Servo Corporation of America v. General Electric Company

393 F.2d 551, 157 U.S.P.Q. (BNA) 470, 1968 U.S. App. LEXIS 7702

• Court: Court of Appeals for the Fourth Circuit
• Decided: March 14, 1968
• Docket: 11209_1
• Status: Published

Opinion

WINTER, Circuit Judge:

As more particularly set out in Servo Corporation of America v. General Electric Company, 337 F.2d 716 (4 Cir. 1964), we held U. S. Patents Nos. 2,880,309 and 2,903,575, of which Servo was the assignee, invalid and, hence, incapable of infringement by General Electric. The patents were for devices known as “hot box detectors,” useful in the railroad industry to detect an abnormal quantity of heat emanating from freight car journal boxes. At the same time, we concluded that General Electric had been unjustly enriched by its wilful copying of Servo’s device, purportedly constructed under U. S. Patent No. 2,880,309, in breach of a confidential relationship between Servo and Southern Railway Company, for which Servo had installed a device for testing and experimentation; and we remanded the ease for a determination of Servo’s damages in its claim for unjust enrichment. General Electric, which had abandoned its efforts to develop a workable hot box detector near the close of 1955, was shown by the evidence to have obtained Servo’s proprietary interest in its claimed invention on April 10, 1958, when its engineering personnel inspected Servo’s installation, took photographs, made notes, and obtained drawings and engineering data. We made no determination of the extent and scope of Servo’s proprietary interest or General Electric’s misappropriation. Reference to the opinion in that case is made for a fuller statement of the facts and our holding.

Later, in Servo Corporation of America v. General Electric Company, 342 F.2d 993 (4 Cir. 1965), we denied a petition for rehearing, but clarified the scope of our previous holding and our directions to the district court on remand, by stating:

“We think it consistent with the opinion heretofore filed that the district court should allow Servo damages only for General Electric’s appropriation of that portion of its work which was not publicly disclosed prior to the visit to Salisbury on April 10, 1958, and that these damages should be measured by General Electric’s cost of acquiring the same information by its own experimentation.” Id., p. 993. 1

Thus, as the parties in this appeal agree, we directed the district court to address itself to the three questions: (1) what portion of Servo’s work was not publicly disclosed prior to April 10, 1958, (2) what portion of the undisclosed work was appropriated by General Electric as a result of its April 10, 1958 inspection, and (3) what would it have cost General Electric to acquire the same information by its own experimentation ?

The district court, with the aid of a Special Master appointed to assist in the determination of damages, 2 who participated in the proceeding, determined that Servo had not publicly disclosed its device in its entirety, that the essential undisclosed feature which General Electric appropriated was the orientation angles, 3 and that damages of $50,000.00 “would be proper, fair and reasonable.” Judgement accordingly was entered and both parties have appealed. We reverse, and remand for further proceedings.

In the proceedings on remand in the district court, and in this Court, Servo insists that the combination of all of the elements of the hot box detector was undisclosed and, impliedly although far from explicitly, that General Electric did not theretofore know, and hence appropriated, all that Servo had developed. General Electric, on the other hand, sought to establish that all of the elements of the detector had been publicly disclosed in trade papers and industrial fairs prior to April 10, 1958, and, again impliedly but not explicitly, that General Electric had developed its competing product from such sources of information, rather than the April 10, 1958 inspection.

Servo sought to take the deposition of a patent attorney and other employees of International Telephone & Telegraph Company. I. T. & T. had expended great effort and considerable sums of money in an unsuccessful effort to develop a hot box detector of its own. The purpose of the deposition was to discover I. T. & T.’s cost of development, presumably as some measure of what General Electric’s costs might have been had it been established that General Electric appropriated some of Servo’s accomplishments. General Electric moved to quash the taking of these depositions, and served interrogatories upon Servo, seeking discovery of what disclosure Servo had made prior to April 10, 1958. Servo, in turn, served interrogatories upon General Electric to determine its costs in connection with its pre-April 10, 1958 attempts to develop a workable detector; and General Electric filed objections to these interrogatories.

