finding trade secrets to be propertySummary of this case from In re Zyprexa Injunction
No. 71, Docket 20960.
January 11, 1949.
Appeal from the United States District Court for the Southern District of New York.
Action by the Conmar Products Corporation against the Universal Slide Fastener Company, Inc., and others for infringement of Claims 43, 44, 45, 47, 48, 49, 52 of Patent No. 2,201,068 issued to George Wintritz on May 14, 1940, and for the infringement of Claims 1, 2, 4, 9, 10, 11, 17, 21, 22, and 23 of Patent No. 2,221,740, issued to Frederick Ulrich on November 12, 1940. The complaint also included a third cause of action based upon the defendants' inducement of the plaintiff's employees to divulge its trade secrets. From a judgment dismissing the complaint plaintiff appeals.
Davis, Hoxie Faithfull, James Franklin and William H. Davis, all of New York City (Maxwell James, John Hoxie and Harold James, all of New York City, of counsel), for appellant.
Pennie, Edmonds, Morton Barrows, Morris Kirschstein, and George E. Middleton, all of New York City, for Universal Slide Fastener (Serval, Inc.) and another.
Before L. HAND, Chief Judge, and AUGUSTUS N. HAND and CLARK, Circuit Judges.
The plaintiff appeals from a judgment so far as it dismissed its complaint for the infringement of Claims 43, 44, 45, 47, 48, 49 and 52 of Patent No. 2,201,068, issued to George Wintritz on May 14, 1940, for the infringement of Claims 1, 2, 4, 9, 10, 11, 17, 21, 22 and 23 of Patent No. 2,221,740, issued to Frederick Ulrich on November 12, 1940; and upon a third cause of action for inducing the plaintiff's employees to divulge its trade secrets. The art to which both the patents and the trade secrets belong was that of "slide fasteners," or "zippers." These date from the invention of a Swede, named Sundback, who in 1914 filed his first application in this country, which issued in Patent No. 1,219,881 on March 20, 1917, a patent which substantially dominated the art until it expired in 1934; and although improvements appeared from time to time before that date, nearly all those relevant to this action were filed about that time. The "zipper" is in such widespread use that we shall assume an acquaintance with it, and confine our statement to the processes which are the subject of this action. Two methods of making the metal parts — "elements" — and of fastening them to the tapes to make "stringers," had been in use before Wintritz filed his application in October, 1934. One of these was to make the "elements" separately and feed them from a hopper to the tape to which they were to be attached; the other was to "coin" or to "stamp" the "elements," feed them to the tape, and attach them — all in a single machine. Wintritz's patent was a variant; one machine "coined" or swaged out of a piece of wire the "elements" as "embryos" in a continuous strip; another machine fed the wire so "coined" to the tape where at one stroke a punch finished the fabrication of that "element" which was next to the end, and cut from the strip the end "element," which was at the same time clamped upon the tape. Ulrich was a later inventor; his disclosure was of a method and a new "strip"; it was an improvement upon Wintritz, the chief purpose of which was to eliminate waste metal. The cause of action for inducing the betrayal of the plaintiff's trade secrets we defer for the moment.
Claims 43, 44 and 45 of the Wintritz patent were for the machine which feeds the partially formed "elements" to the tape, completes, cuts and attaches them. They depend, not upon the details of the mechanism which does this, but upon the succession of steps which begins with "a connected series of embryo fastener elements" and ends with the "elements" clamped upon the tape. The "rolling machine," which is not a part of these claims, has already completed the "embryos," except for the metal which remains between the "jaws"; and the "elements" remain all in a continuous strip. The strip is intermittently advanced, pausing as the end "element" straddles the edge of the tape, which is set vertically in its path. At that moment a punch descends upon the strip, and with a single stroke cuts off the end "element" and punches out the metal left between the "jaws" of the "element" just behind it. The "jaws" of the end "element," astraddle of the edge of the tape, are then pressed together, and that "element" is clamped in position. By a separate mechanism the tape is then moved one step upward; and the "strip" is moved one step forward, the "jaws" of the "element," which have just been freed of its intervening metal, now straddling the tape. This completes the cycle.
