLAIMS — 48 — We are in agreement with appellants that this case is not controlled by Abrams, Yuan, or Cochrane v. Deener, as we view those cases. Insofar as we can determine, this is a case of first impression in that the claims in a mechanical, as distinguished from chemical, case have been rejected because of their breadth, notwithstanding that no prior art reference has been cited and that the claims read on admittedly patentable statutory subject matter disclosed in the specification. — 49 — A more exhaustive analysis of Abrams may be found in Judge Smith's previous opinion, supra; however, for the present purposes, Abrams need not be discussed in such depth. In Abrams, no prior art reference was cited in the application entitled "Petroleum Prospecting Method," the claims being rejected as failing to define subject matter properly within the terms of R.S. 4886, the precursor of 35 U.S.C. §§ 101 and 102, in that "the steps in the claims which constitute the heart of the invention are purely mental in character." — 50 — The court referred to Abrams' specification in interpreting the steps of the claims, but Abrams had disclosed no means whatever for performing, without human intervention, two claimed steps of calculation and comparison. No analog device for carrying out the steps was disclosed in the Abrams specification; and at the time Abrams filed (April 28, 1944), general-purpose digital computers were still in the future. Thus, Abrams disclosed and claimed a process which could only be performed in the mind insofar as the teachings of the application were concerned. Abrams therefore presents a significant difference from the factual situation in the present case in which the teachings of the specification provide a full disclosure of at least analog apparatus for carrying out the claimed steps without requiring any steps to be performed in the human mind. — 51 — Yuan, decided about a month after Abrams, contained the following oftquoted statement: — 52 — This court has deemed it to have been thoroughly established by decisions of various courts that purely mental steps22 do not form a process which falls within the scope of patentability as defined by statute.23 [Emphasis added.] — 53 — But, as appellants point out, "Yuan's disclosure was the use made of equations by pencil-and-paper with the mind of the operator at work to interpret the results." Again, as in Abrams, insofar as the disclosure was concerned, the process (or the critical step thereof) was one that required the use of the human mind — indeed, a purely mental process or step. — 54 — That a process was an "art" capable of being patented was already considered beyond dispute24 in Cochrane v. Deener, involving a process or method of manufacturing flour. The issue was infringement, the defendant performing the process using apparatus of somewhat different construction from that disclosed by the patentee. The Supreme Court found the patent valid and infringed. In the course of its opinion, the Court produced the following, often-quoted passage of dictum: — 55 — A process is a mode of treatment of certain materials to produce a given result. It is an act, or a series of acts, performed upon the subject matter to be transformed and reduced to a different state or thing. If new and useful, it is just as patentable as a piece of machinery. In the language of the patent law, it is an art. The machinery pointed out as suitable to perform the process may or may not be new or patentable; whilst the process itself may be altogether new, and produce an entirely new result. The [Cochrane] process requires that certain things should be done with certain substances, and in a certain order; but the tools to be used in doing this may be of secondary consequence. — 56 — This passage has sometimes been misconstrued as a "rule" or "definition" requiring that all processes, to be patentable, must operate physically upon substances. Such a result misapprehends the nature of the passage quoted as dictum, in its context, and the question being discussed by the author of the opinion. To deduce such a rule from the statement would be contrary to its intendment which was not to limit process patentability but to point out that a process is not limited to the means used in performing it.25 See In re Ernst, 71 F.2d 169, 21 CCPA 1235 (1934). — 57 — Thus, it is clear that this case is not controlled by Abrams, Yuan, or Cochrane v. Deener. However, we do not feel that the distinction which appellants have very clearly pointed out is entitled to the significance which they would attribute thereto. Appellants feel that, since they have disclosed apparatus for performing the process wholly without human intervention and since they are seeking coverage only for the machine-implemented process,26 they have avoided the so-called "mental steps" doctrine. Although in view of our decision here we find it unnecessary to analyze and/or review in depth the so-called "mental steps" doctrine, it would appear that the disclosure of apparatus for performing the process wholly without human intervention merely shows that the disclosed process does not fall within the so-called "mental steps" exclusion.27 Of course, we have already pointed out that there is no dispute, as between the principal