The various .motions and objections came on for hearing on July 19, 1966. The district court ruled on none of them, but instead appointed the second Special Master to aid the court “ * * * in the determination of the damages which might be due Servo, in the event that the Court should be of the opinion that some portion of Servo’s work was not publicly disclosed prior to the visit on April 10, 1958 * * * by representatives of General Electric Company * * As part of the order which was entered, Servo was directed to file a bill of particulars setting forth that portion of Servo’s work which was not publicly disclosed prior to April 10, 1958, and a pretrial conference was directed to be held on October 18, 1966.

Servo filed its bill of particulars on the date fixed, containing a lengthy recital of claimed combinations.' General Electric filed a responsive pleading, in which it asserted that every significant aspect of the detector had been disclosed prior to April 10, 1958.

At the pretrial conference held on October 18, 1966, considerable discussion between counsel and the court occurred, dealing with the relevancy of the discovery data sought to be elicited from I. T. & T. and the power of the court to require it, and also the propriety of General Electric’s being required to divulge the costs which it had incurred in an attempt to develop a hot box detector prior to April 10, 1958. After hearing from counsel, the district judge summarily announced that he thought that the seven years that the litigation had been going on was long enough, that he had conferred during an earlier recess with the Special Master, and that they had collaborated and written an opinion and judgment, containing the essential provisions before described, which he read to counsel.

As an answer to each of the three questions that we directed be determined on remand, we find the district judge’s opinion and judgment unsatisfactory, substantively and proeedurally.

In remanding the case for a re-determination of the first two questions which, the appeal presents, it is appropriate for us to state the substantive legal principles which will govern their outcome. Servo contends that the trade secret which General Electric appropriated from it lay in the “working combination” of various elements which it had developed, and urges us to rule that in order for there to have been public disclosure of the secret, there must have been a disclosure of all the components of the combination within the four corners of an integrated document. That Servo’s trade secret may have been a combination of components, each of which was in the public domain, cannot be doubted. Imperial Chem. Indus. Ltd. v. National Distillers & Chem. Corp., 342 F.2d 737 (2 Cir. 1965). But it does not follow that public disclosure of the working combination may be accomplished only through a single integrated document. That doctrine is the defense of anticipation in the field of patent law. Dewey & Almy Chem. Co. v. Mimex Co., 124 F.2d 986 (2 Cir. 1942). It has no necessary application to a trade secrets case where the question is whether, taking into account all of the plaintiff’s relevant disclosures, it is reasonable to conclude that a competitor could have ascertained the working combination from an examination of those disclosures. Sarkes Tarzian, Inc. v. Audio Devices, Inc., 166 F.Supp. 250 (S.D.Cal.1958); Cummings, Some Aspects of Trade Secrets and Their Protection: The Public Domain and the “Unified Description” Requirement, 54 Ken.L.J. 190, 197-201 (1966).

In holding that Servo’s trade secret might consist of several discrete elements, any one of which could have been discovered by study of. material available to the public, we recognize that in our earlier opinion denying a rehearing in this case we said that “the district court should allow Servo damages only for General Electric’s appropriation of that portion of its work which was not publicly disclosed prior to the visit to Salisbury on April 10,1958.” This statement was made in response to General Electric’s argument that the relationship between Servo and Southern could be deenled to be confidential only to the ex tent that the information communicated to Southern by Servo had not already been placed in the public domain, because, as we had said in American Potato Dryers v. Peters, 184 F.2d 165, 172 (4 Cir. 1950), "* * * it would be absurd to designate as a confidential disclosure that which is publicly known.” However, we also stated that we felt that this position was “consistent with [our earlier] opinion,” in which we had found a confidential relationship to exist between Servo and Southern. We interpret the language in the per curiam opinion to mean that Servo cannot recover for appropriation of information which had been so completely disclosed to the public as to dispel the existence of a trade secret and thus to negate the confidential relationship which had been established by Servo and Southern between themselves. We do not believe that in those instances where the extent of disclosure is arguable, and where the information had not been clearly placed in the public domain, General Electric should be allowed to avoid the consequences of the breach of confidence by piecing together in retrospect bits of information which had been disclosed in a variety of places and which as a combination were not clearly a matter of public knowledge.

