finding that the Supreme Court's opinion in Seattle Times Co. v. Rhinehart, 467 U.S. 20, 104 S.Ct. 2199, 81 L.Ed.2d 17, demonstrates that the Court has considered admission of evidence to be crucial factor in regard to public's right of access to judicial recordsSummary of this case from United States v. Beckham
Nos. 82-1820, 82-1821 and 83-1744.
Argued February 10, 1984.
Decided September 20, 1985.
Anthony C. Epstein, Washington, D.C., with whom Bruce G. Joseph, Jack C. Landau and Judy D. Lynch, Washington, D.C., were on brief, for appellants.
Loren Kieve, Washington, D.C., with whom George A. Birrell, New York City, and Judah Best, Washington, D.C., were on brief, for appellee Mobil Corp.
Appeal from the United States District Court for the District of Columbia (Civil Action Nos. 80-3032 80-2387).
Before WRIGHT and SCALIA, Circuit Judges, and MacKINNON, Senior Circuit Judge.
Opinion for the Court filed by Circuit Judge SCALIA.
Opinion concurring in part and dissenting in part filed by Circuit Judge J. SKELLY WRIGHT.
The Reporters Committee for Freedom of the Press and four individual reporters appeal from two District Court orders delaying until after trial and entry of judgment the public's access to court records consisting of documents produced by, and depositions furnished by the officers of, Mobil Oil Corp. in the course of third-party discovery in a civil suit, and used in connection with summary judgment and trial proceedings. The case presents preliminary issues of mootness and finality, and the merits issue of whether there is a First Amendment right of public access to court records pertaining to private civil actions prior to judgment.
The matters contested here arose in the course of litigation to which neither appellants nor appellee was a party — a libel suit brought by William Tavoulareas, the President of Mobil Oil Corp., and his son Peter, against the Washington Post Co. and a number of individuals connected with the Post, and a slander suit by the same plaintiffs against Philip Piro. In the course of those suits, which were consolidated, the Post sought to discover large numbers of documents from Mobil, and to take the depositions of Mobil-related witnesses. Before conducting a document search, Mobil requested a protective order to prevent public dissemination of the information to be obtained, on the ground that much of it was sensitive and confidential. It supported this request with an affidavit from Mobil Vice-President Walter E. MacDonald describing in general terms the negative effect release of the materials as a whole would have on Mobil's business in Saudi Arabia and its competitive position in shipping. The affidavit also stated that it would be impractical for Mobil to go through documents one by one during the discovery process to determine whether they contained any confidential information.
The District Court entered a protective order on November 5, 1981. It held that the MacDonald affidavit adequately identified the harm which disclosure of the information would cause Mobil, and that it would be undesirable to have Mobil specify, and the court rule on, objections to disclosure of particular documents, since that would slow discovery enormously and involve the court excessively in the discovery process. Instead, it adopted the following procedure:
Mobil would simply note the confidentiality of certain documents produced, based upon a relatively cursory review of the contents or source of the documents, and would then turn them over to The Post for use in this litigation. In the event The Post then wished to challenge Mobil's designation, W[ashington] P[ost] C[o.] counsel could seek an accommodation with Mobil or, as a last alternative, a court ruling that the material is neither confidential nor otherwise privileged.
Tavoulareas v. Piro, 93 F.R.D. 24, 29 (D.D.C. 1981) (footnotes omitted). If the Post did make such a challenge, "Mobil . . . would then bear the burden of establishing 'good cause' for the document's continued protection." Id. at 29-30 n. 3. The court also subjected to the protective order depositions in which information or documents designated confidential might be disclosed. Id. at 34.
On December 16, 1981, Mobil was granted leave to intervene to protect its interest in confidentiality. Discovery proceeded, in the course of which Mobil designated approximately 3,800 pages of deposition and an unspecified number of documents as confidential. Plaintiffs and defendants made cross-motions for summary judgment, in connection with which they filed under seal substantial portions of the materials designated confidential. The District Court denied those motions on June 30, 1982, a week before the trial was to begin. Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387 (D.D.C. June 30, 1982) (Order).
On July 2, 1982, Mobil moved to extend the November 5 protective order to the trial itself. In an order filed on July 8, without requiring any supporting materials in addition to those Mobil had furnished to obtain the original protective order, the District Court granted Mobil's request. It ruled that documents or depositions that Mobil continued to classify as confidential
will continue to be subject to the terms of the protective orders previously entered by the Court, but may be used by the parties for any proper trial purpose. They may be used to question or impeach witnesses and may be read or otherwise introduced into evidence. The document itself, however, will continue to remain subject to the protective orders of this Court.
Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387, slip op. at 2 (D.D.C. July 8, 1982) (Memorandum-Order). As we understand this order, those portions of each protected document or deposition used to question or impeach witnesses or otherwise read into evidence were made available to the public as part of the transcript, but the rest, although part of the exhibit sent to the jury, was not. On the basis of this order, Mobil designated 52 trial exhibits as confidential.
During this entire period, the Post had not exercised its right to challenge Mobil's confidentiality designations. On July 12, 1982, however, appellants here, four individual newspaper reporters and The Reporters Committee for Freedom of the Press, an unincorporated association ("the reporters"), filed a motion to intervene to ask the District Court to reconsider its November 5 and July 8 orders. On July 20, 1982, the District Court granted the motion to intervene but denied the motion to reconsider the orders because continued sealing was "essential to the efficient and expeditious conduct of the trial," Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387, slip op. at 4 (D.D.C. July 20, 1982) (Order). It did, however, agree to review the need for continuation of its protective order after the close of trial, and stated that it would require Mobil to identify the documents it considered confidential and explain the basis of its secrecy claim within thirty days after the end of trial. Id.
In fact, it did not do so until September 14, 1982, well after completion of the trial. At that point it moved to unseal all documents filed with the court. The court ultimately granted that motion at the same time that it granted the reporters' request to unseal the last of the trial and summary judgment exhibits. Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387 (D.D.C. June 21, 1983) (Order). Mobil challenged in a separate appeal the District Court's unsealing of deposition exhibits not used at trial. A panel of this court reversed the District Court and ordered the seal reinstated. Tavoulareas v. Washington Post Co., 724 F.2d 1010 (D.C. Cir. 1984). The court en banc vacated the panel opinion and remanded the case to the District Court for reconsideration in light of Seattle Times Co. v. Rhinehart, 467 U.S. 20, 104 S.Ct. 2199, 81 L.Ed.2d 17 (1984), and FED.R. CIV.P. 26(c). Tavoulareas v. Washington Post Co., 737 F.2d 1170 (D.C. Cir. 1984).
The reporters sought review of that order in separate appeals (consolidated here) pertaining to the two consolidated cases below. In accompanying motions for expedited appeal and mandamus, they urged us to take emergency action on the ground that, as to any documents used at trial or which were part of the summary judgment record, maintenance of the protective order during trial without a prior hearing on the need for that order as to each individual document denied their First Amendment right of immediate access to judicial records. We denied both motions. Tavoulareas v. Piro, Nos. 82-1820 82-1821 (D.C. Cir. July 30, 1982) (Order); In re The Reporters Committee for Freedom of the Press, Nos. 82-1819, 82-1820 82-1821 (D.C. Cir. July 26, 1982) (Order). The appeals were subsequently stayed with the agreement of the parties, pending the District Court's post-trial reexamination of the need for continued sealing of the documents.
By the summary judgment record, the reporters mean all documents on file with the court at the time of its summary judgment ruling which could have been considered by the court under FED.R.CIV.P. 56(c). Brief for Appellants at 31 n. 21.
The jury trial ended July 30, 1982, with a verdict for William Tavoulareas against the Washington Post defendants, for the Washington Post defendants against Peter Tavoulareas, and for William and Peter Tavoulareas against Philip Piro. As the District Court had directed, thirty days later, on August 30, Mobil filed a memorandum identifying those documents whose confidentiality it still wished preserved and explaining the basis for its request. These included only 18 of the 52 trial exhibits originally designated as confidential and an unidentified number of documents that were part of the summary judgment record.
On May 2, 1983, the District Court granted judgment non obstante veredicto ("judgment n.o.v.") for the Washington Post defendants. Tavoulareas v. Washington Post Co., 567 F. Supp. 651 (D.D.C. 1983). On May 18, 1983, the court released 14 of the 18 trial exhibits, because they either had never been subject to the protective order or were already publicly available. Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387, slip op. at 2 (D.D.C. May 18, 1983) (Order). As to the remaining four trial exhibits and those materials that were part of the summary judgment record, the court was not satisfied with Mobil's showing, but gave Mobil additional opportunity to make a more specific demonstration of need. Id. In response, Mobil waived its claim as to all but one trial exhibit, and as to all but two summary judgment documents. On May 20, 1983, the court completed its rulings on post-trial motions going to the merits of the consolidated cases by granting a judgment n.o.v. to Piro. Tavoulareas v. Piro, Civil Nos. 80-2387 80-3032 (D.D.C. May 20, 1983) (Memorandum-Order). On June 21, it ruled on the last three documents for which Mobil continued to advance a claim of confidentiality. The trial exhibit proved to be among those the District Court had already ordered unsealed because it had never been subject to the protective order. The court examined the two summary judgment documents in camera, concluded that Mobil had not substantiated its claim of irreparable injury, and therefore ordered them unsealed as well. Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387, slip op. at 4 (D.D.C. June 21, 1983) (Memorandum). Mobil did not appeal from that decision, but the reporters did, in appeal No. 83-1744. On September 8, 1983, this appeal was consolidated with their appeals from the District Court order of July 20, 1982, which were unstayed on the same date. Tavoulareas v. Piro, Nos. 82-1820, 82-1821 83-1744 (D.C. Cir. Sept. 8, 1983) (Order).
This case comes to us in a peculiar posture. The reporters appeal from two District Court orders: (1) in appeals Nos. 82-1820 and 82-1821, from the District Court's July 20, 1982 refusal to reexamine its order forbidding disclosure of documents on file with the court at the time the summary judgment motions were denied, and of documents to be used at trial, other than the portions read into evidence or used to examine witnesses; and (2) in appeal No. 83-1744, from the June 21, 1983 order releasing all documents. The latter is plainly unappealable by these appellants, since it granted them all the relief they requested at that point. "A party may not appeal from a judgment or decree in his favor." Electrical Fittings Corp. v. Thomas Betts Co., 307 U.S. 241, 242, 59 S.Ct. 860, 860, 83 L.Ed. 1263 (1939). Appeal No. 83-1744 is dismissed.
As for the appeals from the first order in each of the two cases, the ultimate release of the documents appears at first to moot any claims the reporters may have. Insofar as the reporters object to the documents' sealing in the past-during the trial proceedings (which seems to be their principal grievance) and after trial, prior to issuance of the second order — that complaint is beyond remedy. And the second order eliminates any grievance they might have had regarding sealing in the future. Claims otherwise moot, however, will nonetheless be entertained if "the underlying dispute between the parties is . . . 'capable of repetition, yet evading review,'" Nebraska Press Ass'n v. Stuart, 427 U.S. 539, 546, 96 S.Ct. 2791, 2797, 49 L.Ed.2d 683 (1976), quoting Southern Pacific Terminal Co. v. ICC, 219 U.S. 498, 515, 31 S.Ct. 279, 283, 55 L.Ed. 310 (1911). That exception applies if "there [is] a reasonable expectation that the same complaining party [will] be subjected to the same action again" and "the challenged action was in its duration too short to be fully litigated prior to its cessation or expiration." Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 349, 46 L.Ed.2d 350 (1975).
