I am in agreement with the ultimate conclusion of Judge Smith’s opinion, that the motion to dissolve should have been granted because Oliver cannot make the counts, but I arrived at that conclusion by somewhat different reasoning.
The main problem in this case, as I see it, is what the counts mean; and a secondary problem is how we go about determining what they mean. The key expression is “reference surfaces.” The gist of my position is that I consider that expression to be ambiguous and therefore I follow the rule, too well settled to require citation of authority, that under that circumstance we look to the application in which the counts originated in order to determine the meaning of the ambiguous expression.
This question of ambiguity of interference counts has always been a troublesome one. Semantics is involved in its broadest sense: the science dealing with the meanings of words as symbols and with human behavior in reaction to those symbols, including unconscious attitudes and linquistic assumptions. What has happened and what has been said in this interference amply exemplifies these attitudes and assumptions.
The examiner, when confronted with the motion to dissolve and the argument that Oliver does not disclose the invention which the counts define, reacted by saying:
The wording “reference surfaces” means just that; the wording is not ambiguous and it will be given its broadest interpretation in relation to its import.
I find that statement to be a reaction to symbols full of undisclosed linguistic assumptions. Means just what? What is its “import”? Apparently the examiner had his own ideas on the matter but our difficulty is that he nowhere disclosed them other than in his conclusion that “it would be natural and logical to identify the surfaces 28, of Oliver, as ‘reference surfaces’.” He did this despite the fact that Oliver did not so designate them, nor did Oliver use any other terms of similar import.
The board, in agreeing with the examiner’s dismissal of the motion, pointed out numerous limitations which are not included in the counts, some of which appear in other McCutchen claims, but this is little help as we are not concerned with missing limitations or with scope in that sense but with the meaning of the limitation which is there, the limitation to “reference surfaces.” In its initial opinion the board says nothing about what its idea of the meaning of “reference surfaces” is. After a request for reconsideration extensively arguing the question of meaning, the board made this somewhat baffling statement:
The counts do not require reference to the history of the prosecution of the McCutchen et al. patent application file to interpret them. Their meaning is ascertainable without ambiguity from reading the McCutchen et al. specification and claims.
Did the board find it necessary to refer to McCutchen’s specification (including the other claims) to find out what “reference surfaces” means or, like the examiner, did it deem that limitation to have a clear meaning standing by itself? Did it follow the rule on count construction?
I have found nothing in the record or briefs to show that standing by itself “reference surfaces” has any definite meaning, or any accepted technical meaning in this art, or that any of the prior art of record uses the term.
I agree generally with Judge Smith’s approach when he says:
* * * the counts of the present interference are so worded that reference to the specifications of the respective parties becomes necessary to ascertain the meaning of certain language therein. One cannot arrive at an understanding of the present counts as they relate to the subject matter of the interference without something more than the counts themselves. [My emphasis.]
*618I read into that statement the thought, necessary to comply with the time-honored rule about interference count interpretation, that the counts are in fact and in law ambiguous, as he suggests in other passages referring to latent ambiguity. The reason I have said above that the problem of ambiguity is a difficult one is that the cases are as full of disagreement on the question of ambiguity as they are on the issue of meaning. It is my view that with a term such as “reference surfaces,” which has no definite dictionary definition or accepted meaning in the relevant art, although it may at first glance appear to be a clear expression merely of great breadth, it is always proper to look to the usage of the term by the parties, as they read it on their respective disclosures, in order to determine, first, whether or not it is ambiguous. If, on doing this, we find, as I find here, that one party has adopted the term to mean one thing and the other party is using it to mean something else, then, perforce, the term is ambiguous. Once that conclusion has been reached, then we must follow the rule and look to the specification in which it originated and take the meaning there ascribed to it. Following this procedure, what do we find?
From a reading of McCutchen’s application, with or without the prosecution thereof, the conclusion is inescapable that he uses the term “reference surfaces” — and with the frequency Judge Smith reports — -in the sense of mounting surfaces which form an accessible part of the head assembly and abut against other surfaces in the device where the head is employed with the effect that the reference surfaces per se align the head, without more. Thus by “reference surfaces” he means aligning surfaces in a functional sense, not merely surfaces from which the location of other surfaces may be measured or computed by “reference” thereto.
From a reading of Oliver’s application we find that he would give an entirely different meaning to the term. He has a surface, as does McCutchen, lying in the plane of the head gaps and forming the face of half of his head before it is assembled; but when he gets his head assembled that surface vanishes for all practical or functional purposes, buried in the middle of the head and of no use whatever in mounting the head. It amounts to a mathematical abstraction. He, like the examiner, takes the position it is logical to denote this surface as a “reference surface,” presumably for the reason that a back surface 29 of the head, some distance removed from the surface we are speaking of, is carefully located by measurement so as to be parallel to it, the back surface being the one that is used as a “reference surface” to locate the head in its mounting.
At this point I would comment on the supposed common “invention” the parties are said in the principal dissenting opinion to have made. Preliminarily, however, let us consider the supposedly unambiguous word “surface” by itself, which also presumably “means just that.” What is its “import”?