parties, that appellants have disclosed a patentable statutory process. — 58 — However, that is quite another question from the one before us, namely, whether appellants are entitled to claims of the breadth of those they seek here. As already noted, we view appellants' position to be that they are not seeking patent coverage of any purely mental process or any mental process coupled only with pencil and paper markings,28 but they are seeking coverage of the operation of a properly programmed general-purpose digital computer performing their process,29 as well as that of an analog device of the type disclosed. Appellants feel that they are entitled to a claim of the scope of claim 9 since, when read in the light of the specification, claim 9 does not cover a mental process. As to the other method claims, appellants feel "that generation of physical representations is conventional language in a step to be performed by a machine and absent any function of the mind in its performance," "that a physical representation includes more than mental operations," and that "[t]his takes it away from the purely mental concept." Claim 9 — 59 — With respect to claim 9, the broadest claim, appellants acknowledged in their original brief that: — 60 — Viewed apart from appellants' disclosure this claim is broad enough to cover the method if practiced by the use of pencil and paper by an operator, the operator directing the pencil. Such a person may divide each contributing factor by a normalizing function. He may solve for the value of the determinant for each set of equations. He may compare the results to select the set of equations having the greatest determinant for identification of that set of equations least susceptible to error. [Emphasis added.] — 61 — However, appellants urge that claim 9 cannot be read in a vacuum but instead must be read in the light of the specification. We agree. — 62 — Nevertheless, "reading a claim in the light of the specification," to thereby interpret limitations explicitly recited in the claim, is a quite different thing from "reading limitations of the specification into a claim," to thereby narrow the scope of the claim by implicitly adding disclosed limitations which have no express basis in the claim. This distinction is difficult to draw and is often confused by courts; but it is even more difficult for attorneys, attempting to work within the framework of the former, not to cross over into the latter. — 63 — In our view, appellants would really like us to read a limitation of the specification into the claims, not merely interpret the claims in the light of the specification. When read in the light of the specification, claim 9 does read on a mental process augmented by pencil and paper markings. We find no express limitation in claim 9 which, even when interpreted in the light of the specification, would support the conclusion that the claim is limited to a "machine process" or "machine-implemented process." This is particularly important in this case since the board noted that, in their brief before the board, appellants acknowledged that "[t]hough not practical for most of the needed applications, their method, theoretically, can be practiced by hand." — 64 — Inasmuch as claim 9, thus interpreted, reads on subject matter for which appellants do not seek coverage, and therefore tacitly admit to be beyond that which "applicant regards as his invention," we feel that the claim fails to comply with 35 U.S.C. § 112 which requires that "[t]he specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention." (Emphasis added.) This is true notwithstanding appellants' disclosure of a machine-implemented process. There are quite sound reasons why, in an infringement suit on an issued patent, courts may sometimes "interpret patent claims in the light of the specification" so as to protect only that phase of the claimed invention that constitutes patentable subject matter and thus do justice and equity between the parties.30 However, this court has consistently taken the tack that claims yet unpatented are to be given the broadest reasonable interpretation consistent with the specification during the examination of a patent application since the applicant may then amend his claims, the thought being to reduce the possibility that, after the patent is granted, the claims may be interpreted as giving broader coverage than is justified.31 In re Sweet, 393 F.2d 837, 55 CCPA 1191, (1968); In re Soderquist, 326 F.2d 1016, 51 CCPA 969 (1964); In re Tibbals, 316 F.2d 955, 50 CCPA 1260 (1963); In re Henatsch, 298 F.2d 955, 49 CCPA 915 (1962); In re Lundberg, 244 F.2d 543, 44 CCPA 909 (1957); In re Kebrich, 201 F.2d 951, 40 CCPA 780 (1953). We are not persuaded by any sound reason why, at any time before the patent is granted, an applicant should have limitations of the specifications read into a claim where no express statement of the limitation is included in the claim. — 65 — Thus, with respect to claim 9, appellants have not particularly pointed out and distinctly claimed the subject matter which they regard as their invention as required by 35 U.S.C. § 112, it therefore being unnecessary for us to consider 35 U.S.C. § 101 or its applicability here. Accordingly, the board's decision as to claim 