The "capable of repetition" requirement is clearly met here. In Gannett Co. v. DePasquale, 443 U.S. 368, 99 S.Ct. 2898, 61 L.Ed.2d 608 (1979), the Supreme Court held that Gannett's objection to a trial court order closing a suppression hearing in a criminal trial to the press presented a controversy "capable of repetition" because "it is reasonably to be expected that the petitioner, as publisher of two New York newspapers, will be subjected to similar closure orders entered by New York courts in compliance with the judgment of that State's Court of Appeals." Id. at 377-78, 99 S.Ct. at 2904. Similarly, it is reasonably to be expected that at least the Reporters Committee for Freedom of the Press, which is headquartered in the District of Columbia and regularly participates in litigation involving media access to judicial proceedings, will again be confronted by an order of this sort issued by the United States District Court for the District of Columbia. See Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 603, 102 S.Ct. 2613, 2618, 73 L.Ed.2d 248 (1982).
It is perhaps not likely that future orders restricting the Reporters Committee's access to summary judgment and trial materials will be entered at Mobil's behest. Gannett, however, considers the court entering the order to be the relevant factor. 443 U.S. at 377-78, 99 S.Ct. at 2904-05. If the requesting party were relevant, the facts here would present a stronger case for review than Gannett, where the order was entered at the request of a criminal defendant whose case had ultimately been disposed of by a plea bargain. Id. at 376 n. 4, 99 S.Ct. at 2904 n. 4.
Whether the "evading review" requirement is satisfied is less obvious. The issue, as we see it, is whether, without considering the possibility of expedited review (which would of course make the "evading review" test virtually impossible to meet), a sealing order is normally insusceptible of review before completion of trial in the case in which it is entered. Although in this case the issue of the sealing of the documents during trial came up very close to trial, because Mobil did not make a motion to keep them under seal until that time, Mobil could have moved for a protective order that would last through trial when it asked for its original protective order for discovery purposes. But even accepting the beginning of discovery as the starting point for computing the availability of review with regard to sealing both before and during trial, we still find that the reporters' claim would evade review. Globe Newspaper Co. v. Superior Court found that the issue of the constitutionality of closure of a criminal trial to the press evaded review "because criminal trials are typically of 'short duration.'" 457 U.S. at 603, 102 S.Ct. at 2618 (emphasis added) (citation omitted). The interval of time that typically elapses between the beginning of discovery and the completion of trial in a civil case is hard to measure, but it is safe to conclude that it is less than two years, which the Supreme Court has held short enough to cause action which would be mooted if not reviewed within that time to evade review. Southern Pacific Terminal Co. v. ICC, 219 U.S. at 514-16, 31 S.Ct. at 283-84.
We regard completion of trial as the cut-off date both because the order on appeal envisioned giving appellants what they desired (document-by-document justification for the sealing) shortly after trial (Mobil was to provide such justification within thirty days thereafter); and because the reporters' principal grievance is the denial of access during trial, when the material was current news. Mobil argues that the reporters' claims do not evade review because in a future case the District Court might decide never to unseal the documents. That is true, but irrelevant. If the case came to us in that posture, it would present a live controversy as to the future status of the documents, but with regard to their prior sealing would be just as moot as it is here. Furthermore, since the timing and length of the denial of access may determine its permissibility, see infra pages 1338-1339, review of the permanent sealing of documents would not present the same issue.
We find, therefore, that the reporters' claims arising out of the first (July 20) order are not barred by the mootness doctrine.
Mobil also argues that the July 20 order was not a "final decision," and hence we have no jurisdiction to entertain an appeal from it under 28 U.S.C. § 1291 (1982). That section provides: "The courts of appeals . . . shall have jurisdiction of appeals from all final decisions of the district courts of the United States." 28 U.S.C. § 1291. Ordinarily a district court renders a final decision from which appeal may be taken under § 1291 only when it has finished adjudicating the case before it. There is, however, a "small class [of decisions] which finally determine claims of right separable from, and collateral to, rights asserted in the action, too important to be denied review and too independent of the cause itself to require that appellate consideration be deferred until the whole case is adjudicated." Cohen v. Beneficial Industrial Loan Corp., 337 U.S. 541, 546, 69 S.Ct. 1221, 1225, 93 L.Ed. 1528 (1949). The reporters claim that the July 20 order is in that class. They contend that it finally determined that they were not entitled to a document-by-document assessment of confidentiality before sealing of the information, and resulted in their being denied access to it without that assessment during a critical period, the trial and its immediate aftermath.
We agree. In United States v. Hubbard, 650 F.2d 293 (D.C. Cir. 1980), this court held that a district court order to unseal documents obtained in connection with a criminal trial was a collateral order capable of being appealed immediately. There, as here, the party challenging the order was not a party to the underlying criminal proceeding. We found that the order was "'separable from, and collateral to' the rights of the parties to the . . . proceedings"; that because public access to the documents would irreparably damage the interests asserted," the unsealing order would as a practical matter "'finally determine' the claim"; and that consideration of the issues raised would neither halt nor disturb the orderly progress of the criminal proceeding. 650 F.2d at 314, quoting Cohen v. Beneficial Industrial Loan Corp., 337 U.S. at 546, 69 S.Ct. at 1225.
Mutatis mutandis, this analysis holds for the counterpart right of access to information asserted by the reporters. The reporters' claims arising from the July 20 order are just as separable from, and collateral to, the rights of the parties to the underlying proceeding. None of those parties is even involved in this appeal. Since the essence of the reporters' claims is that they had a right to the unprivileged information during trial, when it had greater news value, and not, as the July 20 order's provision for later submissions held forth as a prospect, after the conclusion of proceedings, the order "irreparably damage[d] the interest asserted," id., and as a practical matter finally determined the claims. Finally, appellate consideration of the reporters' claims would not disrupt the trial, even if it were still now in progress. Mobil's counsel, who were presumably somewhat involved in the trial representing Mobil witnesses, would be distracted; but a similar objection could have been made to treating the unsealing order as collateral in the Hubbard case.
Although appellants in cases of this sort typically assert a common law right of access to court records, the reporters place the issue before us in purely constitutional terms. They have framed and argued their appeal as follows: "Does the First Amendment prohibit a trial court from indefinitely postponing public access to the complete record of civil trials and summary judgment proceedings without a prior judicial determination that temporary sealing of specific documents is the least restrictive means necessary to protect a compelling governmental interest?" Brief for Appellants at 1 (Statement of Issue Presented).
No Supreme Court decision deals with the precise issue of the public's First Amendment rights to court records in civil cases. It was not until 1980 that the Supreme Court found that the "common core purpose of assuring freedom of communication on matters relating to the functioning of government" shared by the various clauses of the First Amendment created a right to observe criminal trial proceedings, Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 575, 100 S.Ct. 2814, 2826, 65 L.Ed.2d 973 (1980). The precise contours of that right are in the process of being drawn. See Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 102 S.Ct. 2613, 73 L.Ed.2d 248 (1982) (no exception for victims in rape trials); Press-Enterprise Co. v. Superior Court, 464 U.S. 501, 104 S.Ct. 819, 78 L.Ed.2d 629 (1984) (right extends to voir dire proceedings); cf. Waller v. Georgia, 467 U.S. 39, 104 S.Ct. 2210, 81 L.Ed.2d 31 (1984) (Sixth Amendment right extends to evidence suppression hearings). The principle has not yet been applied to access to civil trials (though the Court has perhaps intimated that it obtains there, see Richmond Newspapers, 448 U.S. at 580 n. 17, 100 S.Ct. at 2829 n. 17), much less to access to records in civil trials — or, for that matter, even records in criminal trials.
On its facts, perhaps the closest Supreme Court case in point is Nixon v. Warner Communications, Inc., 435 U.S. 589, 98 S.Ct. 1306, 55 L.Ed.2d 570 (1978), relating to the court records in a criminal case. The Court rejected press claims of a right to physical access to the so-called "Watergate tapes" introduced and played at the trial of former Attorney General John Mitchell. The holding is not, however, particularly helpful here. In addition to the fact that it was expressly tied closely to "all the circumstances of this concededly singular case," 435 U.S. at 608, 98 S.Ct. at 1317, it did not resolve the issue before us, even in the context of those circumstances. The Court found no common law right of access to the tapes, id., no right under the Sixth Amendment, id. at 610, 98 S.Ct. at 1318, and no right under the "freedom of press" clause of the First Amendment, id. at 608-10, 98 S.Ct. at 1317-19. The latter claim was curtly rejected on the ground that "the public [had] never had physical access" to the tapes in question, and that "[t]he First Amendment generally grants the press no right to information about a trial superior to that of the general public." Id. at 609, 98 S.Ct. at 1318. Presumably the Court believed that it was constitutional that the public had been denied access (one would surmise that from its finding of no common law right of access); but the point of a First Amendment right of public access was not raised and not explicitly decided.
The only other Supreme Court case dealing with First Amendment rights to court records (or at least potential court records) is Seattle Times Co. v. Rhinehart, 467 U.S. 20, 104 S.Ct. 2199, 81 L.Ed.2d 17 (1984). Though this was closer to the present case in that it involved a civil rather than a criminal trial, it was more remote in that it did not involve a claim by the public to access, but rather a claim by one of the parties of the right to disseminate information acquired in the course of pretrial discovery. The Court upheld the trial court's imposition of a protective order that restrained such dissemination. The information in question, however, had not, evidently, been filed with the court, much less introduced into evidence.
Since the Supreme Court has not spoken to the existence of a First Amendment right to court records of civil proceedings, we must resolve the question on the basis of the analysis which the Court has brought to bear in this general field. In deciding whether the public has a First Amendment right of access to judicial proceedings, the Court has made two inquiries: (1) whether the proceeding has historically been open, see Press-Enterprise Co. v. Superior Court, 104 S.Ct. at 822-23; Globe Newspaper Co. v. Superior Court, 457 U.S. at 605, 102 S.Ct. at 2619; Richmond Newspapers, Inc. v. Virginia, 448 U.S. at 569-70, 100 S.Ct. at 2823-24; and (2) whether the right of access plays an essential role in the proper functioning of the judicial process and the government as a whole, see Press-Enterprise, 104 S.Ct. at 823-24; Globe Newspaper, 457 U.S. at 606, 102 S.Ct. at 2619; Richmond Newspapers, 448 U.S. at 569-73, 100 S.Ct. at 2823-26. Apparently, both these questions must be answered affirmatively before a constitutional requirement of access can be imposed. An historical tradition of at least some duration is obviously necessary, particularly to support a holding based upon the remote implications of a constitutional text — which implications potentially include, in the case of the free speech clause of the First Amendment, not merely a right of access but all that furthers the "common core purpose of assuring freedom of communication on matters relating to the functioning of government." With neither the constraint of text nor the constraint of historical practice, nothing would separate the judicial task of constitutional interpretation from the political task of enacting laws currently deemed essential. The further requirement that the historical practice play "an essential role" in the proper functioning of government is also needed, since otherwise the most trivial and unimportant historical practices — for example, the courts' earlier practice of reading their judgments aloud in open session — would be chiselled in constitutional stone. Cf. Williams v. Florida, 399 U.S. 78, 102, 90 S.Ct. 1893, 1907, 26 L.Ed.2d 446 (1970) (no Sixth Amendment requirement of trial by a jury of twelve because "the fact that the jury at common law was composed of precisely 12 is a historical accident, unnecessary to effect the purposes of the jury system").
If the holding falls directly within the text of the constitutional provision — as in Seattle Times, where not an implied right of access but the expressly protected right of "freedom of speech" was at issue — historical practice may be relegated to a lesser role, serving not as a sine qua non but as an important tool for interpretation.
As to the historical tradition: The Supreme Court opinion in Seattle Times contains the statement that "to the extent that courthouse records could serve as a source of public information, access to that source customarily is subject to the control of the trial court," 104 S.Ct. at 2207-08 n. 19. This was said, however, in the context of a discussion of discovery materials not yet introduced into evidence, and the Court gave as illustrations of the "control" it had in mind the ability of the trial court to exempt such materials from filing requirements, or to provide for filing under seal. Despite the breadth of its expression, this statement cannot reasonably be considered support for the proposition that courts have traditionally had plenary power to close civil records in general to public access. That is clear from the qualification included within the Court's later statement that "restraints placed on discovered, but not yet admitted, information are not a restriction on a traditionally public source of information," id. at 2208 (emphasis added).