Unless upon remand the district court should find that Servo’s public disclosures were so broad as to render ineffective its confidential relationship with Southern, the extent to which General Electric may invoke Servo’s prior publications as a defense to its misappropriation directly depends upon the extent to which it relied on them and not on its inspection of April 10, 1958. The gravamen in a trade secrets case is a breach of confidence, rather than an infringement of a property right; hence, reliance on innocent sources of information involving no breach of duty, is an essential element of the defense that the secrets were previously disclosed. Franke v. Wiltschek, 209 F.2d 493 (2 Cir. 1953); Standard Brands, Inc. v. Zumpe, 264 F.Supp. 254 (E.D.La.1967), Nims, Unfair Competition and Trade Marks, § 142 (1947 Ed.). Recently, the Supreme Court of Pennsylvania held that the first determination which must be made in a trade secret case is “whether, in fact, there was a trade secret to be misappropriated,” and that a public disclosure dispels the secret, irrespective of reliance. Van Prod. Co. v. General Welding & Fabricating Co., 419 Pa. 248, 213 A.2d 769 (1965). 4 We agree with this statement to the extent that complete public disclosure of a secret would destroy the legal effectiveness of, or prevent the creation of, any confidential relationship based upon that secret, the breach of which is the essence of a trade secret's action. But if Van Products holds that the mere presence in the public domain of the information upon which a trade secret is based precludes recovery for breach of a confidential relationship, we decline to follow it. Thus, should it appear that Servo made disclosures prior to April 10, 1958, unless these disclosures made its secret so obvious as to render meaningless the confidential relationship which had been established by Servo and Southern, they become irrelevant unless General Electric can show that it relied on them and did not appropriate the secret in its April 10, 1958 inspection. Only if proof is offered to show that General Electric obtained its knowledge from them and not the April 10, 1958 inspection, need the question of their sufficiency, singly or collectively, to disclose the working combination be considered. In that determination, we hold, further, that the burden of proving innocent reliance on non-confidential sources rests on General Electric.

Of course, with respect to the amount of damages, the matter of a disclosure of something less than a disclosure so obvious as to render meaningless the claimed confidential relationship may be significant. On this record, we do not now know what secret belonging to Servo was misappropriated by General Electric. From the arguments which have been advanced, the district court may well conclude that the secret was a combination of several components and, further, the district court may well find that one or more of these components had been publicly disclosed by Servo, but that all were not disclosed such as to destroy the confidential relationship. In the latter event, we think that the damages which Servo may recover would be the cost of experimentation to discover the component or components not disclosed and to discover how to combine all components, but not the cost of experimentation to discover the component or components which were disclosed, notwithstanding that General Electric did not rely on the disclosures but, rather, derived its information from the April 10, 1958 inspection. As to elements disclosed by Servo, but unknown to General Electric, General Electric should be liable only for the cost of discovering Servo’s disclosure, but not the cost of experimentation to develop that which was disclosed.

In applying these substantive principles, the district court on remand should comply strictly with Rule 52, because we are unable to ascertain, with certainty, how the district court arrived at its conclusions. We directed, first, a determination of what portion of Servo’s work was not publicly disclosed prior to April 10, 1958. Servo contended that none was disclosed in the bill of particulars it was directed to file; General Electric contended all were disclosed, although it concedes on appeal that the district judge correctly determined that the aiming of the angle of the detector was not disclosed. 5 While the first Special Master found that General Electric had copied three elements of the Servo device, we cannot read his report as making any finding of what Servo had' disclosed, if anything, prior to April 10, 1958. The second Special Master was not permitted to make any report.