9 is affirmed. Claims 1, 6-8, 12, and 17-21 — 66 — Claim 17, supra, is representative of the claims falling within this group. — 67 — In arguing that claims 1, 6-8, 12, and 17-21 do not read on a purely mental process or a mental process coupled with pencil and paper marking, appellants further advise us that: — 68 — Webster's Third New International Dictionary (1964) defines "physical" in the sense: — 69 — "* * * of or relating to natural or material things as opposed to things mental, moral, spiritual, or imaginary * * *" — 70 — Having given careful consideration to appellants' additional arguments concerning the proper interpretation to be given to claims 1, 6-8, 12, and 17-21, we feel that the rejection must nevertheless be affirmed. "Generating physical representations" appears to us to be broad enough, even when read in the light of the specification, to encompass pencil and paper markings which a mathematician might make in documenting or recording his mental calculations. Hence, since appellants do not seek such broad coverage as hereinbefore discussed, we think that claims 1, 6-8, 12, and 17-21 must also fail under 35 U.S.C. § 112 for the reasons expressed with regard to claim 9; with respect to those claims also, the board's decision is affirmed. APPARATUS CLAIM — 71 — Apparatus claim 10 presents quite a different question and, in our view, requires a different answer. As we see it, the underlying statutory basis for the rejection of apparatus claim 10 is 35 U.S.C. § 103 which precludes the grant of a patent if and only if "the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art." Appellants' discovery, discussed in the second paragraph under the heading "THE INVENTION," supra, is, it seems to us, part of their contribution to the art. On that basis, appellants' discovery should be considered as part of "the subject matter as a whole" and not part of the prior art. It is conceded by the Patent Office that that discovery is both new and unobvious. Thus, based on the record before us, we do not perceive any reasonable basis for concluding that "the subject matter as a whole," as defined by apparatus claim 10, would have been obvious at the time of appellants' invention. — 72 — We have carefully considered the basic position of the Patent Office that it would be obvious to program a general-purpose digital computer to practice appellants' invention and that apparatus claim 10 reads on such a computer, as well as the disclosed analog device. We find that position fatally defective in that it, in effect, assumes the existence as prior art of appellants' discovery that the relationship indicative of error amplification "is related to, and may be expressed in terms of, the determinants of the subsets of equations, the determinant of largest magnitude indicating the subset of equations involving least error amplification."32 Perhaps today, after reading appellants' disclosure, the public dissemination of which the patent system fosters and encourages, it might be obvious to program a general-purpose digital computer to practice the invention. But 35 U.S.C. § 103 requires an analysis of the prior art at the time the invention was made to determine whether the invention was obvious. Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966). Assuming the existence, at the time of the invention, of general-purpose digital computers as well as typical programming techniques therefor, it is nevertheless plain that appellants' invention, as defined in apparatus claim 10, was not obvious under 35 U.S.C. § 103 because one not having knowledge of appellants' discovery simply would not know what to program the computer to do. See Ex parte King, 146 USPQ 590 (Pat.Off.Bd.App.1964). — 73 — We do not perceive of any "mental steps" issue in regard to apparatus claim 10. It is quite clear that claim 10, in typical means-plus-function language as expressly permitted by the third paragraph of 35 U.S.C. § 112, does not encompass the human being as the "means" or any part thereof. Cf. Brown v. Davis, 116 U.S. 237, 6 S.Ct. 379, 29 L.Ed. 659 (1886); Republic Iron & Steel Co. v. Youngstown Sheet & Tube Co., 272 F. 386 (6th Cir. 1921); Permutit Co. v. Village of Poynette, 61 F.Supp. 305 (W. D.Wisc.1945) aff'd per curiam 158 F.2d 799 (7th Cir. 1947); Wilcox v. Danner, 53 F.2d 711, 19 CCPA 802 (1931); Mabon v. Sherman, 161 F.2d 255, 34 CCPA 991 (1947). The pencil, paper, and ruler — referred to by the board in regard to 35 U.S.C. § 102 — do not anticipate the claimed "means" since the former additionally require human manipulation. — 74 — Accordingly, the board's decision as to claim 10 is reversed. SUMMARY — 75 — As to claims 1, 6-9, 12, and 17-21, the board's decision is affirmed; as to claim 10, the board's decision is reversed. MODIFIED — 76 — WORLEY, Chief Judge (concurring). — 77 — The court is indebted to counsel for the respective parties and amici curiae for the able and earnest fashion in which they have, in both argument and briefs, presented their understandably diverse positions. — 78 — One of the more interesting points raised is how far Congress intended to go in conferring patentable status on mental steps as they are intertwined in computer programs generally. It is questionable whether prior decisions denying