Nixon v. Warner Communications contains much more discussion relevant here. Though, as noted earlier, the case did not reach the First Amendment issue, it did consider the traditional practice with regard to court records at some length, in connection with its holding on the common law right of access. The practice immediately relevant was that relating to criminal trials, but the discussion goes well beyond that, and warrants quotation at some length:
It is clear that the courts of this country recognize a general right to inspect and copy public records and documents, including judicial records and documents. In contrast to the English practice, see, e.g., Browne v. Cumming, 10 B. C. 70, 109 Eng.Rep. 377 (K.B. 1829), American decisions generally do not condition enforcement of this right on a proprietary interest in the document or upon a need for it as evidence in a lawsuit. The interest necessary to support the issuance of a writ compelling access has been found, for example, in the citizen's desire to keep a watchful eye on the workings of public agencies, see, e.g., State ex rel. Colscott v. King, 154 Ind. 621, 621-27, 57 N.E. 535, 536-38 (1900); State ex rel. Ferry v. Williams, 41 N.J.L. 332, 336-39 (1879), and in a newspaper publisher's intention to public information concerning the operation of government, see, e.g., State ex rel. Youmans v. Owens, 28 Wis.2d 672, 677, 137 N.W.2d 470, 472 (1965), modified on other grounds, 28 Wis.2d 685a, 139 N.W.2d 241 (1966). But see Burton v. Reynolds, 110 Mich. 354, 68 N.W. 217 (1896).
It is uncontested, however, that the right to inspect and copy judicial records is not absolute. Every court has supervisory power over its own records and files, and access has been denied where court files might have become a vehicle for improper purposes. For example, the common-law right of inspection has bowed before the power of a court to insure that its records are not "used to gratify private spite or promote public scandal" through the publication of "the painful and sometimes disgusting details of a divorce case." In re Caswell, 18 R.I. 835, 836, 29 A. 259 (1893). Accord, e.g., C. v. C., 320 A.2d 717, 723, 727 (Del. 1974). See also King v. King, 25 Wyo. 275, 168 P. 730 (1917). Similarly, courts have refused to permit their files to serve as reservoirs of libelous statements for press consumption, Park v. Detroit Free Press Co., 72 Mich. 560, 568, 40 N.W. 731, 734-35 (1888); see Cowley v. Pulsifer, 137 Mass. 392, 395 (1884) (per Holmes, J.); Munzer v. Blaisdell, 268 App.Div. 9, 11, 48 N.Y.S.2d 355, 356 (1944); see also Sanford v. Boston Herald-Traveler Corp., 318 Mass. 156, 158, 61 N.E.2d 5, 6 (1945), or as sources of business information that might harm a litigant's competitive standing, see, e.g., Schmedding v. May, 85 Mich. 1, 5-6, 48 N.W. 201, 202 (1891); Flexmir, Inc. v. Herman, 40 A.2d 799, 800 (N.J.Ch. 1945).
On the basis of this discussion, we take it as a given that there is a tradition of public access to court records, and that that right is not absolute. The factor most obviously distinguishing the request for records in the present case from the requests at issue in the vast majority of reported cases — and the factor that obviously caused the District Court to deny access without the document-by-document examination that ordinarily accompanies Rule 26(c) protective orders — was the pendency of the litigation at the time the request was made. We must consider, therefore, whether the tradition of public access includes pre-judgment access.
An early opinion of this court speaks to the point. Ex parte Drawbaugh, 2 App. D.C. 404 (1894), ruled on a motion made in connection with an appeal from a decision of the Patent Office. The appellant, "upon filing the transcript of the papers pertaining to the appeal with the clerk of this court, . . . moved the court . . . 'that the files relating to the above appeal be preserved in secrecy, and that the clerk of this court be directed not to permit said files, nor any part thereof, to be inspected, nor any copy to be taken from said files, except upon request of the appellant or his attorneys, or upon special order of this court first obtained, after due notice to appellant's attorneys.'" Id. at 404. The motion was denied. In the course of its opinion, the court made the following observations relevant to the issue now before us:
But the record or transcript brought into this court on appeal, after judgment entered on the proceedings by the tribunal appealed from, do not stand upon the footing of original papers placed in the files of a court of original jurisdiction, and where there has been no trial had or judgment entered thereon.
Id. at 406 (emphasis added).
[T]here is also a distinction made in some of the cases between the right to inspect judicial records after trial, and the right to inspect and take copies from papers merely filed, but before any action had thereon by the court. In the latter case, it has been held, in one instance at least, that the court might withhold from a publisher of a newspaper the right to inspect and take copies of papers or documents on file, for publication before the trial of the cause. That was done in the case of Schmedding v. May, 85 Mich., 1 [48 N.W. 201], and the same power was intimated to exist, rather than decided, in the case of Cowley v. Pulsifer, 137 Mass., 392. But those decisions have no application to this case.
Id. at 407.
The two cases cited in the foregoing excerpt — both of which were also referred to by the Supreme Court in Nixon, see supra page 1333 — assert as the common law rule that there is no right of public access to prejudgment records in civil cases. The first of them, Schmedding v. May, 85 Mich. 1, 48 N.W. 201 (1891), was a holding by the Michigan Supreme Court, which was soon followed by another holding of the same court to the same effect, Burton v. Reynolds, 110 Mich. 354, 68 N.W. 217 (1896). The second case, Cowley v. Pulsifer, 137 Mass. 392 (1884), contained a dictum on the point by Justice Holmes. A holding to the same effect appears in a 1953 decision of the United States District Court for the Northern District of Illinois, Birnbaum v. Wilcox-Gay Corp., 17 F.R.D. 133, 139, 140 (N.D.Ill. 1953). And a number of cases, including several federal cases, refer to the pre-judgment rule adopted by other courts, without confirming that it represents general common law, but likewise without disapproving it, see Ex parte Drawbaugh, 2 App.D.C. at 407, quoted supra at page 1333; In re Sackett, 136 F.2d 248, 249 (C.C.P.A. 1943); Werfel v. Fitzgerald, 23 A.D.2d 306, 309, 260 N.Y.S.2d 791, 795 (1965); or express what they regard as the common law rule in a fashion that allows for the prejudgment exception, see Ex parte Drawbaugh, 2 App.D.C. at 406, quoted supra at page 1333.
The dissent contends that Schmedding, Burton and Birnbaum do not constitute holdings because on their facts they excluded access only pre-trial and not after the commencement of trial but pre-judgment. See Dissent at 1348-49, 1350 n. 20. This is a quibble over the meaning of "holding." The rationale on which the court rested its decision in all three of these cases was that there was no right of access until judgment. If one chooses not to consider that a holding, it is no less forceful — especially given the absence of any contrary expression in any case we have been able to discover — as an indication of the common-law rule.
The dissent seeks to make much of the fact that the language used to describe the rule in the foregoing cases is not consistent, but sometimes refers to the absence of a right of access "before trial," or "until after trial," rather than "before judgment." See Dissent at 1348-51. But the former variation ("before trial") is entirely consistent with the pre-judgment access rule, which, as described above, holds that access is not a matter of right before judgment except to the extent that material is disclosed at trial. It is an entirely natural and accurate description of the consequences of the rule to say that there is no right of access "before the trial of the cause," Drawbaugh, 2 App.D.C. at 407, or "prior to trial," Werfel, 23 A.D.2d at 309, 260 N.Y.S.2d at 795, or "before trial," In re Sackett, 136 F.2d at 249. In fact Schmedding itself uses this simplified formulation, see 85 Mich. at 4, 48 N.W. at 202 (characterizing issue as whether there is a "right to an examination of the records and papers in a cause . . ., before trial or hearing, or before they become public by proceedings in open court"). As for the latter variation (no right of access until "after trial"), see, e.g., Drawbaugh, 2 App.D.C. at 407; Birnbaum, 17 F.R.D. at 139, that is admittedly an inaccurate description of the rule, though only by implication — i.e., it is entirely true that under the pre-judgment rule there is no right of access until trial is completed, but the implication that there is a right of access after trial is completed is wrong. The implication is in any event not inevitable ( cf. the longstanding controversy among biblical exegetes over "And [Joseph] knew her not till she had brought forth her first-born son," Matthew 1:25), and is particularly weak when one considers that the time between conclusion of trial and judgment is ordinarily brief, and that none of the cases using this formulation involved that interval, so that in the writer's mind the conclusion of trial and judgment may have been synonymous. Since all of the cases using the variations cite Schmedding or annotations based on Schmedding (except Sackett, whose reference to "some cases," 136 F.2d at 249, must allude to the same line of authority) as support for the rule they are applying or describing, there is no reasonable basis to believe that they are setting forth anything but the Schmedding rule itself. And Schmedding, despite the dissent's selective quotation to establish the contrary, is unequivocal that entry of judgment is the crucial stage.
It is, moreover, extravagant to believe, as the dissent evidently does, see Dissent at 1348 n. 17, that Schmedding's exception for material "made public . . . by proceedings in open court," 85 Mich. at 7, 48 N.W. at 202 (citing Cowley v. Pulsifer), meant to generate access, at the time of trial, to all material entered into evidence at the trial, whether or not read aloud or displayed to the jury in a fashion visible to the public. To think that, one would have to accept that the court, writing in the days before photostatic copying, envisioned the passing around of documentary exhibits from the jury to the audience, or the manual copying of all of them. The more natural meaning of the phrase "made public by proceedings" and the reason of the matter unite to establish that the court had in mind only those disclosures to the viewing audience automatically produced by the oral proceedings in a traditionally open trial.
A suspicion that the pre-judgment nonaccess rule represented the general common law is aroused by the natural connection between such a policy of nonaccess and another common law rule that was indisputably (at least prior to 1927) "supported by an almost unbroken line of authority," Annot., 52 A.L.R. 1438, 1438 (1928): the rule that the "public records privilege" — the privilege against liability for defamation in the accurate reporting of public records — did not extend to "accusations contained in papers filed by a party and not yet brought before a judge or magistrate for official action." Sanford v. Boston Herald-Traveler Corp., 318 Mass. 156, 158, 61 N.E.2d 5, 6 (1945). See also Park v. Detroit Free Press Co., 72 Mich. 560, 568-69, 40 N.W. 731, 734 (1888). It would be strange, if not unthinkable, see Cowley v. Pulsifer, 137 Mass. at 396, to assess civil liability for bringing to the public's attention government records which the public is entitled to see. The reasons that support the one rule support the other as well. Schmedding's perception that matters in a civil lawsuit "involve private dealings between private parties" until made public in open court, or until "their truthfulness has been determined by the judgment or decree of the court," 85 Mich. at 5, 48 N.W. at 202, was the very basis for that court's earlier holding that pre-judgment civil records do not come within the public records privilege:
One of the reasons why parties are privileged from suit for accusations made in their pleadings is that the pleadings are addressed to courts where the facts can be fairly tried, and to no other readers . . . . The public have no rights to any information on private suits till they come up for public hearing or action in open court; and, when any publication is made involving such matters, they possess no privilege, and the publication must rest on either nonlibelous character or truth to defend it. A suit thus brought with scandalous accusations may be discontinued without any attempt to try it, or on trial the case may easily fail of proof or probability. The law has never authorized any such mischief.