The defendant's chief reliance to invalidate these claims is Patent No. 2,169,176 to Noel J. Poux, applied for on December 16, 1933, and issued on August 9, 1939. The claims of this patent were for a method of making a continuous strip of "elements," and for the strips themselves; and the chief difference between its strip and Wintritz's, when Wintritz's emerges from the "rolling machine," is that Wintritz has only struck up the outlines of the "embryo," leaving in place the metal between the "jaws," while Poux in one way or another has almost completely opened the "jaws." As a result, when the "element" next to the end is beneath the punch, all that remains to be cut away is a narrow piece of metal connecting the ends of the "jaws." Poux's punch at once cuts off the last "element," and cuts away this connecting piece in the following "element." Another mechanism clamps the "jaws" to the tape in substantially the same way as Wintritz.
Poux did not describe the machine in detail which was to advance the strip, drive the punch, and clamp the "elements" to the tape; as we have said, his disclosure was only of a method and of the strip. However, it was permissible for him to presuppose as part of his disclosure whatever was within the knowledge of the art at the time; and the examiner must have supposed that there existed accessible mechanical trains to make practical the disclosure as it stood; else he would not have passed it. It is true that Poux was not a part of the prior art when Wintritz filed his application; but he was a prior inventor, and the validity of Wintritz's invention depends upon the same considerations as though Poux had been in the prior art: i.e., whether the steps he took beyond Poux, — his variations upon Poux's theme, — were of themselves an invention. The only change was that in Poux's prefabricated strip the "embryos" had come a little nearer to maturity, so to say; part of the metal between the "jaws" had already been removed. The variant at once strikes an observer as being of a kind which any art throws up in its inevitable proliferation; and, although such impressions are often unreliable, and should always yield to any evidence drawn from the history of the art, here there is no such corrective, for Poux's disclosure did not become public until after Wintritz had filed his application. We should not be justified in ascribing any greater invention to Wintritz's advance beyond Poux than an untutored appraisal gives it.
The only basis for even a plausible argument is that the "jaws" in Poux must be distorted, because he first punches or cuts out part of the metal to form them, and later "coins" or swages the recesses. This stands only upon the opinion of the plaintiff's expert that this must occur because the recesses are so close to the "jaws"; for nobody professes to have observed anything of the sort; and even were the defect demonstrated, the advance would not be great enough to base a patent upon. However, it is an unwarranted limitation upon Poux's disclosure anyway, because in one form of his invention (Figures 30 and 31), he declared that the "jaws" might be formed after the recesses. The plaintiff answers that in this form the specifications disclose a slot made in advance of the recess, and later expanded to make the "jaws." The answer will not serve. The "coining" of the recess may distort the slot, but the plaintiff certainly did not prove that, after the punch, 116 of Figure No. 31, had fashioned the "jaws" out of the slot, any distortion produced in the slot by the formation of the recess would remain in the "jaws." The argument appears to us no more than a bit of expert's ingenuity; and we hold Claims 43, 44 and 45 invalid.
The plaintiff does not seek to support Claims 47 and 48 because they contain those particular elements, which on their face appear to distinguish them from the immediately preceding claims (47, by the single cam shaft and the "counter"; 48, by the lightness, smallness and shortness in stroke of the machine). The argument is that, since Poux did not disclose any machine for making his strip, Wintritz was the first inventor to disclose a complete machine which secured the advantages of the new method. As we have already said, the very grant of Poux's patent presupposes that the examiner thought it practicable as it read; and we are not confined to that presumption. Sundback's patent of 1920 — No. 1,331,884 — shows adequate mechanisms to feed and attach the "elements" to the tape, although these are disclosed in combination with the "coining" or swaging processes which operate upon the wire and form the "embryos." The plaintiff replies that the difficulty is just that: Sundback's machines did combine the two operations, and it is the separation of the two which is the kernel of Wintritz. That argument, even if Wintritz had been the first to divide Sundback's process into two, would be of doubtful validity; but, taken merely as a conception, Poux was the prior inventor, for Poux's disclosure was for a process in two separate steps. A competent workman in the field who wished to follow Poux would have found the means at hand in Sundback's machine. These claims are also invalid.
