In Re ARTHUR ANDERSEN & CO., Petitioner

COFFIN, Chief Judge. This is an unusual petition for mandamus seeking reversal of several interlocutory orders entered by the district court. The orders arise out of a lawsuit by two classes of purchasers of securities of a company (Viatron) against various officers and directors of the company, and its suppliers, auditors, underwriters and consultants because of the alleged furnishing of false information on which they relied to their detriment. Petitioner, Arthur & Andersen & Co., the auditor of Viatron, seeks mandamus review of orders denying motions to see and copy proofs of claim of alleged purchasers of Viatron debentures and common stock. Petitioner also seeks mandamus review of an order denying a motion for leave to file a cross-claim and third party complaint naming Arthur D. Little, Inc. (Little) as a defendant. *38 One of the plaintiff classes brought its lawsuit in 1971, the other in 1973, Andersen being a defendant in both cases. Little was made a defendant in 1975. In 1978 plaintiffs entered into a settlement with Via-tron’s underwriters. In later 1979 and early 1980 another settlement was negotiated between plaintiffs and Little and various of Viatron’s suppliers. On May 7, 1979, the district court issued an order to implement the settlement with the underwriters, fixing the procedures for testing the fairness and adequacy of the settlement and for determining distribution of the partial settlement fund. In connection with the latter, the court required that each class member desiring to participate file a proof of claim, postmarked no later than October 3, 1979, by sending it to the Clerk of the court at a post office box rented at the expense of plaintiffs’ counsel “for the purpose of receiving proofs of claim and other communications from class members with regard to the [notice of hearing on fairness and stipulation of settlement].” The post office box to be used was in the General Post Office in the city of New York. The notice sent to class members pursuant to the May 7 order concluded by saying, under the caption “Court File Available”, “All documents filed in the Actions may be inspected at the office of the Clerk of the United States District Court for the District of Massachusetts . .” Andersen moved, on November 21, 1979, for access to inspect and copy the proofs of claim on the ground that they were court records. This was denied without opinion on December 6,1979. A motion for rehearing was denied on January 11, 1980. On January 15, 1980, Andersen again sought access to the proofs of claim via a motion to reopen discovery (which the court on June 5, 1979, ordered completed by August 17, 1979) on the ground that proof of damages, presumably derivable from data in the proofs of claim concerning net proceeds of sale, is “an integral part of plaintiffs’ alleged causes of action”. This motion was denied on January 16, 1980. Meanwhile, Andersen had moved for leave to file a cross-claim and third party complaint against Little on January 11, 1980. 1 Pointing out, with a sense of inverse relevance, see n. 1, that if its motion were denied, it would be the only one to receive this disposition, Andersen urged that a separate suit against Little could be avoided and that Little could “in no way” be prejudiced. On January 18, 1980, the district court, rehearsing the history of four years of discovery leading to the settlements and leaving only essentially the litigation between plaintiffs and Andersen, denied the motion. Perhaps the most palpable impression we distill from our reading of the record in this case is that of what might be called a megadefense. The extent to which even in this small salient of a much wider front a stream of motions, memoranda, requests for clarification and reconsideration can be filed on multiple issues without surcease is truly impressive. To begin with, it ought not be, but it is, necessary to observe the severe standards which govern the issuance of mandamus. Andersen properly invokes Kerr v. United States District Court, 426 U.S. 394, 96 S.Ct. 2119, 48 L.Ed.2d 725 (1976), citing to some of the prerequisites of “extraordinary circumstances”: the lack of other means to attain relief; and establishing an indisputable right to the writ. Id. at 403, 96 S.Ct. at 2124. The Court uses even stronger language in equating “exceptional circumstances” to “a judicial ‘usurpation of power’ ”. Id. at 402, 96 S.Ct. at 2123. We have echoed these criteria and specifically have *39 noted their inapplicability “to obtain review of an otherwise unreviewable interlocutory order”. USM Corp. v. GKN Fasteners, Ltd., 574 F.2d 17, 23 (1st Cir. 1978). With these brooding admonitions hovering over us, we consider each of the three purported bases for mandamus. The first is the motion for access to the proofs of claim based simply on the grounds of their alleged status as court records. Andersen’s authorities are, to say the most, less than dispositive. International Business Machines Corp. v. Edelstein, 526 F.2d 37, 45 (2d Cir. 1975), says simply that “the trial record is composed of the proceedings in the District Court including all papers, exhibits, and affidavits on file with the court”, adding, “It is not the type of paper submitted but rather the fact of filing which determines whether a particular item will be included in the record.” Id. For this latter proposition, it cites Todd v. Nello L. Teer Co., 308 F.2d 397 (5th Cir. 1962), which involves a letter from an attorney to a trial judge attempting to retract a concession as to a point of law. The letter was marked “filed”, bore notations presumably by the judge indicating his denial of a motion enclosed with the letter, and was in fact placed in the court file. That such a letter was held part of the record can shed little light on our present inquiry. References to the treatises suggest that some closer nexus to the judge or clerk must be intended before a paper becomes part of a case record, at least for purposes of FRAP Rule 10. 