Park v. Detroit Free Press, 72 Mich. at 568-69, 40 N.W. at 734. And Justice Holmes' expression of the reasons why public policy does not require that publication of pre-judgment records enjoy the same immunity from defamation liability as publication of post-judgment records applies as well to a distinction between public access to the two:
The chief advantage to the country which we can discern [from application of the public records privilege to judicial records] . . . is the security which publicity gives for the proper administration of justice. . . . It is desirable that the trial of causes should take place under the public eye, not because the controversies of one citizen with another are of public concern, but because it is of the highest moment that those who administer justice should always act under the sense of public responsibility, and that every citizen should be able to satisfy himself with his own eyes as to the mode in which a public duty is performed.
[I]t is clear that [these grounds] have no application whatever to the contents of a preliminary written statement of a claim or charge. These do not constitute a proceeding in open court. Knowledge of them throws no light upon the administration of justice. Both form and contents depend wholly on the will of a private individual, who may not be even an officer of the court.
Cowley v. Pulsifer, 137 Mass. at 394.
Because of their sparseness, the authorities discussed above are perhaps weak support for a general common law rule of nonaccess to pre-judgment records in private civil cases. But when laid beside our inability to find any historical authority, holding or dictum, to the contrary, they are more than enough to rule out a general tradition of access to such records. It is possible that modern American practice is to the contrary — though that is less than clear from the cases, and it is risky to generalize from one's familiarity with the practice in a few jurisdictions, or, for that matter, to assume that a practice of granting access where no objection is made establishes the existence of an acknowledged right to access. See, e.g., In re McLean, 16 F.Cas. 237, 239 (C.C.S.D. Ohio 1879) (No. 8,877) (in a case where no one but the clerk of court resisted the provision of records to a person who had no qualifying private interest in them, "[t]he judges afterwards granted ex gratia what they ruled the petitioner was not entitled to, as a matter of law"). But even accepting as representative of current law the modern dicta which, without distinguishing between pre-judgment and post-judgment, announce a "public right of access" in civil cases, see, e.g., Ottaway Newspapers, Inc. v. Appeals Court, 372 Mass. 539, 362 N.E.2d 1189 (1977), we cannot discern an historic practice of such clarity, generality and duration as to justify the pronouncement of a constitutional rule preventing federal courts and the states from treating the records of private civil actions as private matters until trial or judgment.
See Times-Call Publishing Co. v. Wingfield, 159 Colo. 172, 410 P.2d 511 (1966) (en banc) (interpreting Colorado statute that denies pre-judgment access); State ex rel. Williston Herald, Inc. v. O'Connell, 151 N.W.2d 758, 762-63 (N.D. 1967) (pre-judgment access denied, even in criminal case); C. v. C., 320 A.2d 717, 724 (Del. 1974) ("It is generally held that judicial records are subject to inspection after completion of the proceedings") (dictum). The closely related pre-judgment exception to the public records privilege rule, discussed above, continues to be the majority rule in this country. See Newell v. Field Enterprises, Inc., 91 Ill.App.3d 735, 745-46, 47 Ill.Dec. 429, 438-39, 415 N.E.2d 434, 443-44 (1980).
Many statutes express an absolute right of access to court records. See, e.g., ILL.ANN.STAT. ch. 25, § 16(6) (Smith-Hurd Supp. 1984-85); VA. CODE § 17-43 (1982); W. VA. CODE § 51-4-2 (1981). But the purpose of most if not all of these was evidently to eliminate the English rule that no civil court records, pre-judgment or post-judgment, were accessible to persons without some private interest in them. See, e.g., Sloan Filter Co. v. El Paso Reduction Co., 117 F. 504, 507 (C.C.D.Colo. 1902); In re McLean, 16 F.Cas. 237, 238 (C.C.S.D. Ohio 1879) (No. 8,877). They have not been read to prevent judicial withholding entirely, see, e.g., Deere Co. v. Finley, 103 Ill.App.3d 774, 775-76, 59 Ill.Dec. 444, 445-46, 431 N.E.2d 1201, 1202-03 (1981), and thus need not be read to exclude judicial withholding prior to judgment.
Our substantial, though perhaps not exhaustive, research has disclosed no case, even in recent times, explicitly discussing and rejecting the pre-judgment nonaccess rule. Indeed, we find few cases without such discussion that reverse a trial court's denial of access (or even affirm a trial court's grant of access) in a civil matter prior to merits judgment, and in most of these either (1) a government agency was a party, see The Miami Herald Publishing Co. v. Collazo, 329 So.2d 333 (Fla.Dist.Ct.App. 1976), cert. denied, 342 So.2d 1100 (Fla. 1976); Charlottesville Newspapers, Inc. v. Berry, 215 Va. 116, 206 S.E.2d 267 (1974), or (2) at least with respect to some of the parties, the action had been dismissed (and probably dismissed with prejudice), see In re Continental Illinois Securities Litigation, 732 F.2d 1302 (7th Cir. 1984); Mary R. v. B. R. Corp., 149 Cal.App.3d 308, 196 Cal.Rptr. 871 (Cal.Ct.App. 1983); The Miami Herald Publishing Co. v. Collazo, supra. The former category of case appears always to have been outside the pre-judgment nonaccess rule, see Schmedding, 85 Mich. at 5, 48 N.W. at 202 (referring to civil suits that involve "private dealings between private parties," 85 Mich. at 5, 48 N.W. at 202); and perhaps "with prejudice" dismissals were regarded as judgments (insofar as the sought-after records pertained to parties and claims dismissed) — at least where the dismissal was for a reason other than settlement, see 85 Mich. at 6, 48 N.W. at 202 (justifying the non-access rule on the ground that "[t]hese suits, involving private transactions, may never come to trial or hearing. The troubles may be settled, and the charges withdrawn").
The same conclusion would result from application of the second of the Supreme Court's tests for First Amendment entitlement to access: whether access plays an essential role in the proper functioning of the judicial process. The Court has found that open criminal trials enhance the quality and safeguard the integrity of factfinding, see Globe Newspaper, 457 U.S. at 606, 102 S.Ct. at 2619; Richmond Newspapers, 448 U.S. at 569, 100 S.Ct. at 2823; assure an appearance of fairness, see Press-Enterprise, 104 S.Ct. at 823; Globe Newspaper, 457 U.S. at 606, 102 S.Ct. at 2619; Richmond Newspapers, 448 U.S. at 569-70, 100 S.Ct. at 2823-24; function as a check on the judicial and governmental process, see Globe Newspaper, 457 U.S. at 606, 102 S.Ct. at 2619; Richmond Newspapers, 448 U.S. at 569, 100 S.Ct. at 2823; and play a cathartic role in permitting the community to observe justice being done, see Press-Enterprise, 104 S.Ct. at 823-24; Richmond Newspapers, 448 U.S. at 570-72, 100 S.Ct. at 2823-25. Even assuming, as seems unlikely, that these functions are as important in the context of civil suits between private parties as they are in criminal prosecutions, they are not greatly enhanced by access to documents (which, unlike live proceedings, do not contain unrecordable subtleties) before judgment rather than after. In fact, such access appears to be rarely requested, and its denial does not provoke the kind of outcry which closing of a trial usually excites.
The dissent's response to this point is to appeal to Bridges v. California, 314 U.S. 252, 62 S.Ct. 190, 86 L.Ed. 192 (1941), which stresses that the public interest in obtaining news is an interest in obtaining contemporaneous news. See Dissent at 1352-53. Quite so. But the dissent is mistaken in assuming that the public interest in obtaining news is the focus of inquiry in right-of-access cases, as it was in Bridges, which involved the attempted suppression of editorial comment concerning pending litigation. Rather, as Globe, Richmond Newspapers, and Press Enterprise make clear, the focus is upon the public's ability to assure proper functioning of the courts. Contemporaneity of access to written material does not significantly enhance that ability.
We have been assuming in the foregoing discussion that the relevant inquiry is, as the reporters have framed it, whether there is an important tradition of public access to court records. The more precise inquiry, however, is a functional rather than classificational one: whether information of the sort at issue here — regardless of its prior or current classification as court records — was traditionally open to public scrutiny. An historical tradition of access to civil and criminal judgments, adequate to sustain a constitutional claim, could hardly be defeated, for example, by a state statute providing that henceforth such judgments will not become part of court records. Contrariwise, a new constitutional right of public access to litigants' personal financial data would hardly be created by a state statute requiring such data to be filed with the court, simply because all other court records have been (let us assume) traditionally open. This distinction is significant here with regard to the pre-trial materials sought by the reporters — i.e., those materials never introduced at trial — the vast majority of which consisted of depositions and discovery documents.
Traditionally, absent a statute or court order, even parties to the case did not have the right to inspect depositions taken at the behest of their opponents. See 18 C.J. Depositions § 312 (1919). Requirements for the filing and "publication" of depositions were designed to make their contents known to the litigants, and even these requirements were far from absolute. See id. at § 300. And when filed, unless delivered to the court personally by the magistrate who took them, depositions were filed under seal, and "[u]nless by consent . . . [could not] be opened out of court. . . ." Id. at § 306. See Beale v. Thompson, 12 U.S. (8 Cranch) 70, 3 L.Ed. 491 (1814). The tradition that the opening ( i.e., publication) of depositions merely made them available to parties was carried forward following passage of the Federal Rules of Civil Procedure in 1938, as most district courts adopted local rules providing that depositions, which the rules required to be filed "securely seal[ed] . . . in an envelope," FED. R.CIV.P. 30(f)(1), were to remain sealed except "upon request of any counsel in the case." Cohn, Federal Discovery: A Survey of Local Rules and Practices in View of Proposed Changes to the Federal Rules, 63 MINN.L.REV. 253, 284 (1979). The current Federal Rules require deposition transcripts to be filed, but the court may on its own motion dispense with the requirement, FED.R.CIV.P. 5(d). Many district courts have adopted local rules making such dispensation standard except where the depositions are needed in connection with motion proceedings, and excluding public access even where filing occurs. See Marcus, Myth and Reality in Protective Order Litigation, 69 CORNELL L.REV. 1, 13-14 nn. 61-62 (1983). It can thus hardly be said that there was a tradition, or is even now a general practice, of public access to pretrial depositions. As for discovery documents: Besides the fact that they (at least in their currently permitted scope — including all third-party documents such as those at issue here) are a relatively modern phenomenon, see 18 C.J. Discovery §§ 111-16 (1919), the Federal Rules continue to contain no provision for their filing. See Marcus, op. cit., at 13.
Publication meant "the open showing of depositions and giving copies of them to the parties by the clerks or examiners in whose custody they are." 18 C.J. Depositions § 311 at 721 n. 59 (1919) (emphasis added).
Chief Justice Burger, in a passage of his concurring opinion in Gannett cited approvingly by the Court in Seattle Times, described the past and present practice with regard to pre-trial proceedings in general as follows:
Even though the draftsmen of the Constitution could not anticipate the 20th-century pretrial proceedings to suppress evidence, pretrial proceedings were not wholly unknown in that day. Written interrogatories were used pretrial in 18th-century litigation, especially in admiralty cases. . . . Yet, no one ever suggested that there was any "right" of the public to be present at such pretrial proceedings as were available in that time; until the trial it could not be known whether and to what extent the pretrial evidence would be offered or received.
Similarly, during the last 40 years in which the pretrial processes have been enormously expanded, it has never occurred to anyone, as far as I am aware, that a pretrial deposition or pretrial interrogatories were other than wholly private to the litigants.