Absent a report from either master, the conclusion of the district judge that “the Servo detector in its entirety had not been publicly disclosed,” whatever was meant manifestly lacks the specificity required by Rule 52, Fed.R. Civ.P. Without prejudice to the right of either party to conduct further discovery and to supplement the overall record with additional evidence, the district judge on remand will make detailed and intelligible findings if it is necessary to reach this question in the light of the principles we have stated.

In regard to the question of General Electric’s appropriation, the first Special Master found that General Electric, as a result of its April 10, 1958 inspection, had copied the orientation angles, the wheel trip and the shutter. His finding was adopted by the district judge in respect to the orientation angles. We infer that the district judge rejected his finding in regard to the wheel trip and the shutter, but if this inference is a correct one, we do not know why the latter finding was rejected. Again, we think that the district judge failed to comply with Rule 52 and, again without prejudice to the right of either party to conduct further discovery and to supplement the overall record with additional evidence in regard to this second question which should have been determined on remand, we direct the district judge to make detailed findings, initially or by reference, in regard thereto.

To turn to the question of damages, it appears that evidence was presented to the district judge on behalf of General Electric that its cost of development of that which it copied was $6,000.00, and evidence on behalf of Servo that its costs were many times that figure. This factual dispute rendered the case a peculiarly inappropriate one for the summary treatment which it received. The parties should have been afforded the opportunity to present relevant evidence in open court, or before a Special Master, as to damages; and a determination of damages, in compliance with Rule 52, made. In this connection, evidence of the development costs of I. T. & T. may be relevant, and appropriate discovery, limited by relevancy to the determination made of the second issue which should have been resolved on remand, permitted. Admittedly, the assessment of damages under the rule which we stated on remand cannot be done with complete exactness; but should it appear that I. T. & T.’s costs of developing that which may be determined to have been copied by General Electric can be identified and segregated, such data might be a persuasive indicator of the weight to be afforded other evidence of costs. What we say about permissible discovery dependent upon the determination of the second issue, however, is not intended by us to foreclose resistance to discovery by I. T. & T., if and when sought, on the grounds of unreasonableness or oppressiveness, or to condition compliance on the payment of costs, Rule 45(b), or to indicate how the district judge should decide such questions if they arise.

Because of the successive and dependent nature of the several questions which should have been determined on remand, we suggest to the district judge that, in an effort to conserve judicial time and time and costs to the parties, the several questions should be heard and determined, seriatim. Thus, discovery and evidence relevant to the second question may await, or may precede, the determination of the first question, depending upon further development of the issues by discovery and pretrial conferences; and discovery and evidence relevant to the third question, if reached, should await determination of the second question. This is not to intimate that we intend to hear interlocutory appeals from the determinations of the several questions to be decided, but simply to suggest that in this, as well as other protracted litigation, the separate determination of multiple issues can reduce the litigation to manageable size and bring about greater judicial efficiency.

We reverse the judgment below and remand the casé for further proceedings in accordance with the views expressed herein. We direct that each party shall pay its own costs on appeal.

Reversed and remanded.

1 . Certiorari was denied in Servo Corporation of America v. General Electric Co., 383 U.S. 934, 86 S.Ct. 1061, 15 L.Ed.2d 851 (1966), rehearing denied, 384 U.S. 914, 88 S.Ct. 1333, 16 L.Ed.2d 366 (1966).

2 . This was the second Special Master who had served in the case. The first Special Master had acted until the first appeal to this Court.

3 . The first Special Master in his consideration of the validity of the patents had found the essential elements of the inventions to be orientation angles, the wheel trip and the shutter. Apparently it was from this source that the district judge made his findings, although he made no mention of the latter two elements.

4 . In so ruling, the Pennsylvania court rejected the contrary prediction of Pennsylvania law in Smith v. Dravo Corp., 203 F.2d 369 (7 Cir. 1953).

5 . Whether this is a permissible reading of the opinion of the district judge that “the Court thus finds that the Servo detector in its entirety had not been publicly disclosed” is highly questionable. We are inclined to read it differently from General Electric.