As to Claims 49 and 52, we have nothing to add to what we said in Metallizing Engineering Co. v. Kenyon Bearing Auto Parts Co., Inc., and we take this occasion to reaffirm the doctrine there laid down. In spite of the repeated admonitions of the Supreme Court that the denial of a petition for certiorari is not to be taken as indicating approval of the decision below, we cannot avoid attaching some significance to the denial in that case. We had very frankly overruled at least one of our former decisions, and we had deliberately refused to follow the Court of Customs and Patent Appeals; the petition was presented with the utmost earnestness by a most distinguished leader of the patent bar as an unhappy departure from established law, the result of our misunderstanding of the Supreme Court's own decisions. All things are possible; but the situation as a whole was so favorable to a consideration of the merits, if the court had been in doubt, that we feel bound to treat the issue as closed, and so we shall, until the court directs us otherwise.
2 Cir., 153 F.2d 516.
The Ulrich Patent
The claims in suit of this patent need not detain us long, for they are all plainly invalid. We are to judge them as though all the prior art were before Ulrich when he filed his application on December 11, 1937. Buchwald and Kretsch's Austrian patent had been published on October 10, 1935; the Legat patent issued on October 26, 1937. Each disclosed the "nesting" feature of Ulrich's specifications; and so far as that was an advance it was therefore already in the art. It is true that neither of these patents disclosed a continuous strip of "embryo" "elements," and we may assume for argument that to transform either of them in such a strip with appropriate feeding, punching, cutting and attaching mechanical train would have demanded invention. However, although neither Poux nor Wintritz was in the prior art in December, 1937, they were both prior inventors, and Ulrich had the burden of showing that, with the disclosures of Buchwald and Kretsch and of Legat before him it demanded invention to introduce their metal saving into the strips of the two. We cannot add to the force of this bare statement of the question. As in the case of Wintritz's own patent, only a history of long unsatisfied delay before Ulrich appeared, and of immediate and universal substitution when he did, would satisfy us that such a combination of existing elements justified a patent. All the claims of this patent are invalid, regardless of any question of disclaimer.
The Trade Secrets
As its third cause of action the plaintiff alleged that the defendants enticed away several of its employees and made use of information which they had acquired while working for it. The plaintiff recognized that this alone would not be enough, and therefore it asserted, as was the fact, that these employees, and particularly the most important among them, Voity, had executed a contract in which they promised not to divulge anything which they might learn of the plaintiff's methods; and that the defendants, knowing of this contract, or at least having notice of its existence, induced the men to divulge the secrets they had learned. It was Voity who devised the offending machine, which the plaintiff says embodied some seven of its secrets, of which six and a part of the seventh were to be found in the disclosures of the two patents in suit. The other part of the seventh was disclosed in two later patents issued to the plaintiff.
The upshot of the evidence taken on this cause of action was that in May, 1939, Voity was in the plaintiff's employ in a position that made him familiar with all the details of its "zipper" manufacture, under a contract by which he had promised "not to disclose any information." During the summer after Voity left the plaintiff and began to work for the defendants, they engaged him in making "zipper" "elements" to be attached by hopper machines — this being a new part of the defendants' business. The plaintiff became concerned when it learned that Voity had gone to the defendants, and in August, 1939, according to the testimony of Blume, its vice-president, he had an interview with Shaw, president of one of the defendants, in which he told Shaw of the secrecy contract. This Shaw denied, and although the judge made no express finding as to that particular talk, he did find that the defendants had no knowledge of the secrecy contract when they hired any of the plaintiff's employees. This is conclusive, because one of the plaintiff's former employees, Griffiths, the defendants did not hire until 1940. Thus the finding that Shaw had no knowledge of the secrecy contract when he took any of the plaintiff's employees, necessarily precludes the truth of Blume's version; and certainly the finding is not "clearly erroneous." Nor was there any evidence to put the defendants on notice that Voity or Griffiths or any of the others was subject to a secrecy contract. The plaintiff was a very active exploiter of patents; in this record alone nine were put in evidence; apparently it depended in large measure upon them for protection. The record contains no evidence that it was customary in this art to require such contracts of employees, or that there was an implied understanding or confidence that employees should not divulge what they learned in a factory; and here, as always, the plaintiff was charged with the burden of making out its case. What the judge said generally of the plaintiff's arguments upon the third cause of action is especially applicable: that they were "made up of inferences and conclusions necessary to piece out the testimony of witnesses, and counsel assumes many facts to be true of which there is no direct evidence and would require the court to draw inferences not justified on the whole record."