Gannett Co. v. DePasquale, 443 U.S. at 396, 99 S.Ct. at 2914 (Burger, C.J., concurring), cited with approval in Seattle Times Co. v. Rhinehart, 104 S.Ct. at 2208. It is true that in the present case the reporters were only seeking those pretrial materials that could have been considered by the court in its disposition of the Rule 56(c) motion — what they call the "summary judgment record." Chief Justice Burger's analysis does not make any such distinction, though it would be an obvious one to make if it were relevant. We are certainly unaware of any tradition of public access (pre- or post-judgment) to all documents consulted (or, as appellants would have it, consultable) by a court in ruling on pre-trial motions. If such a tradition existed, public files would presumably be filled with complaints stricken as scurrilous and with proffered evidence ruled inadmissible. The passage of Seattle Times which cites Chief Justice Burger's analysis with approval evidently considers the admission of evidence the touchstone of a First Amendment right to public access: "Therefore, restraints placed on discovered, but not yet admitted, information are not a restriction on a traditionally public source of information." 104 S.Ct. at 2208 (emphasis added). Even if one were to expand this perception to include all admissible evidence, it would still lead to the conclusion that material placed before the court in connection with summary judgment motions is not constitutionally required to be open to the public — unless we are to subject trial courts to the constitutional necessity of ruling, either pre-trial or post-trial, on the admissibility of voluminous material that is filed, and perhaps even referred to in the summary judgment motion, but not sought to be introduced.
Since the functional analysis we have just undertaken does not resolve the case as to all the documents at issue (certainly not as to those introduced into evidence), we return to the pre-judgment distinction we discussed earlier, to determine whether that suffices to dispose of the entire matter. Final judgment in the libel case against the Post was entered upon the trial court's grant of judgment n.o.v. in favor of the defendants on May 2, 1983, and in the slander case against Piro upon the grant of judgment n.o.v. in defendant's favor on May 20, 1983. We therefore find that until May 2, 1983 and May 20, 1983, respectively, the District Court could, without violating the First Amendment to the Constitution, categorically refuse the reporters access to the materials at issue in the two cases. The reporters thus had no First Amendment entitlement to a document-by-document determination of the need for confidentiality prior to those dates.
Part of the reporters' grievance, however, consists of the District Court's continuing failure to grant the requested access until June 21, 1983, more than a month after the entry of final judgment in the suit against Piro and more than seven weeks after the final judgment in the suit against the Post. It is questionable, to begin with, whether a challenge to the July 20, 1982 denial of the motion to reconsider the sealing order — which denial did not even mention the date on which unsealing would be ruled upon, much less specify that that date would be after entry of judgment — is the proper vehicle to place in issue the post-judgment delay. Even if a new motion to be accorded immediate access was not necessary, and a ruling on the earlier motion could be regarded as having been merely deferred, surely the appropriate means of appealing the allegedly unconstitutional continuation of the deferral to a point too long after the entry of judgment would have been a petition for mandamus rather than the appeal of the earlier order. Assuming that the point is properly before us, however, we think it readily disposed of.
To the extent a First Amendment right to post-judgment civil records exists, it does not exceed, for the reasons discussed earlier, the traditional common law right. That was not conceivably violated here. The District Court had, as early as July 20, 1982, determined that its final ruling on confidentiality would turn on Mobil's ability to provide specific, document-by-document justification for the claim of confidentiality — which is precisely the substantive standard, though not the time frame, insisted upon by the reporters. Thus, the only issue with regard to post-judgment access is the propriety of retaining the seals on the documents after entry of judgment in the two cases, pending the final (document-specific) determination. For obvious reasons, courts have uniformly approved the practice of provisionally sealing documents pending assessment of justification for a request to seal. See In re Knight Publishing Co., 743 F.2d 231, 235 n. 1 (4th Cir. 1984); Publicker Industries, Inc. v. Cohen, 733 F.2d 1059, 1073 (3d Cir. 1984); Hubbard, 650 F.2d at 324-25. Mobil's claim, supported by the MacDonald affidavit, that disclosure of the material would be likely to injure its competitive standing established a prima facie case for placing it under seal until the confidentiality claims could be ruled on. It is true that Mobil had, by the time the judgments n.o.v. were entered, abandoned the claim of confidentiality with regard to almost all of the documents initially designated under the protective order. That could have been regarded, as appellants suggest, as casting doubt on the totality of the MacDonald affidavit. But it did not have to be so regarded — and could indeed have been thought to suggest a particularly high need for confidentiality with respect to the documents remaining. What view to take was assuredly one of those matters regarding access "best left to the sound discretion of the trial court." Nixon, 435 U.S. at 599, 98 S.Ct. at 1312. The reporters make much of the fact that at the end of the day all of Mobil's confidentiality claims were either abandoned or found to be unwarranted. That, however, goes only to the merits of the claims, not to whether the affidavit asserting them could reasonably be regarded as sufficient to overcome the presumption of access pending a determination of the merits. The only remaining question, then, is whether the provisional seals were retained after judgment for an unreasonable length of time; in making its final ruling and setting aside the provisional seals in the two cases within fifty and thirty-two days, respectively, of the entries of judgment n.o.v., the District Court did not, in our estimation, exceed the limits of its discretion.
* * * * *
We have addressed in this opinion only the assignment of error presented, holding that the District Court's action did not violate the First Amendment to the Constitution of the United States. Conceivably, it violated the federal common law, which can of course go beyond constitutional prescriptions — either because the pre-judgment nonaccess rule was never so general as to include the federal courts, or because the federal rule (by reason of adoption of the Federal Rules of Civil Procedure or otherwise) has changed.
Although these points were neither raised, briefed nor argued, the dissent takes us to task for not considering them — perhaps after requesting supplemental briefing — and for thus reaching the constitutional question unnecessarily. see Dissent at 1342. That charge is unfounded. We decline to consider the existence of a common law right to access in these circumstances for the same reason (presumably) appellants declined to raise and argue it: In the absence of some overriding constitutional command to provide access, the eminently reasonable action of the trial court in deferring his ruling on these complex matters unrelated to the dispute between the parties, declining to interrupt an ongoing trial for that purpose, seemed to us unquestionably lawful. The prudential rule against reaching constitutional issues unnecessarily, see, e.g., Rescue Army v. Municipal Court of Los Angeles, 331 U.S. 549, 568-72, 67 S.Ct. 1409, 1419-21, 91 L.Ed. 1666 (1947), does not require a vain excursion through areas of law considered potentially helpful neither by the parties nor the court.
The dissent's excursion into the modern common law of access does not show our estimation of its futility to have been wrong. The dissent concludes that current law does not provide a right of access except with respect to documents designated as exhibits at trial — so with respect to the remaining documents at issue here, even in the dissent's view the constitutional issue must be reached. Even as to designated exhibits, however, the dissent's avoidance of the constitutional question is illusory. What the dissent would prescribe is a "minimal" common-law requirement that the trial judge demand "document-by-document justification of . . . confidentiality" before he renders designated trial exhibits inaccessible to the public pending later judgment; but no requirement (which the dissent evidently agrees would be unreasonable in a case such as this, involving "massive documentation") that the trial judge immediately rule upon the claim of confidentiality. Dissent at 1344. If this is correct (which is debatable, since document-by-document justification will often disrupt the responding party's ability to conduct a pending case as much as document-by-document ruling on the "provisional sealing" will disrupt the judge's), then the dissent has assuredly provided useful guidance to our district court for the future. But it has not determined the invalidity of the July 20, 1982 denial, in this case, of the motion here on appeal — which sought not a more precise statement of Mobil's basis for its claims of confidentiality (followed by a deferral of the ruling on those claims), but a ruling, then and there, and a release of the subject documents. It would be extraordinary to reverse a trial judge for failing to provide something that was not remotely the relief requested, and not remotely the relief sought on appeal. In other words, the dissent demonstrates that our answer to the question of the utility of the common law is wrong only by changing the question to one that is not presented here. Since, even on the dissent's analysis, the district court would have to be affirmed with regard to all of the documents on common-law grounds, consideration of constitutional grounds for reversal was essential.
Moreover, even if the petitioners could be deemed properly to have requested something short of a conclusive ruling on Mobil's confidentiality claims, we cannot imagine what request they would have been entitled to, under the most favorable assumptions of modern common law, that would warrant a reversal here. The dissent's categorical rule that no claim for confidentiality of trial exhibits can be sustained unless accompanied by a document-by-document justification is simply unworkable. At the time the requirement to provide such justification would have arisen in the present case, Mobil's counsel were engaged in preparing for and attending depositions of Mobil witnesses; according to Mobil's uncontested representation, depositions noticed by the Post, most of which involved Mobil officers and employees, "took up the entire two months prior to" the trial, and even continued mornings and nights during the trial itself. Brief for Appellees at 17. See also Transcript at 2351 (July 16, 1982) (at argument on Reporters' request for access, lead counsel for Mobil stated the request had to be discussed with his client by another attorney, since "I attended the deposition last night that lasted until 10:00 o'clock"). It does not suffice to say that this problem was Mobil's fault for making its sealing request at the "eleventh hour." Dissent at 1355 n. 27. Any request for such protection, other than one for blanket sealing of all trial exhibits (which we presume the dissent does not intend to promote) must be made close to the date of trial, since only then are the parties' lists of proposed trial exhibits available. Here those lists were filed on July 2, 1982, and trial began five days later (never mind that two of those days were a weekend, and a third an official holiday). Given the other matters on which counsel were occupied, it was effort enough (and surely a demonstration of good faith) to review the approximately 600 Mobil documents (comprising a much larger number of pages) included in plaintiffs' and defendants' proposed trial exhibits, and to limit the confidentiality claims to such degree that of the approximately 150 Mobil documents considered important enough ultimately to be used at trial, only 52 were within the remaining scope of the confidentiality claim. It would have been excessive to expect in addition, at that state of the proceedings, the crafting of a document-by-document specification of bases for the claim that would sustain post-trial legal attack. And the difficulties encountered in the present case are as nothing compared with those that a major antitrust trial would present. In sum, the dissent's made-for-the-occasion categorical rule of instantaneous document-by-document justification is utterly infeasible; and any feasible rule, which would have to accord the district judge a reasonable degree of discretion, could not conceivably have been violated here, where the document-by-document justification was required within 30 days after conclusion of the trial.
It appears that the proposed lists were transmitted to Mobil's counsel some time before their filing, but close enough beforehand that, as of July 1, review of the documents simply to determine whether they were subject to the preexisting protective orders was "not as yet complete." Letter of July 1, 1982, from Loren Kieve to John J. Walsh, et al. at 1, Tavoulareas v. Washington Post Co., Civil Nos. 80-3032 80-2387 (D.D.C. filed July 2, 1982). It was envisioned that after this review the documents would be classified into two categories, those as to which the protection would continue and those as to which it would not. Id.
The appeal in No. 83-1744 is dismissed. The ruling in appeals Nos. 82-1820 and 82-1821 is
I concur in the judgment of the court insofar as it rejects appellants' claims to a right of access to documents considered by the District Court in denying a summary judgment motion. But because I believe that appellants have a right of access to designated trial exhibits, I must respectfully dissent from that part of the majority's opinion affirming the provisional sealing of such exhibits.
It is my view that appellants' right of access to designated trial exhibits is secured by federal common law. Unfortunately, the majority has seized this occasion to issue an advisory opinion on the scope of the First Amendment. Ordinarily, I would not comment on such unnecessary constitutional ruminations. The dangerous drift of the majority's analysis, however, leaves me little alternative. Contrary to the view of other United States Courts of Appeals, the majority finds that the First Amendment does not provide the slightest protection to the public's interest in contemporaneous access to evidentiary exhibits in civil proceedings. In so doing, the majority opens the door to the abuse of provisional seals entered on the basis of mere summary affidavits. Proponents of such seals will be able to make broad, unsubstantiated, claims of confidentiality and prevent public access to critical evidentiary exhibits until public interest in such documents has long faded. I fear that public understanding of judicial decision-making will suffer accordingly.