It has been made clear that the transducer head is made of two halves, like a peanut butter sandwich. Suppose we spread two slices of bread with peanut butter and then put the slices together. The butter was spread on two surfaces. How many surfaces has the sandwich, 2, 3, or 4? Suppose another laminated structure, a 4-ply auto tire, the plies being vulcanized together. Mounted on the wheel, how many surfaces has it, 1, 2, 5, or 8? Or take a piece of plywood with five laminated sheets of veneer. Has-it 2, 6, or 10 surfaces? What did McCutchen mean by “surfaces” and what does Oliver mean, with reference to the finished transducer head defined by the language of the counts? It seems to me that when we are discussing the finished head, assembled from its two halves, we should not be contemplating surfaces which no longer exist, which existed only at one stage in the manufacture,' were fugitive and have been obscured, made inaccessible and hence can no longer be used for anything, like the original surfaces of the individual plies *619in the tire or the plywood which cannot be used for any purpose.
Now the principal dissenting opinion takes the position, which is to me untenable, that
The “gist” of the invention defined by the present counts, as I see it, is simply the provision of surfaces at either end of the recording head which are accurately ground, or lapped, coplanar with the pole tips of one of the pole pieces of the head. That invention is common to both parties. The present counts define that invention clearly.
There are various things wrong with this view. In the first place, the counts refer to the finished head ready for use, not to some phase of its manufacture in which it was in a different condition, as when it was in two unassembled halves. The counts define the sandwich, not the ingredients. Therefore the “head” of the count, which has the “surfaces at either end,” is the whole head, ready to be mounted, not merely the pole-piece portion thereof, as the dissent would have it in order to find the surfaces “at either end * * The fact is Oliver’s head has no “surfaces at either end of the recording head,” “coplanar with the pole tips” when his head is finished. He had coplanar surfaces at either end of the row of pole pieces before assembly, but they do not appear in the finished head defined in the counts. In McCutehen’s head it is otherwise. For the invention we must look to the counts, as the dissent emphasizes.
In the second place, we presume, in an interference, that the invention of the counts is a patentable invention, patentable to both parties, and patentability is not open to question. Patentability requires utility. It would therefore seem that the surface, the provision of which is said to be the gist of the invention, must be a surface serving some useful purpose in the finished head of the count. In McCutchen’s head it does just that. It mates with a positioning surface in the recording/reproducing equipment and locates the gap or line of gaps where it is wanted. The reference surface the dissent finds in Oliver does nothing of the sort. It does nothing. It has become useless. If that be the gist of the invention, I find it difficult to reconcile the view that this now useless surface is a patentable invention with the dissenter’s insistence in other notable cases that patentable inventions must have a practical, presently existing utility. See the dissenting opinions in In re Nelson, 280 F.2d 172, 47 CCPA 1081, and in In re Manson, 333 F.2d 234, 52 CCPA 739. One could go further and say that to assume a construction for a patent claim which causes it to read on a structure lacking utility is out of tune with the Supreme Court’s decision in Brenner v. Manson, 383 U.S. 519, 86 S.Ct. 1033, 16 L.Ed.2d 69 (1966).
■ Before leaving the principal dissent, I would point out one other flaw I find in its reasoning. The opinion says:
It seems to me that the term “reference surfaces” is not ambiguous and that it is plainly applicable to the Oliver structure for the reasons given by the Patent Office. The term “reference surfaces” refers to the surfaces lapped coplanar with the pole tips, whether that term be applied to the McCutchen or Oliver construction. The question of how far that reference surface extends, i. e. whether it extends far enough to be exposed, is not placed in issue by the broad language of the counts.
Like the reasoning of the examiner and the board, it confounds the issue of the meaning of “reference surfaces” with the question of the absence of other limitations from the counts. The conclusion as to what “reference surfaces” refers to is no more than a judicial fiat. The conclusion that the term is not ambiguous is mere assumption not based on an investigation into the facts about its several possible inconsistent meanings.
Ambiguity or the lack of it is not something about which one can make a decision merely by looking at the words, unless we are dealing with terminology which permits only of one definite mean*620ing. “Reference surfaces” is not such a term.
Judge Almond’s dissent, like the board’s opinion on rehearing, has the fault of looking to MeCutchen’s application to determine the meaning of the counts while at the same time apparently asserting lack of ambiguity (by joining the principal dissenting opinion which so asserts), an impermissible procedure. I see no more justification, absent ambiguity, for comparing the counts with other McCutchen claims than for studying his specification. That amounts to interpreting the counts as though they were claims in an infringement suit. If that is done, it is clear to me that appellants are right in saying the claims which became the counts would never be held to be infringed by Oliver’s structure by reason of the circumstances of their allowance as defining accessible surfaces. That being so, there is no reason for Oliver to be in interference with McCutchen on these claims for if what Oliver discloses is not an infringement of the counts, they do not define a common invention.
To summarize my reasons for voting to reverse: in compliance with the rules in interferences as to count construction I first determine whether the count is ambiguous by looking for its possible meanings and the meanings actually ascribed to it in this case and doing so I find ambiguity; I therefore construe the counts in accordance with the meaning they have in the application where they originated and discover that a “reference surface” is one which is available in the finished transducer head to facilitate its mounting orientation in use; I then determine from reading Oliver’s disclosure that he does not have a reference surface, according to that meaning, which complies with the other limitations of the counts since the reference surface he has, which is his back surface 29 (as well as an end reference surface), is not coplanar with the pole tip(s). I conclude, therefore, the Patent Office was in error in its construction of the counts and in failing to grant the motion to dissolve. I do so only on the basis of the above reasoning.
My conclusions make it unnecessary to consider the issue of priority as I consider that there is no such issue.