patentability of purely mental steps, or the statute, read singly or together, can support a broad rule either sanctioning or prohibiting the patentability of such steps in relation to computer programs. Where the line will be drawn can only be determined on a case by case basis in building, as best we can, sound and intelligible precedent. While I agree here with the results reached by my colleagues I do not necessarily subscribe to all that is said. — 79 — Congress, of course, had no way of knowing in 1952 what lay in the test tube then or what would become a reality tomorrow. But it devised a statute — a model of legislative craftsmanship and foresight — broad enough to anticipate and nourish the technological explosion we have witnessed. Our patent system is a delicate balance of interests; it protects the fruits of the extraordinary efforts it demands of inventors compatibly with the public interest. It has largely fostered the favorable climate resulting in the tremendous strides this country has made in reaching the pinnacle in the worldwide competition in the arts and sciences — from atomic energy to antihistamines, computers to catalysts, lasers to lunar landings.1 — 80 — So it is with no little surprise and concern that one learns of proposed patent "reforms". I can appreciate, from my own Congressional experience, the difficulty of enacting legislation acceptable to all concerned. However, after having dealt with the patent statutes, particularly the 1952 Patent Act, for nearly twenty years as a member of this court, I have grave misgivings concerning the desirability or need for any substantive change in a system that has worked so well in following the constitutional mandate "to promote the Progress of Science and useful Arts." Footnotes [1] Notes: [1] In addition to the principal parties' oral arguments and briefs, Bell Telephone Laboratories, Incorporated (Bell Labs), the real party in interest in Patent Appeal No. 8216, filed a brief amicus curiae [2] Consisting of Chief Judge Worley and Judges Rich, Smith, Almond and Kirkpatrick, Senior District Judge, Eastern District of Pennsylvania, sitting by designation [3] Reversed, with opinion authored by the late Judge Smith and joined by Judges Rich and Almond. Judge Kirkpatrick concurred in the result, and Chief Judge Worley did not participate in the merits of the decision. In re Prater, 415 F.2d 1378, 56 CCPA ___, (1968). The decision was handed down the day Judge Smith died. [4] Consisting of Chief Judge Worley and Judges Rich, Almond, Baldwin, and Kirkpatrick, Judge Rich dissenting with an opinion in which Judge Almond joined. In re Prater, 415 F.2d 1390, 56 CCPA ___ (1969) [5] In addition to new briefs of the principal parties and Bell Labs, International Business Machines Corporation (IBM) and Honeywell Inc. (Honeywell) filed briefs amicus curiae, and Applied Data Research, Inc. (Applied Data) and the Association of Independent Software Companies (Software Association) filed a brief amici curiae [6] The principal parties argued the appeal. All of the amici curiae, with the exception of Honeywell, presented brief oral arguments [7] Consisting of Chief Judge Worley and Judges Rich, Almond, Baldwin, and McGuire, Senior District Judge, District of Columbia, sitting by designation [8] Serial No. 155,236, filed November 20, 1961, for "Reduction of Data from Spectral Analysis," allegedly a continuation-in-part of serial No. 49,921, filed August 16, 1960, now abandoned. The real party in interest appears to be Mobil Oil Corporation, formerly Socony Mobil Oil Company, Inc [9] 35 U.S.C. §§ 101, 102, 103, and 112 are the relevant sections of the statute [10] Claims 1, 6-9, 12, and 17-21 are directed to the "method," or "process." See 35 U.S.C. § 100(b) and 35 U.S.C. § 101 [11] Claim 10 is directed to the "apparatus," or its statutory counterpart "machine," the terms being synonymous and used interchangeably here. See 35 U.S.C. § 101 [12] The set of equations is of the general form: Y1 = a11x1 + a12x2 + a13x3 . . . a1mxm Y2 = a21x1 + a22x2 + a23x3 . . . a2mxm . . . Yn = an1x1 + an2x2 + an3x3 . . . anmxm, where Y1, Y2 . . . Yn = peak heights of peaks 1, 2 . . . n, respectively, of the spectrogram, x1, x2, x3 . . . xm = concentrations of components 1, 2, 3 . . . m, respectively, present in the mixture, a11, a12, a13 . . . anm = constant coefficients representing the contribution that each component 1, 2, 3 . . . m, respectively, makes to the height of each peak 1, 2 . . . n, and n and m are both integers, n being larger than m. [13] "For example, [states the specification,] in a spectrogram obtained from the analysis of a 10-component mixture in which there are 20 peaks from which to choose, there are 184,756 possible subsets of 10 equations available." [14] Of course, if the spectrographic data were error-free, or exactly 100% accurate, then each subset of equations would yield the same concentration values. However, any electrical measuring instrument, including the spectrograph, has some degree of error in it and cannot produce error-free data. Thus, a variety of concentration values may be produced by the different subsets of equations [15] "Erroramplification," which occurs in the data analysis, is not to be confused with the measurement error discussed