I. THE COMMON LAW RIGHT OF ACCESS
As the Supreme Court has recently reminded us, federal courts ought "not decide a constitutional question if there is some other ground upon which to dispose of the case." Lowe v. SEC, ___ U.S. ___, 105 S.Ct. 2557, 2563, 86 L.Ed.2d 130 (1985), quoting Escambia County, Florida v. McMillan, 466 U.S. 48, 104 S.Ct. 1577, 1579, 80 L.Ed.2d 36 (1984). See also Ashwander v. TVA, 297 U.S. 288, 347, 56 S.Ct. 466, 483, 80 L.Ed. 688 (1936) (Brandeis, J., concurring). Presumably mindful of this well established principle, the majority justifies its declaration of constitutional law by reference to the parties' failure to brief the common law issues in this case. See majority opinion (maj. op.) at 1340.
The Supreme Court, however, has also indicated that the mere failure of the parties to brief a non-constitutional issue does not legitimate constitutional adjudication. Escambia County, supra, 104 S.Ct. at 1579. An order directing the parties to submit supplemental briefs on the common law issues would have addressed the majority's concern. Although such a procedure might be inadvisable if it deprived the litigants of timely justice, all concede that our judgment in this case is merely prospective. Consequently, there would have been little to lose and much to gain by prudent exercise of our authority to require supplemental briefing of the common law issues in this case.
This is essentially the course followed by the Escambia Court. In Escambia, however, the Court not only required the parties to rebrief the issues but, because the issue of avoiding a constitutional decision first arose in the context of Supreme Court review, it also forced the parties to argue their case at the appellate level.
Even without the benefit of briefing, it requires little research to determine that federal courts have extended a conditional common law right of access, prior to judgment, to evidentiary materials on which a court relies in issuing an order determining a litigant's substantive rights. See Matter of Continental Illinois Securities Litigation, 732 F.2d 1302, 1308 (7th Cir. 1984); (contemporaneous right of access to materials on which court relied in dismissing derivative claims against corporate directors); Brown Williamson Tobacco Corp. v. FTC, 710 F.2d 1165 (6th Cir. 1983), cert. denied, ___ U.S. ___, 104 S.Ct. 1595, 80 L.Ed.2d 127 (1984) (right of access to trial record in civil proceeding contesting agency action); In re Coordinated Pretrial Proceedings in Petroleum Products Antitrust Litigation, 101 F.R.D. 34, 43 (C.D.Cal. 1984) (right of access attaches as soon as documents are submitted to a court in connection with a motion); Zenith Radio Corp. v. Matsushita Electric Industrial Co., 529 F. Supp. 866, 899 (E.D.Pa. 1981) (presumptive right of access applies to documents once there is an evidentiary hearing). Nor would I ignore the views of several state courts that have clearly articulated a pre-judgment presumptive right of access to the records of civil trials. See State of Florida ex rel. Gore Newspapers Co. v. Tyson, 313 So.2d 777 (Fla.App. 1975); Charlottesville Newspapers v. Berry, 215 Va. 116, 206 S.E.2d 267 (1974); Des Moines Register Tribune Co. v. Hildreth, 181 N.W.2d 216 (Iowa 1970); Times-Call Publishing Co. v. Wingfield, 159 Colo. 172, 410 P.2d 511, 512 (1966) (dicta).
Nothing in this opinion should be taken to suggest that there is an absolute right of access to disputed evidentiary documents regardless of their content. On the contrary, my analysis assumes that at some point the proponent of a seal should have the opportunity to present a document-by-document justification of the need for confidentiality. Consequently, throughout this opinion I will only refer to a "conditional" or "presumptive" right of access.
The status of the common law right of access to discovery documents is less clear. The Seventh Circuit has found that a presumptive right of contemporaneous access attaches before trial, as long as the case is at an "adjudicative stage." Matter of Continental Illinois Securities Litigation, 732 F.2d 1302, 1314 (7th Cir. 1984). The Second Circuit has found a post-summary judgment right of access. Joy v. North, 692 F.2d 880, 893 (2d Cir. 1982). In both of these cases, however, the court made a final determination of private rights in the course of deciding a pre-trial motion. In our case, however, the court denied the summary judgment motion, essentially postponing a final determination of substantive legal rights. The public interest in obtaining contemporaneous access to such documents is not as pressing. Consequently, I fail to find a common law right of access to the documents to which the District Court may have referred in denying the summary judgment motions. My concurrence, however, should not be read to signify that I think there would be no federal common law right of access in a case where a pre-trial motion resulted in a final judgment.
The majority suggests that the conclusion of the Seventh Circuit is distinguishable because the court effectively dismissed some of the claims in the case at the time it granted access to the disputed documents. See maj.op. at 1336 n. 8. Assuming, arguendo, that such a dismissal were relevant, it should be noted that the Continental Illinois court only reviewed the effective dismissal of derivative claims against some of the defendants. 732 F.2d at 1306. The disputed evidentiary document, the Special Litigation Report, would continue to affect the rights of the remaining defendants. Nonetheless, the Seventh Circuit affirmed the pre-judgment disclosure of that Report.
The majority claims that this case is distinguishable on the ground that a government agency was a party. This is unpersuasive for two reasons. First, the court reversed an order that would have barred public access to all papers in any new civil action filed in that court until 21 days had elapsed from the date of such filing. Although the court also restricted access to hearings concerning a challenge by the Board of Supervisors to a special grand jury, the order sealing court records was not, by its terms, restricted to that case. Second, the common law right of access vindicates the public's interest in receiving information about the operations of the judicial branch of government. The public interest in access is implicated whether or not another government agency is involved in the case.
The majority attempts to distinguish Wingfield on the ground that the court in that case merely construed the scope of a state statute. The statute in question provided that only "parties in interest or their attorneys shall have the right to examine pleadings or other papers filed in any cause pending in such court." 410 P.2d at 512, citing C.R.S. 1963, 35-1-1. In holding that the statute did not require sealing of the pleadings in a civil suit contesting a school bond election, the court apparently relied on its understanding of the American common law presumptions in 1885 when the statute was enacted. Because the court believed that the general understanding in 1885 in Colorado would have been to provide access to civil actions "involving matters of public interest" the court carved out an exception to the otherwise restrictive state statute. Id. Thus Wingfield expresses both modern and historical concerns for the presumptive openness of the records of civil procedures.
Admittedly, few courts have directly considered whether a conditional right of contemporaneous access applies when a trial court enters a provisional seal. Provisional seals, unlike permanent seals, merely delay final determination of the confidential status of the contested documents. In gauging the public interest in access to civil proceedings, however, the timing of such access simply cannot be ignored. Indeed, the Supreme Court has found that the question of timing can rise to constitutional magnitude. And at least one United States Court of Appeals has explicitly concluded that the right of access is presumptively a right of "contemporaneous" access. Continental Illinois Securities, supra, 732 F.2d at 1310. Consequently, it would seem that the common law right of access would require, at a minimum, that a party seeking a provisional seal of trial exhibits should have to make a document-by-document showing as to the need for such a seal.
Courts that have spoken approvingly of provisional seals have only done so in cases where the underlying litigation turned on the confidentiality of information that would be made public if no provisional seal were entered. Thus in Des Moines Register Tribune Co. v. Hildreth, 181 N.W.2d 216 (Iowa 1970), the appellate court granted a writ of mandamus, forcing the trial court to unseal the record of a pending civil case. The court, however, did suggest that a provisional seal might be acceptable. Id. at 221. But the reason the Hildreth court believed that such a provisional seal might defeat an application for a writ of mandamus was that the underlying litigation in that case concerned the confidentiality of certain government records. Consequently, it would make sense to allow for a provisional seal so as not to moot the substantive issue of the underlying litigation. No such consideration presents itself in this case. See also Publicker Industries, Inc. v. Cohen, 733 F.2d 1059, 1073 (3d Cir. 1984).
See text infra at 1352-53 (discussing the importance of contemporaneous access under the First Amendment).
Although the Supreme Court has held that the common law right of access to evidentiary exhibits in civil cases does not provide an absolute right to copy and broadcast such exhibits, Nixon v. Warner Communications, Inc., 435 U.S. 589, 608, 98 S.Ct. 1306, 1317, 55 L.Ed.2d 570 (1978), the Court never questioned the common law right to inspect such exhibits at the time of trial. See id. at 609. See also Continental Illinois Securities, supra note 7, 732 F.2d at 1309 n. 11.
Presumably, the majority seeks to preserve the flexibility of trial courts in cases involving massive documentation. This is a worthy concern. But the majority fails to explain why a District Court would be unduly burdened if it were to require the proponent of a provisional seal to come forward with a pre-trial document-by-document justification of the confidentiality of the contested material. Once such a pre-trial showing were entered, nothing would then prevent a District Court from postponing a final determination of the confidential status of the contested material by issuing a provisional seal. In the meantime, however, the proponent of confidentiality would have to substantiate its concerns.
In the usual case such a minimal requirement would vindicate the public's interest in contemporaneous access to trial exhibits in civil proceedings and allow the District Court to retain administrative flexibility. In particular, it would protect the public from abuse of provisional seals. Such abuse tends to occur whenever a party can obtain a provisional seal on the basis of a rather general affidavit describing the harm it fears from disclosure of its documents. By contrast, the majority's approach nullifies the public's common law right of contemporaneous access. It would allow a summary affidavit to justify provisional sealing of trial exhibits in civil proceedings and therefore would do nothing to prevent abuse of such orders.
Cf., e.g., Zenith Radio Corp. v. Matsushita Electric Industrial Co., 529 F. Supp. 866, 873-74 (E.D.Pa. 1981); 513 F. Supp. 1100 (E.D.Pa. 1981) (plaintiff designated 100,000 pages of discovery documents as "confidential" under an umbrella protective order). See also Marcus, Myth and Reality in Protective Order Litigation, 69 CORNELL L.REV. 1, 11 n. 51 (1983).
This case provides an object lesson in the dangers of such an approach. In this case the District Court did not require Mobil to make a document-by-document showing as to the need for a provisional seal. Instead it was content to rely on the June 12, 1981 affidavit which merely identified the harm that might befall Mobil from disclosure in general. The affidavit said nothing as to how the alleged harm might result from the disclosure of specified individual documents. Mobil consequently claimed thousands of pages of discovery documents to be "confidential." Once the trial ended and the intensity of public attention abated, Mobil conceded the non-confidentiality of all but a handful of its documents. In the end it was unable to sustain its claims of confidentiality for a single document. Because this course of action effectively subverted the public's common law right of contemporaneous access to trial exhibits, I would reverse the District Court's decision to extend the terms of its umbrella protective order to the trial itself.
I am not unaware of the practical difficulties that faced that District Court in this case, given appellants' mid-trial intervention. But for the reasons stated, I do not think this justified the July 20, 1982 order. My analysis leads me to conclude that the District Court ought not to have allowed the conditions for appellants' midtrial intervention to have come to fruition, i.e., it ought to have denied Mobil's July 5, 1982 motion seeking to extend the terms of the November 5, 1981 protective order to the trial. Thus the rule I suggest today would not force disruption of ongoing trials. Rather, its general effect would be to require District Court judges to deny eve-of-trial motions to seal trial exhibits absent a document-by-document justification while providing an incentive to proponents of provisional seals to provide such justifications before the trial began.
The foregoing analysis leads me to the conclusion that, under the facts of this case, the District Court should have granted the Reporters Committee's request for immediate access to the designated trial exhibits. My conclusion is based on two predicates. First, Mobil had effectively waived its right to demand a document-by-document analysis of the confidentiality of the disputed trial exhibits. Second, the District Court's determination that a document-by-document analysis would disrupt the trial justified an order summarily granting access to such documents.
Nothing in my reasoning suggests that the federal common law right of access necessarily bars parties from presenting a document-by-document justification in the midst of trial. I only address the narrow case where (a) proponents of a provisional seal had the chance to make a document-by-document justification and instead presented a summary affidavit, and (b) the District Court determined that a mid-trial document-by-document analysis would be unduly disruptive.