Some time in February, 1940, Voity made drawings for the offending machine, and an experimental one was produced in March or April of that year, and went into production in July. Fourteen were ordered in that month of which two were delivered and in use in October; and by the middle of November the defendants had either spent, or committed themselves to spend, $40,000 upon the venture. On November 16 the plaintiff's attorneys wrote a letter to the defendants, complaining of the employment, not of Voity, but of two of its former employees, Hirsch and Griffiths, and enclosing the important parts of the secrecy contract. In his answering letter Shaw said that "about three months ago" Blume had called him on the telephone and complained about Griffiths, and had told him of "an agreement" with Griffiths. Whether that agreement was the secrecy agreement is not clear — on the stand Shaw said that it was not — but in any event the talk was only about Griffiths. Even though Blume did mention the secrecy agreement, it told Shaw nothing as to Voity; and indeed it would probably have misled him; for to single out Griffiths would naturally lead Shaw to suppose that only Griffiths was subject to such a contract. The same thing is also true of the letter of November 16, for it said no more than that Hirsch and Griffiths had executed the contract. The plaintiff's silence is inexplicable, and confirms the judge's conclusion that Blume did not tell Shaw in 1939 that Voity had signed such a contract. Whether the notice in this letter that Hirsch and Griffiths had done so, was enough to put Shaw more on notice as to Voity than the talk had, is indeed doubtful; but arguendo we will assume that it was; and if so, then on November 16, 1940, the defendants should have inquired further and would have learned that Voity too had signed.
Courts have been accustomed to speak of trade secrets as "property," and at times to deal with them as if they were. That may be permissible and to some extent desirable, when the question is whether the wrongdoer has got access to them by some wrongful means, like breaking into a factory, or copying formulae or blue prints. When, however, the dispute turns, as it does here, upon whether the wrongdoer has acquired a secret from the employee who has himself acquired it lawfully, the wrong consists in inducing him to break his contract, or to be disloyal to a confidence reposed in him; and in either case it is a species of the tort — recognized now for over a century — of inducing an obligor to default upon an obligation. Since the specifications of the patents in suit disclosed the first six secrets and part of the seventh, that much of the secrets upon issue of the patents fell into the public demesne; and, prima facie, the defendants were free to use them. The Seventh Circuit, and apparently the Sixth as well, have, however, held that if before issue one has unlawfully obtained and used information which the specifications later disclose, he will not be free to continue to do so after issue; his wrong deprives him of the right which he would otherwise have had as a member of the public. We have twice refused to follow this doctrine; and we adhere to our decisions. Conceivably an employer might exact from his employees a contract not to disclose the information even after the patent issued. Of what possible value such a contract could be, we find it hard to conceive; but, if an employer did exact it, others would perhaps be obliged to turn to the specifications, if they would use the information. Be that as it may, we should not so construe any secrecy contract unless the intent were put in the most inescapable terms; and the plaintiff's contract had none such. In their absence we do not see why a wrongful inducement to divulge the disclosure before issue should deprive the wrongdoer of his right to avail himself of the patentee's dedication; for, as we have just said, the contract is to be construed as imposing secrecy only until issue. The doctrine must rest upon the theory that it is a proper penalty for the original wrong to deny the wrongdoer resort to the patent; and for that we can find no support in principle. Thus, any possible liability for exploiting whatever the patents in suit disclosed, ended with their issue. Since the earliest notice was on November 16, 1940, it is not necessary to resort to this reasoning as to the first six secrets and as to part of the seventh, because the two patents in suit issued before November 16, 1940; but the doctrine does become important as to the remaining part of the seventh secret.
Du Pont de Nemours Powder Co. v. Masland, 244 U.S. 100, 37 S.Ct. 575, 61 L.Ed. 1016; Robins Dry Dock Repair Co. v. Flint, 275 U.S. 303, 48 S.Ct. 134, 72 L.Ed. 290; Becher v. Contoure Laboratories, 279 U.S. 388, 49 S.Ct. 356, 73 L.Ed. 752.
Shellmar Products Co. v. Allen-Qualley Co., 87 F.2d 104.
A.O. Smith Corporation v. Petroleum Iron Works Co., 74 F.2d 934.