in n. 14, supra, or the magnitude of the latter. So it is that the mere use of the peaks having the most accurate heights per se does not ensure the most accurate concentration values. [16] So stated the board after it quoted from the examiner's Answer that: Appellants have discovered that those sets of simultaneous equations having the largest determinants are among those sets which can be solved with greatest accuracy. * * * Indeed, the solicitor has not controverted the novelty or non-obviousness of appellants' discovery. [17] As disclosed in the specification, the machine includes a battery-energized motor which drives the armatures of ganged rotary switch sections through a mechanical linkage. The motor is stopped when the solenoid of a relay is energized by the amplified output of a photocell which scans an oscilloscope screen and is responsive to a maximum signal appearing thereon. The signals produced on the oscilloscope screen correspond to the magnitude of the determinants of the several subsets of equations, and are brought together on the screen within the view of the photocell. Movement of the photocell downwardly across the face of the oscilloscope is accomplished by the aforementioned motor via a gear box; and, when the photocell photo-electrically senses the upper end of a determinant trace, its electrical response is amplified and it energizes the relay to stop the motor. At this point, the machine has identified the "determinant of greatest magnitude." Stopping of the motor simultaneously stops operation of the motor-driven rotary switches, and the values of the circuits which are established at that point are displayed on meters. The meter readings reveal the component concentrations The various mathematical coefficients to be dealt with in utilizing appellants' method are represented in the disclosed machine by various voltages in various circuits. In this manner, the coefficients of the relevant mathematical equations appear as electro-mechanical elements in appellants' machine. The specification sets forth in detail the electrical and mechanical components of this machine and their respective functions to relate the constant coefficients inherent in the present invention to a step in the method and to a means in the machine which directly performs the functions which underlie the present invention. The specification describes the operation of the machine in relation to the mathematical content of the invention and points out that, as the motor rotates the armatures of the ganged rotary switch banks through pre-set positions, the values of the determinants for the subsets represented by the positions can be ascertained. [18] The specification states that "the flexibility of a digital computer lends itself to the solution of problems of the type above described and in most instances represents an instrumentality preferred for the carrying out of the method of the present invention." Nevertheless, no program for a general-purpose digital computer, either in block diagram or machine language form, has been disclosed [19] Directing himself at rehearing to appellants' allegation that they havedisclosed a statutory invention, the solicitor orally stated, "Nobody questions that. The examiner conceded it, and the board has conceded [it]. But the question is, what do the claims define?" [20] The solicitor, in his petition for rehearing, argues that "the grant of a patent containing process claims of such breadth as to confer upon a patentee the right to exclude others from thinking in a certain manner" would run afoul of the First, Ninth, and Tenth Amendments to the Constitution. He urges that Article I, Section 8 must be construed in the light of the other constitutionally assured rights and that freedom of mind or thought may not be abridged by the patent laws. However, amicus Bell Labs points out "that a claim may verbally read on that which does not infringe without being itself invalid for that reason alone." Bell Labs notes, first, that thebreadth of a claim may be judicially restricted so as to be no broader than the actual invention when the claim is read in connection with the specification, suggesting that no court would hold purely mental activity to be an infringement even though the claim may "read on" such mental activity; second, that no court would construe purely mental activity to be structure or acts equivalent to the present machine and its function as described in the specification, 35 U.S.C. § 112, third paragraph; and, third, that purely mental activity would not be infringement since there would be no harm to the patentee, an essential element for any tort. In Bell Labs' view, freedom of mind or thought would not necessarily be abridged by the grant of a patent with claims broad enough to read on mental processes. As will be seen, we find it unnecessary to further discuss this aspect of the case. [21] In their Objections to the Grant of the Petition for Rehearing, appellants stated that this court had not held "that a disclosure of thought processes by way of equations and the like in the absence of any machine, system or apparatus making use of them would be patentable." Appellants go on: [T]he mere fact that a claim absent the disclosure is broad enough to read upon