As I have suggested, I believe the federal common law presumption of access imposes, at a minimum, an obligation on the proponent of a provisional seal to provide a document-by-document justification for such a seal. But federal common law does not provide an absolute right to provide such a justification. Whatever the right to provide such justification, circumstances may indicate that a party has waived its right in a particular case. In a case where this right has been waived there is no longer any reason to delay a ruling on the issue of access. I believe this to have been such a case.
In this case Mobil could and should have provided its document-by-document justification in its eve-of-trial motion. It chose not to. In so doing Mobil could have anticipated that any challenge to the provisional seal would arise at a time when a document-by-document review would disrupt the trial. In a typical case the prospect of such a disruption might decisively shift the balance of competing interests against an immediate ruling on the issue of confidentiality. If such a ruling were delayed, the proponent of the provisional seal would benefit from its own failure to present a document-by-document justification before the trial began. Such a result would encourage tactical abuse of provisional seals and vitiate the federal common law right of contemporaneous access. Only a policy of summarily sustaining the challenge to a provisional seal in such circumstances would effectively deter this abuse. Accordingly, I would hold that whenever a party has an opportunity to present its document-by-document justification before trial and fails to do so, that party has constructively waived its right to present such a justification should the issue be raised mid-trial.
Notwithstanding such a waiver of the right to present a document-by-document justification, the District Court of course retained discretion to require Mobil to provide such a justification. The District Court, however, determined that a document-by-document ruling in the midst of trial would be unduly disruptive. Given this conclusion, the appropriate course of action would have been to grant immediate access without such a document-by-document analysis.
In sum, having failed to provide a document-by-document justification of the need for confidentiality in its eve-of-trial motion, Mobil bore the risk that public intervention in the midst of trial might spur the District court to take summary action. Any rule to the contrary would effectively encourage proponents of provisional seals to manipulate the balance of competing interests to obtain a near-automatic delay of any ruling on the confidentiality of their documents. I am fearful that the majority's opinion invites precisely this sort of abusive behavior. Such a legal regime will unacceptably force the public to secure efficient trials only by sacrificing the openness of such proceedings. I therefore believe that the federal common law right of access required the District Court to grant access to the designated trial exhibits in this case.
It might be argued that it would be unfair to place such a burden on Mobil, given that the law of this circuit was not clarified at the time Mobil entered its eve-of-trial motion. If our holding today were anything other than a purely prospective statement of the law, such an argument would have merit. But given the prospective character of our ruling, Mobil's arguable lack of notice is irrelevant.
Indeed, the majority's critique of the practicality of the rule I suggest today, see maj.op. at 1340-41, can be traced to its failure to view the proponent of a provisional seal as being fully aware of its common law obligations. Thus the majority fails to appreciate the fact that Mobil had close to eight months to prepare its case justifying a provisional seal. Discovery of Mobil's documents had proceeded under the District Court's protective order since November 5, 1981. Nonetheless, the majority pretends that the first time Mobil could or should have thought of the need to provide a document-by-document justification of the need for a provisional seal was on the date on which it received a list of designated trial exhibits. Surely, a reasonable would-be proponent of a provisional seal, aware of its common law obligation to provide a document-by-document justification, could be expected to begin preparing its case well before the eve of trial.
I am well aware that the list of discovery documents was far longer than the list of designated trial exhibits. But at the very least, Mobil would have been alerted to the allegedly most sensitive documents and could have prepared a few brief paragraphs on each one (as that was all that it prepared in its ultimate justification presented to the District Court after completion of trial). Then, when the list of designated trial exhibits was issued, Mobil could have culled the relevant documents from the complete universe of discovery materials and filled the remaining gaps. In the unusual case where there was inadequate time for such last minute gap-filling, the District Court could easily provide for additional time. In any event, I fail to understand why the public should bear the burden of Mobil's failure to take the proper notes necessary to make its case when it was Mobil that sought the benefits of a provisional seal.
Thus, although majority's sympathetic portrait of Mobil's overburdened counsel is touching, the facts are to the contrary. Indeed, if one were to accept the majority's view one would have to believe that Mobil did not know why it was going through the exercise of requesting a provisional seal in the first place.
Thus, contrary to the majority's suggestion, see maj.op. at 1340, my analysis of the federal common law in this case leads me to argue in favor of granting the relief requested below regarding the designated trial exhibits. My concern to avoid the constitutional issues surrounding the right of access to designated trial exhibits follows as a direct result.
II. THE FIRST AMENDMENT AND THE PRESUMPTIVE RIGHT OF CONTEMPORANEOUS ACCESS TO TRIAL EXHIBITS
Notwithstanding appellants' colorable common law claims, the majority proceeds to declare that it finds no conditional First Amendment right of contemporaneous access to trial exhibits in civil proceedings. The majority's conclusion rests on two premises. First, the majority claims that before the First Amendment can provide the slightest protection to the public's interest in access to trial exhibits, such prejudgment access must be shown to have been firmly established in nineteenth century common law. Maj. op. at 1332. Second, the majority claims that, as a matter of historical fact, there was no common law right of access before a judgment issued. Maj. op. at 1332-34. Supreme Court precedent contradicts the first proposition. Careful reading of the common law undermines the second. Consequently, I find myself in agreement with those Courts of Appeals that have found that the First Amendment provides a presumptive right of contemporaneous access to materials admitted into evidence in a civil proceeding. See Continental Illinois Securities, supra; Brown Williamson, supra.
The majority implies that the doctrine of constitutional avoidance does not prevent it from reaching the issue of the right of access to designated trial exhibits because there is a genuine constitutional issue concerning access to pre-trial discovery documents. It has always been my understanding that the doctrine of constitutional avoidance cautions us to be quite discriminating in deciding whether to reach a particular constitutional issue. I therefore fail to understand why the majority paints with so broad a brush.
It does not appear that any other circuit has adopted the view expressed by the majority opinion. Belo Broadcasting Corp. v. Clark, 654 F.2d 423 (5th Cir. 1981), found no First Amendment right to copy or broadcast trial exhibits, but the court noted that the public had been afforded the right to inspect the exhibits in question. It argued that the Constitution is "fully satisfied by the kind of untrammeled access" afforded in that case. Id. at 427. Publicker Industries, supra note 7, 733 F.2d at 1073, includes dicta approving of a provisional seal. But as in Hildreth, see note 7 supra, the underlying litigation concerned the confidentiality of information that would be made available at the closed hearing. Under these circumstances the Third Circuit indicated that a provisional seal might be warranted so as not to moot the underlying litigation.
A. The Scope of the First Amendment Right of Access to Trial Exhibits
1. The role of historical practice in delineating the right of access. The majority asserts that before it would find that the First Amendment extends a pre-judgment right of access to documents in a civil proceeding it would have to be assured that such a right of access was clearly and specifically established by "historical" common law precedent. Maj. op. at 1332. As I read the Supreme Court's discussion of the public's right of access under the First Amendment, I find no such requirement.
The Court's most recent consideration of the degree to which the First Amendment affects the open character of judicial proceedings, Seattle Times Co. v. Rhinehart, 467 U.S. 20, 104 S.Ct. 2199, 81 L.Ed.2d 17 (1984), devoted precisely one sentence to the historical status of pre-trial proceedings, id. 104 S.Ct. at 2207-08. Far from declaring the lack of historical support for a right to disseminate pre-trial discovery documents to be dispositive, the Court merely weighed that consideration along with many other factors. The statutory source of the right to conduct discovery, the general custom governing publicity at pre-trial proceedings, the irrelevance of many discovery documents to the underlying cause of action, and the limited effect of pre-trial protective orders on an individual's capacity for self-expression were all weighed in the balance. Id. at 2207-08 n. 19. Had all other indicia been to the contrary, it is far from clear that the Court would have found historical practice, standing alone, to be conclusive. In light of such an eclectic approach, it is not surprising that a panel of the Second Circuit recently concluded that a majority of the Justices have come to rely on functional arguments in determining the scope of the right of access. See Application of the Herald Co., 734 F.2d 93, 98 (2d Cir. 1984).
There can be no question but that the Court has indicated that historical practice can shed light on the general contours of the right of access and that history can illuminate the underlying policy concerns surrounding the right. See Press-Enterprise Co. v. Superior Court, 464 U.S. 501, 104 S.Ct. 819, 822, 78 L.Ed.2d 629 (1984) (terming the historical data "helpful"). See also Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 605, 102 S.Ct. 2613, 2619, 73 L.Ed.2d 248 (1982); Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 564-69, 100 S.Ct. 2814, 2820-23, 65 L.Ed.2d 973 (1980) (plurality opinion). Nonetheless, I cannot agree with the majority's assertion that the Court has fossilized the First Amendment. Nowhere does the Court equate the scope of its protections with those precise practices that can be shown to have been firmly established at common law. Instead, the Court appears to have weighed historical practice as one factor among many in evaluating the appropriate scope of the right of access.
In the context of this case, the majority would find the general historical presumption of the open character of trial exhibits in civil proceedings to be irrelevant. Only a specific tradition of pre-judgment access would suffice. By contrast, I would find such a general historical tradition illuminating. Unless the common law were plainly established to the contrary, I would weigh that tradition, along with other factors, in determining whether the First Amendment provides a presumptive right of access to trial exhibits in civil proceedings before a judgment has been entered.
2. The general common law presumption of access. As the majority implicitly concedes, there is a longstanding common law tradition assuring a presumption of access to trial exhibits in civil proceedings. See Richmond Newspapers, supra, 448 U.S. at 580 n. 17, 100 S.Ct. at 2829 n. 17 ("[H]istorically both civil and criminal trials have been presumptively open."); Gannett Co. v. DePasquale, 443 U.S. 368, 386 n. 15, 99 S.Ct. 2898, 2908 n. 15, 61 L.Ed.2d 608 (1979) (same); Nixon v. Warner Communications, Inc., 435 U.S. 589, 597-98, 98 S.Ct. 1306, 1311-12, 55 L.Ed.2d 570 (1978) (there is a long-standing conditional common law right of access to trial exhibits). The underlying rationale for this presumption of openness was succinctly stated by the then Judge Holmes: "[I]t is of the highest moment that those who administer justice should always act under the sense of public responsibility, and that every citizen should be able to satisfy himself with his own eyes as to the mode in which a public duty is performed." Cowley v. Pulsifer, 137 Mass. 392, 394 (1884).
The majority contends, however, that the historical common law right of access was not implicated prejudgment. Maj. op. at 1335-36. My own reading of the cases cited by the majority fails to persuade me of this proposition. Instead, these cases suggest that the right of access to trial exhibits attached once such exhibits were introduced at trial and thus became, in the words of Chief Justice Burger, "part of the [presumptively open] trial." Gannett Co., supra, 443 U.S. at 396, 99 S.Ct. at 2914 (Burger, C.J., concurring).
The majority cites only two "historical" cases directly addressing the right of access. The earlier of the two, Schmedding v. May, 85 Mich. 1, 48 N.W. 201 (1891), concerned the attempt of a reporter to obtain access to pre-trial pleadings. The court framed the issue as follows:
The question is therefore fairly presented, have parties the right to an examination of the records and papers in a cause for the purpose of publishing statements in regard thereto in the newspapers, before trial or hearing, or before they become public by proceedings taken in open court? * * *
Id. 48 N.W. at 202 (emphasis added). The Michigan court's holding was framed in similarly narrow terms. "[O]ur conclusion is that the parties to suits in court may, under the direction of the court, lawfully withhold the records and papers in the case * * * until they are made public * * * by proceedings in open court." Id. 48 N.W. at 203. At no point does the court suggest that the trial court has plenary discretion to seal trial exhibits until judgment issues.