2 In 9 Moore’s Federal Practice, ¶ 210.04, at 10-14, we read, “The term original papers refers, then, to all papers presented to the district court and filed in the record and to all papers filed by the district court itself.” In 16 Wright and Miller, Federal Practice and Procedure, § 3956, at 386, the definition, if anything, is narrower: “Under [Rule 10(a)], only those matters that were in fact presented to the district court are considered part of the record on appeal; matters that were merely lodged with the clerk or never offered in evidence are excluded from the definition.” Whether the proofs of claim, even if they are potential record material, were “filed with the court” in accordance with Rule 5(e), Fed.R.Civ.P., 3 presents another problem. In 4 Wright and Miller, id. at § 1153, at 599, the authors pointed out that some courts have held that filing was not accomplished even by leaving a notice under the door of the clerk’s office after hours, and that uncertainty is best avoided by personal delivery, whether at the clerk’s office or his home. See also 2 Moore, id., ¶ 5.11, at 1391-92. Moreover, the case authorities on depositions indicate that they do not become part of the record until they are presented to the court. Dictograph Products Co. v. Sonotone Corp., 231 F.2d 867 (2d Cir. 1956); Daiflon, Inc. v. Allied Chemical Corp., 534 F.2d 221 (10th Cir. 1976), cert. denied, 429 U.S. 886, 97 S.Ct. 239, 50 L.Ed.2d 168 (1976); Busse v. United States, 542 F.2d 421 (7th Cir. 1976). Andersen also seeks support from a statute designed, we think, more for archival guidance than for court operations, 28 U.S.C. § 457. This merely provides that court records are to be kept where the court sits. Were this, however, to be taken literally, it would cut against any claim that the proofs of claim ordered to be sent to New York were part of the records of a case in a Massachusetts district court. Or, if the statute be thought to require that such proof ought never have been ordered sent to New York in the first place, the absurdity of the barrier thus erected to efficient processing of claims is apparent. Our distillation from all this is that we are not sure that the proofs of *40 claim sent to the New York postal box are yet case “records”, or that, if they are, they are presently disclosable to Andersen. We doubt that Andersen can have such a present interest in the statements of plaintiffs who have chosen to settle with other defendants, such plaintiffs not by any means necessarily being all of the plaintiffs who are suing Andersen. Moreover, the fact that the court-approved notice to claimants ends with a reference to the “court file” being subject to inspection in Massachusetts suggests an intent that proofs of claim in the processing, winnowing, and correcting stage have not yet reached the status of case records filed in court. But whatever our doubts may be as to these matters we are crystal clear that whether or not Andersen has a right to these records can be determined by proper motions at later stages of this litigation and that the district court’s denial of this first motion cannot, with any composure, be termed a usurpation of power. Andersen's second basis for mandamus lies in the court’s denial of its motion to gain access to the proofs of claim, not because they were court records, but because the information in the hands of plaintiffs’ counsel was discoverable since it bore on verification of proof as to damages. The existence of a megadefense was underscored in the documents submitted to the court revealing that Andersen contemplated calling as witnesses “[rjesearch analysts of every broker, mutual fund or other institutional investor who has filed a proof of claim”, and “[ejvery individual who, and a financial officer of any corporation, partnership, association or other entity which, filed a proof of claim.” Andersen has proposed a minimum of some 65 pages of special questions to be put to the jury, many of them requiring specific findings as to each class member. Another proposed list of some 60 voir dire questions, to be asked every juror, included inquiries as to whether the jurors were related to or acquainted with any of the 2400 plaintiffs or their counsel. In support of its argument that the material sought in the discovery motion is relevant to its defense. Andersen pointed out in its motion that the data contained in the proofs of claim would be relevant to determining damages. In addition, in its argument, Andersen asserts that the combination of data on the state residency of claimants and the dates of purchase and sale would bear on