a pencil and paper execution is not per se a reason why that claim should be rejected. This does not say that a monopoly is in that way granted upon a pencil and paper execution. * * * Infringement is never determined by the language of the claim divorced from the disclosure. In the same paper, appellants stated that they "have not urged * * * that any claim will be infringed by one sitting at a desk and making mental calculations. There is no such thing as mental infringement." In fact, appellants argue that "Thought Is Still Unpatentable." Appellants' reply to IBM's amicus brief states that they "have contended all along that the meaning to be deduced from appellants' specification, as well as the dictionary, suggests that the method claims do not cover purely mental operations." [22] "Purely mental steps" are considered to be steps which may only be performed in, or with the aid of, the human mind. This is quite in contrast to "purely physical steps" which may only be performed by physical means, machinery, or aparatus. Purely mental steps (e. g., "believing") are quite different from purely physical steps (e. g., "heating") in many respects, not the least of which is that the former are much less susceptible to specific definition or delineation. Between the purely mental and purely physical ends of the spectrum there lies an infinite variety of steps that may be either machine-implemented or performed in, or with the aid of, the human mind (e. g., "comparing" and "determining"). In ascertaining whether a particular step is "mental" or "physical," each case must be decided on its own facts, considering all of the surrounding circumstances, to determine which end of the spectrum that step is nearer. It may well be that the step of "comparing" may be "mental" in one process, yet "physical" in another. Disclosure of apparatus for performing the process without human intervention may make out a prima facie case that the disclosed process is not mental and is, therefore, statutory. See Kayton, Patent Protectability of Software: Background and Current Law, in The Law of Software 1968 Proceedings B-25 (1968). Here, of course, the patentability of the disclosed process is not in question; but that is not to say that the claims delineate a patentable process. [23] At the timeAbrams and Yuan were decided, the Patent Act of 1952, under which we decide this case, was still in the future. Couched in terms of the present statute, it may well be that today purely mental steps are unpatentable because purely mental steps may, at present, be too vague and indefinite to comply with 35 U.S.C. § 112. See n. 22, supra. Whether or not a sequence of purely mental steps comes within the bounds of "process" as used in 35 U.S.C. §§ 100 and 101 is, we feel, an issue which has never been squarely decided. [24] "That a process may be patentable, irrespective of the particular form of the instrumentalities used, cannot be disputed." Cochrane v. Deener, 94 U.S. 780, 787. The word "art" appeared in the statute then in force and the word "process" did not. See R.S. 4886 [25] Further expansion of the law relating to process patentability followed shortly after in Tilghman v. Proctor, 102 U.S. 707, 26 L.Ed. 279 (1880) and The Telephone Cases (Dolbear v. American Bell Tel. Co.) 126 U.S. 1, 8 S.Ct. 778, 31 L.Ed. 863 (1887). See also Smith v. Snow, 294 U.S. 1, 55 S.Ct. 279, 79 L.Ed. 721 (1935) and Waxham v. Smith, 294 U.S. 20, 55 S.Ct. 277, 79 L.Ed. 733 (1935), and the analyses thereof in Judge Smith's previous opinion, supra [26] See n. 21, supra [27] See Kayton, Patent Protectability of Software: Background and Current Law, in The Law of Software 1968 Proceedings B-25 (1968) [28] See n. 21, supra, and text associated therewith [29] No reason is now apparent to us why, based on the Constitution, statute, or case law, apparatusand process claims broad enough to encompass the operation of a programmed general-purpose digital computer are necessarily unpatentable. In one sense, a general-purpose digital computer may be regarded as but a storeroom of parts and/or electrical components. But once a program has been introduced, the general-purpose digital computer becomes a special-purpose digital computer (i. e., a specific electrical circuit with or without electro-mechanical components) which, along with the process by which it operates, may be patented subject, of course, to the requirements of novelty, utility, and non-obviousness. Based on the present law, we see no other reasonable conclusion. [30] By construing a claim as covering only patentable subject matter, courts are able, in appropriate cases, to hold claims valid in order to protect the inventive concept or the inventor's contribution to the art. The patenteeat that time usually may not amend the claims to obtain protection commensurate with his actual contribution to the art. [31] This thought, in the public interest, is deemed to be paramount to an applicant's interest, since the applicant is not foreclosed from obtaining the proper coverage by express claim language [32] See the second paragraph under the heading "THE INVENTION," supra [34] Notes: [1-1] 1 Reminiscent of an old timer's lament that he was just beginning to grasp the principle of the flashlight when along came radio