dissenting respectfully.
The majority quite rightly identifies the dispositive issue in this case: whether there is a legally sufficient reason to combine the Guilford reference with the other cited prior art, thereby rendering Beattie’s claims prima facie obvious. That issue, of course, cannot be responsibly addressed without examining the teachings of the pri- or art.
We all agree that without the dodecatonic — i.e., twelve equal tone — system of notation found in the Guilford reference Beat-tie’s claims are not unpatentable under section 103. This is so because Beattie claims the combination of two radically different systems of musical notation on a single marker, not the mere mixing of numbers and letters on a marker to express the same system of notation. The majority correctly points to nothing in the Esche-mann or Barnes references that would provide a suggestion to combine them with Guilford. The majority cannot find such a suggestion in the Eschemann or Barnes markers because both registers of these markers are written in heptatonic notation, i.e., with seven primary tones and five secondary tones. There is no indication in either reference that these markers should or could incorporate the dodecatonic notation disclosed in Guilford. Consequently, if Guilford is to be combined with either Es-chemann or Barnes, it is because Guilford provides the requisite suggestion.
What Guilford teaches is a question of fact. Panduit Corp. v. Dennison Mfg. Co., 810 F.2d 1561, 1579 n. 42, 1 USPQ2d 1593, 1606 n. 42 (Fed.Cir.), cert. denied, 481 U.S. 1052, 107 S.Ct. 2187, 95 L.Ed.2d 843 (1987); Ashland Oil, Inc. v. Delta Resins & Refractories, Inc., 776 F.2d 281, 297 n. 24, 227 USPQ 657, 667 n. 24 (Fed.Cir.1985), cert. denied, 475 U.S. 1017, 106 S.Ct. 1201, 89 L.Ed.2d 315 (1986). Whether Guilford teaches toward combination with other pri- or art, or away from such combination is also a question of fact, Raytheon Co. v. Roper Corp., 724 F.2d 951, 960-61, 220 USPQ 592, 600 (Fed.Cir.1983), cert. denied, 469 U.S. 835, 105 S.Ct. 127, 83 L.Ed.2d 69 (1984), not a mere “notion” as the majority holds today for the first time. These factual questions go to the core of the obviousness analysis.
*1314Guilford itself and seven declarations from persons skilled in the art of music teaching are the only factual evidence in this case addressing whether Guilford teaches toward or away from combination with the Eschemann or Barnes references. Let us first examine Guilford and then turn your attention to the declarations.
The system of notation Guilford discloses is, in a word, distinct. For example, Guil-ford writes the first and last four measures of “Home Sweet Home” thus:
[[Image here]]
Guilford describes traditional pitch notation as “a perplexing and irrational system of sharps, flats, and naturals” such that “the same tone has many different symbols.” He finds the traditional five line staff bothersome because it notates, at most, only one and one-third octaves of the more than seven octave musical range. The traditional staff
necessitates the use of a greater or less number of ledger-lines above and below the staff, thus needlessly increasing the work of composition for the type-setter, besides consuming a large part of the page for only a few measures of the music, and, worse than all, it adds greatly to the difficulty of reading the music itself.
Guilford also states that “the present notation is just as defective in its means for expressing time as it is in its means for expressing pitch.” He finds the current system of notation “illogical” and that it does not express the modern conception of musical time. Guilford emphasizes that he “employ[s] a single uniform series of note-symbols” and that “the use of all so-called ‘accidentals’ [sharps, flats and naturals] is rendered wholly unnecessary.” Guilford dramatically sums up his invention stating:
[w]ith a view to overcoming the above deficiencies, crudities, and other faults as well I have devised a new notation which affords a simple and adequate means for expressing pitch, accent, and duration, which are the elements of modern music.
Guilford more than fails to suggest using the traditional system of notation in combination with his own system of notation as Beattie has done. Rather, Guilford actively rejects the traditional alphabetical notational system and proposes his system as a replacement. In so doing, Guilford removes any basis of finding a suggestion to combine his system of notation with any other system, including the traditional alphabetic system of notation used in the lower portions of both the Eschemann and Barnes markers.
The Board did not examine these unmistakable teachings of Guilford. Instead, it merely asserted that an artisan would have seen the need to merge Guilford with the other prior art. The Board thus committed clear error by misreading the factual content of the prior art.