It is almost, if not quite, impossible to learn what was that part of the seventh secret which did remain undisclosed after the issue of the Ulrich patent in suit; but for argument we will assume that the plaintiff proved that some parts did so remain. Whatever these were, they were all disclosed in two later patents to Ulrich: No. 2,338,884, issued January 11, 1944, and No. 2,370,380, issued February 27, 1945; and in any event the right to an injunction against exploiting any secrets whatever had therefore expired before the judgment was entered in November, 1947. All that could remain was the right to an accounting for profits or to a claim for damages between November 16, 1940, and the date of issue of these patents. We think, however, that the defendants had an excuse for exploiting that secret over that period, if they did so. As we have said, by November 16, 1940, they had invested $40,000 in the offending machine; that is, they had either paid or committed themselves to pay that much. The Restatement of Torts makes it an excuse for continued exploitation of a secret that at the time when one, who has theretofore been innocently exploiting it, first learns that he has induced the breach of an obligation, he has substantially changed his position. The opinions which support this are not very satisfactory; so far as we have found, in all but one or two it is doubtful whether what the judges say is more than dictum. However, they do all point the same way, for they assume that the situation is proper for the application of the doctrine that a bona fide purchaser takes free from a trust. The act of inducing the breach is the wrong, and the inducer's ignorance is an excuse only because one is not ordinarily held liable for consequences which one could not have anticipated. Although it is proper to prevent any continued use of the secret after the inducer has learned of the breach, the remedies must not invade the inducer's immunity over the period while he was ignorant. They may invade it, if the inducer has changed his position on the faith of his ignorance. We agree with the Comment of the Restatement that each case must stand on its facts; the answer depends upon weighing the loss to the inducer against the benefit to the obligee. In the case at bar we have no hesitation in deciding that issue in favor of the defendants. On November 16, 1940, when the defendants were first charged with any duty to desist, they had become free to exploit the major part of the secrets — the first six and an undetermined part of the seventh. Their duty at most extended no further than to change their machines and their methods, so as not to continue to exploit the still protected vestiges of the seventh; and even these would become free when the two other patents issued. Certainly, to compel them to make that change was seriously to invade the immunity they had enjoyed to that time; it would have compelled a disruption of their business and a redesigning of their machines. Opposed to this was a benefit to the plaintiff, the importance of which no one could even guess; for it must be measured by the advantage of suppressing only the use of the yet undisclosed part of the seventh secret, while the defendants remained free to use all the rest. The plaintiff had the burden of showing that the balance was in its favor, and we should be wholly unwarranted in deducing that conclusion from the maze of verbiage which wraps the issue.
§ 758(b) Comment (e).
Morison v. Moat, 9 Hare 241; Edelsten v. Edelsten, 1 DeG. J. S. 185; Stewart v. Hook, 118 Ga. 445, 45 S.E. 369, 63 L.R.A. 255; Homer v. Crown Cork Seal Co., 155 Md. 66, 141 A. 425; Chadwick v. Covell, 151 Mass. 190, 23 N.E. 1068, 6 L.R.A. 839, 21 Am.St. Rep. 442; Vulcan Detinning Co. v. American Can Co., 72 N.J. Eq. 387, 67 A. 339, 12 L.R.A., N.S., 102, Id., 75 N.J. Eq. 542, 73 A. 603; Lamont, Corliss Co. v. Bonnie Blend Chocolate Corp., 135 Misc. 537, 238 N.Y.S. 78.
The plaintiff finally argues that the defendants were liable because Voity was their agent, authorized to make the machines, and that while doing so, he knew of the contract and so supplied the missing factor in the liability and charged them as his principals. This reasoning would confine the excuse of ignorance to cases in which the inducer did not take the obligor into his own employ, and those are by far the greater number of cases in which he is aware of the obligation. For it will be seldom that one will innocently induce another's employees to disclose secrets of their employer's business, and yet not engage the employees in his own employ. That alone should be enough to condemn the reasoning practically. However, it is also wrong in theory. The fact that Voity knew of his own contract is immaterial; that knowledge might well have charged the defendants had he induced the breach of some other employee's obligation; but that he did not do. He did not induce himself to default in his own obligation. The transaction in which they ignorantly induced his breach was one in which Voity was the opposite party and did not represent them in any way. Having acquired the secrets innocently, they were entitled to exploit them till they learned that they had induced the breach of the contract.