the questions of whether each plaintiff purchased in justifiable reliance on alleged misrepresentations or sold with actual knowledge of misstatements or facts sufficient to put one on inquiry. The original ground for discovery assigned, relevance to damages, obscures the basic proposition that plaintiffs bear the burden of proof. When, whether, and to what extent use of data in the proofs of claims may be relevant, e. g., for impeachment purposes, will depend on whether the issue of damages is severed and made the subject of separate proceedings. See 3B Moore’s Federal Practice, ¶ 23.02, p. 23-65, and ¶ 23.46, p. 23-407. The reasons subsequently assigned as justifying immediate and mandatory discovery are no more persuasive. Residence of a plaintiff at the time of purchase, rather than at the time of filing proof of claim would seem to be the only relevant data bearing on the appropriate limitations period. Insofar as reliance is concerned, this may or may not be a matter for individual proof; in the latter event it most likely would be addressed in a severed proceeding. But in any case the mere amounts and dates of purchases and sales are not, at least facially, demonstrably relevant to whether or not a plaintiff used due diligence. The extent to which information from each class member may be used in attempting to establish plaintiffs’ case or may be relevant in defense will depend in large part upon decisions made by the court in its continuing and demanding task of managing this litigation. Cf. Blackie v. Barrack, 524 F.2d 891 (9th Cir. 1975). At this stage, the eve of trial between plaintiffs and Andersen, we see no abuse of discretion, certainly no usurpation of power, and no lack of remedy if Ander *41 sen were subsequently to feel deprived of information essential to its defense. We come, finally, to Andersen’s eleventh hour motion for leave to file a cross-claim and third party complaint against Little. The district court very cogently invoked In re Cessna Distributorship Antitrust Litigation, 532 F.2d 64 (8th Cir. 1976), which is comprehensively apposite. There the delay in filing a cross-claim was only three not four or five years. There, as here, the court faced a complex, hard earned settlement, the benefits of which would have been largely negated by allowing the cross-claim. There, as here, the movant was still able, if losing the instant litigation, to bring a separate action over for indemnity or contribution. In this case trial is now, after prolonged preliminaries, set for May 27, 1980. Were Little to be brought in at this late date, 4 not only would it be deprived of much of the near term peace for which it has bargained but trial would most probably be postponed for a substantial period to allow preparation. And the issues would perforce be extended to cover not only plaintiffs’ diligence and reliance, Andersen’s conduct and the reasonableness thereof, but also Little’s conduct and its reasonableness, which would also trigger an inquiry into the conduct of vendors and their reasonableness. Moreover, if Andersen were to prevail, any action over against Little would be mooted. We said at the outset that this was an unusual petition. We so bespoke ourselves in view of the conjunction of orders which are only interlocutory or subject to broad discretion, or both; the questionable or at least fragile posture of petitioner’s claim; and the fact that other avenues of adequate relief are available. Having in mind the austere standards governing the dispensation of mandamus, we cannot find any arguable justification for challenging a court under these circumstances. Plaintiff-respondents, however, have not sought double costs or counsel fees and we shall not venture where they have been reluctant to tread. The petition for writs of mandamus is denied. Footnotes [1] Some fourteen other cross-claims and third party complaints had, after hearing and/or submission, been granted by a magistrate on October 29, 1979. Despite a colloquy on May 4, 1979, in which the district court, commenting on Andersen’s fear of inconsistent results if suit in an action for indemnity were to be brought in another forum, raised the possibility of a cross-claim in the instant litigation, Andersen took no action for the next eight months. Subsequent to the denial of the motion for leave to file, Andersen initiated two independent actions against Little, one in New York and one in Massachusetts. [2] Rule 10(a) reads: “Composition of the Record on Appeal. The original papers and exhibits filed in the district court, the transcript of proceedings, if any, and a certified copy of the docket entries . shall constitute the record on appeal in all cases.” [3] Rule 5(e) reads: “Filing With the Court Defined. The filing of pleadings and other papers with the court as required by these rules shall be made by filing them with the clerk of the court, except that the judge may permit the papers to be filed with him . . . [4] Andersen cites as reason for its delay a secret agreement which it had with Little tolling the statute of limitation. While this agreement may have made the delay possible, however, it in no way constitutes a justification for the delay in filing the claim in the face of the district court’s repeated entreaties and efforts to conduct the litigation efficiently.