When addressing the declaration evidence, the Board recognized that “each affidavit [contains] ... discussions ... that Guilford teaches away from [combination],” but dismissed the affidavits because they contained conclusions on the ultimate legal issue of obviousness as well. I do not quibble with the Board’s or the majority’s rejection of the experts’ conclusions on the ultimate legal issue. The Board and the majority, however, are not free to ignore declarations in the record from music experts who all agree that the seven-tone *1315approach to music and the twelve-tone approach to music are two incompatible concepts. In these declarations, the experts state, in part:
Although it is well known that music can be thought of as based on seven tones and it can also be thought of as based on twelve tones, Mr. Beattie is carrying this an important step further by showing how to combine the two schools of thought.... It appears to me that this is a significant attempt to combine two separate trends in music which had previously been assumed to be irreconcilable.... [T]he bias in Guilford is clearly toward doing away with all seven tone orientation, not toward finding ways to keep it around.... This, in my view, shows [Guilford’s] desire to eliminate any vestige of seven-orientation from his system which is the exact opposite of the seven-twelve reconciliation which Mr. Beattie discloses. (Joel Mandelbaum, Professor of Music, Queens College.) Since inventor Guilford’s twelve semitone numbers per octave stand all by themselves, and are not correlated in any way with the conventional seven scale degrees, it is clear to me that his objective was to discard the old seven-based system and replace it completely with his new twelve-based system, not combine the two. He offers positively no discussion or hint of making a “heptatonic/dodecatonic” combination. In fact the opposite is true, because he specifically says (page 2, line 93) that his system will make the use of accidentals “wholly unnecessary”, which is to say [Guilford] advocates eliminating the seven-based system from consideration. (Lawrence Widdoes, music instructor, The Julliard School at Lincoln Center.)
[Guilford] appears to have been engaged in a crusade to stamp out the traditional system with its “perplexing and irrational system....” (Paul Sheftel, music teacher, author, lecturer, pianist and composer.)
[I]t is fairly evident that Guilford intends his numerical arrangement to supersede, not reinforce, the traditional diatonic system with its sharps and flats which he describes as “perplexing and irrational” (p. 1, lines 25 and 26), and in fact this is a goal similar to that of the twelve-number systems used by music theoreticians.... (L. Poundie Burstein, faculty member at The Mannes College of Music and Preparatory School and teacher at Queens College.)
Teachings of the prior art simply cannot be combined when the prior art contains no suggestion or motivation to combine them. ACS Hosp. Sys., Inc. v. Montefiore Hosp., 732 F.2d 1572, 1577, 221 USPQ 929, 933 (Fed.Cir.1984). There can be no motivation or suggestion to combine when the prior art in fact teaches away from any combination.
In the face of uncontroverted materially dispositive factual evidence to the contrary, the majority zestlessly declares, ipse dixit, as did the Board, that it is obvious to combine Guilford with Eschemann. It never tells you why its conclusion might otherwise be correct. Furthermore, the majority simply avoids the factual evidence in the record that precludes combination. Rather than come to grips with the factual support the declarations provide for the proposition that Guilford teaches away from combination, the majority simply dismisses the declarations in their entirety as an unsuccessful effort to overcome a prima facie case of obviousness. Just because the declarations address matter pertinent to “secondary considerations”, see Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 694, 15 L.Ed.2d 545 (1966), does not justify calculated avoidance of all other pertinent factual assertions contained therein. Even the Board recognized that the declarations speak to the teaching away issue. Guilford tells the examiner, the Board and the majority not to combine his system of notation with the traditional system he despises, and those skilled in the art agree that Guilford teaches away from combination. These are facts that cannot be ignored.
The majority concludes that Beattie faced a “very strong prima facie case of obviousness.” There can be no such case here unless someone can find in “the prior art taken as a whole” the suggestion to *1316combine Guilford with the other prior art. The mere conclusory approach of the Board, which the majority leaves undisturbed, is unconvincing. I submit that Beattie provided the motivation to combine the two systems of notation. His genius resides in combining two hitherto alien systems of musical notation on one marker. Perhaps “simple as pie,” but not obvious under section 103.
I am at a loss to understand why the majority seems compelled to affirm the clearly erroneous decision of the Board. This court regularly upsets trial court judgments of Article III judges when they fail to consider correctly the pertinent facts. See, e.g., Read Corp. v. Portee, Inc., 970 F.2d 816, 829, 830 (Fed.Cir.1992) (reversing district court denial of JNOV because jury verdict of willful infringement unsupported by substantial evidence); Laitram Corp. v. Rexnord, Inc., 939 F.2d 1533, 1539, 19 USPQ2d 1367, 1372 (Fed.Cir.1991) (reversing district court because conclusory assertions of infringement insufficient); Eli Lilly & Co. v. Medtronic, Inc., 915 F.2d 670, 673-74, 16 USPQ2d 2020, 2023 (Fed.Cir.1990) (reversing district court because factual findings insufficient to establish violation of injunction). Why we should treat the Board of Patent Appeals and Interferences differently is a mystery to me.
Given the choice of crediting the unsupported conclusion of my colleagues or the unmistakable teaching of Guilford and the unchallenged factual statments of experts skilled in an esoteric art, I must part company with my colleagues.