Nor can the phrase "made public by proceedings taken in open court" refer only to those documents actually read aloud in open court. The Schmedding court indicated that this was not its view when it stated its understanding of the rationale underlying the right of access: the need to assure that everyone be able "to satisfy himself with his own eyes as to the mode in which a public duty is performed." Schmedding v. May, 85 Mich. 1, 48 N.W. 201, 202 (1891), quoting Cowley v. Pulsifer, 137 Mass. 392, 394 (1884). If the purpose of the right of access was to assure that the public understood the basis of judicial decisionmaking, it would make little sense to enable the public to obtain the information it needed only in those case where the crucial evidence was given orally rather than in the form of exhibits.
It is true that the Schmedding court also articulated a rationale for a limit on the right of access: the need to protect private parties from scurrilous charges contained in pre-trial pleadings. But the court apparently saw the principal danger as inhering in those pleadings that although filed "may never come to trial or hearing." Thus the court sought to protect litigants' privacy by preventing pre-trial disclosure.
The Schmedding court sought to balance the public's interest in access and the individual's interest in privacy by limiting the right of access to the trial stage of the litigation. As the majority suggests, a complete vindication of the private interest in reputation might have required the court to deny the right of access pre-judgment. But unlike the majority here, the Schmedding court sought to achieve a measure of protection for both the public and the private interest; it did not seek to vindicate the individual's privacy interest at the cost of eviscerating the public interest in self-governance. Consequently, it drew the line at the onset of the trial.
The majority also cities Burton v. Reynolds, 110 Mich. 354, 68 N.W. 217 (1896). In that case the Michigan court again refused to require a trial court to allow a non-litigant to have access to the pre-trial pleadings. Not only does this case merely purport to follow the rule established in Schmedding, but it reiterated the view that trial courts had unfettered discretion to refuse pre-trial requests for access, saying: "[I]t is not the absolute right of persons to make merchandise of the contents and allegations contained in the records of private actions and suits, before trial, for gain." Id. (emphasis added).
The majority also cites Ex parte Drawbaugh, 2 App.D.C. 404 (1894). In dicta the court distinguished the right of access to papers that had been "merely filed" and those as to which the court had taken some action. Id. at 407. The court also distinguished between post-judgment papers and pre-judgment and pre-trial papers. Id. at 406. Consequently, this case cannot provide support, even in dicta, for the majority's alleged "post-judgment rule."
The majority also alludes to a line of cases concerning the scope of the reporter's privilege in libel cases to report judicial proceedings without fear of liability. Maj. op. at 1335. Again, neither of the "historical" cases cited by the majority provide support for a "post-judgment" rule. In Cowley v. Pulsifer, supra note 17, the court refused to extend the privilege when a newspaper published the contents of pre-trial pleadings. The court found that there was no privilege as to such a "preliminary written statement of a claim or charge" because "[k]nowledge of them throws no light on the administration of justice." 137 Mass. at 394. Although Holmes noted that allowing access to the civil trial itself might present some peril to individual privacy and reputation, he concluded that "[f]or the purposes of the present case, it is enough to mark the plain distinction between what takes place in open court, and that which is done out of court by one party alone, or more exactly * * * the contents of a paper filed by him in the clerk's office." Id. at 395.
Park v. The Detroit Free Press, 72 Mich. 560, 40 N.W. 731 (1888), also cited by the majority, see maj. op. at 1333, was another libel suit for publication of a pre-trial filing. In holding that such a publication was not privileged, the court again drew a distinction between pre-trial and trial papers:
If pleadings and other documents can be published to the world by anyone who gets access to them, no more effectual way of doing malicious mischief with impunity could be devised than filing papers containing false and scurrilous charges, and getting those printed as news. The public have no rights to any information on private suits till they come up for public hearing or action in open court[.] * * *
Id. 40 N.W. at 734 (emphasis added).
Finally, it should be noted that recent scholarly treatment of the "historical" cases analyzed by the majority finds that such cases established a pre-trial, not a "post-judgment," rule. See Marcus, supra note 10, 69 CORNELL L.REV. at 33 n. 136.
Unable to muster support from "historical" cases, the majority seeks support in more recent precedent. But only one of the mid-century cases cited by the majority contains dicta suggesting adherence to a "post-judgment rule," State ex rel. Williston Herald, Inc. v. O'Connell, 151 N.W.2d 758 (N.D. 1967). O'Connell, however, concerned the right of access in a criminal case. Its view of the pre-judgment right of access was limited accordingly: "The records of criminal trials conducted in respondent's court are public only after such proceedings are completed and entered in the docket of the court." Id. at 763. Thus it is far from clear that O'Connell endorsed a common law "rule" concerning pre-judgment access in civil cases, a rule that the majority claims was grounded in the need to protect private reputations from private calumny. Rather, the court may simply have been moved by the need to provide a fair trial to a criminal defendant, free of prejudicial pre-judgment publicity.
The O'Connell court did assert that "[w]e believe that it is the right of the public to inspect the records of judicial proceedings after such proceedings are completed and entered in the docket of the court." 151 N.W.2d at 763. Admittedly, this dicta is not restricted to criminal proceedings. But neither does this dicta indicate that there is no pre-judgment right of access pre-judgment. It only asserts that there is a conditional right of access post-judgment. Indeed, the issue presented in O'Connell was whether a newspaper could gain access to post-judgment criminal records and thereby avoid the burden of having a reporter present in the court-room during the trial. Id. at 760. Not only did the court not have occasion to consider the right of access at the time of trial, it may have assumed that such a right did obtain at that time. Thus, the court argued, "The trials in respondent's court are open to the public, and the petitioner's reporters, or anyone else, may attend such trials. We are faced here not with the right to know, but with the very limited question of the right of the petitioner to inspect the criminal records of the court after the trial has been concluded." Id. at 762. Although this language does not expressly indicate that the record was open at the time of trial, a prior statement of the court suggests that this may have been its meaning: "[Petitioner] desires to inspect the criminal records of the county court so that this information can be obtained without having its reporter present throughout the trial of each case. This proceeding involves, therefore, not so much the right of the petitioner to secure the information it seeks as it does the method of getting such information." Id. at 760. Thus the court appears to have assumed that the press could obtain all of the information it desired on the basis of its right of contemporaneous access to the "trial."
Although two of the mid-century cases cited by the majority suggest that access could be unconditionally barred before the end of the "trial" (but not before "judgment"), most of the remaining mid-century cases that the majority claims refer to a "post-judgment" rule actually recite a "pre-trial" rule. And as I have noted, the trend in the cases over the last 15 years has been to declare a presumptive right of access to the records of civil proceedings.
Of the cases cited in support of the "post-judgment rule" only one, C. v. C., 320 A.2d 717 (Del. 1974), concerned a prejudgment request for access to records that might contain trial exhibits in a civil case. In C. v. C. a reporter sought access to a sealed file in a divorce case after the trial had been completed but before judgment had issued. The court noted, in dicta, that there was a general right to access "after completion of the proceedings," but then went on to explain why that right was not absolute. Id. at 724. What is remarkable about the court's discussion, given that C. v. C. was a pre-judgment case, is that it nowhere referred to the alleged "post-judgment rule." Moreover, at no point does the court suggest that there was no right to access before completion of the proceedings; it simply referred to the state of the law governing access to trial records after completion of the proceedings because it was confronted with a post-trial pre-judgment case.
Similarly in Birnbaum v. Wilcox-Gay Corp., 17 F.R.D. 133, 139 (N.D.Ill. 1953), the court stated, in dicta, that "[t]he records of proceedings of a court are not required to be open to public inspection until after the trial." (Emphasis added.) Although the majority is meticulous in insisting that recent cases that do not expressly provide for pre-judgment access to trial exhibits cannot be read to imply such a right, maj. op. at 1336 n. 8, the majority oddly believes that the "post-trial" language of Birnbaum and C. v. C. must be liberally construed to refer to a "post-judgment" rule. Even assuming the majority's reading, however, Birnbaum is readily distinguishable from our case. Once again, the litigation concerned the trial court's power to seal pre-trial pleadings and not trial exhibits. Indeed, the Birnbaum court only sealed the pleadings on a provisional basis so that it might have the opportunity to determine whether they were frivolous. Thus at most Birnbaum stands for the proposition that a court may enter a provisional seal before the trial has begun to prevent public access to potentially frivolous complaints. It says nothing about access to exhibits in causes that actually come to trial.
Thus in Werfel v. Fitzgerald, 23 A.D.2d 306, 260 N.Y.S.2d 791, 794-95 (1965), the passage cited by the majority reads: "[Although some courts have provided for pre-trial access,] [o]n the other hand, other American courts held that records prior to trial were not available to public inspection until there were proceedings in open court, following which the records were accessible to all persons, whether they had a special interest or not[.]" (Citation omitted; emphasis added.)
Similarly, In re Sackett, 136 F.2d 248, 249 (Ct.Cus. Pat.App. 1943), stated, in the passage cited by the majority, that in addition to distinguishing between official records in general and judicial records in particular, "[d]istinction has also been made in some cases between the right to inspect judicial records before trial and the right to inspect the record of the court after trial." (Emphasis added.) Although the court does not appear to have focused on the question of mid-trial access, it is also clear that it did not understand the common law to have enunciated a "post-judgment" rule.
Finally, the more recent libel cases cited by the majority simply do not refer to a post-judgment rule. On the contrary, Sanford v. Boston Herald Traveler Corp., 318 Mass. 156, 61 N.E.2d 5, 6 (1945), states that "the doctrine long established in this Commonwealth is that the right to report proceedings in the court does not extend to reporting accusations contained in papers filed by a party and not yet brought before a judge or magistrate for official action." (Emphasis added.) And in Newell v. Field Enterprises, Inc., 91 Ill.App.3d 735, 47 Ill.Dec. 429, 415 N.E.2d 434 (1980), the court does not assert that the contemporary majority rule is a post-judgment rule. But see maj. op. at 1336 n. 8. Rather it states that the majority view of the judicial records privilege is that it does not attach until "some judicial action" has been taken. The language of the case does not indicate whether the court thought that "judicial action" referred only to final judgments or whether it might refer to evidentiary rulings or preliminary motions. When read against the facts of that case, however, the meaning of the Newell court is clear. Newell concerned the issue of whether the privilege attached to publication of pre-trial pleadings. Thus in evaluating conflicting precedent the Newell court was simply seeking to determine whether the privilege applied pre-trial (as the emerging trend of cases would have it) or whether it did not cover such pleadings (as was the case under the older cases). The question of restricting the privilege to post-judgment publications simply did not present itself.
See text supra at 1342-43.
Not surprisingly, the majority concedes that there is "sparse" authority, whether "historical" or otherwise, supporting the "post-judgment rule." Maj. op. at 1335. Indeed, there is almost none. Instead, a review of common law precedent suggests a presumptive right of contemporaneous access to the records of civil proceedings. To the degree that the common law limited the time at which such access might occur it appears to have shown a preference for access at the time the trial began, not at the time judgment issued. Given the functional arguments supporting the right of access, identified by Holmes over a hundred years ago, it is difficult to imagine how the line could have been drawn anywhere else.
2. The functional basis of a right of access to trial exhibits in civil cases. Open trials serve several important functions. They enhance the quality of factfinding. Cf. Globe Newspaper, supra, 457 U.S. at 606, 102 S.Ct. at 2619 (criminal trial). They assure the appearance of fairness. Cf. Press-Enterprise, supra, 104 S.Ct. at 823 (criminal trial). And they play a cathartic role in permitting the community to observe justice being done. Cf. Richmond Newspapers, supra, 448 U.S. at 570-572, 100 S.Ct. at 2823-25 (plurality opinion) (criminal trial). But most important, open trials are essential to the legitimacy of judicial proceedings. As the Chief Justice observed in Richmond Newspapers:
People in an open society do not demand infallibility from their institutions, but it is difficult for them to accept what they are prohibited from observing. When a criminal trial is conducted in the open, there is at least an opportunity both for understanding of the system in general and its workings in a particular case[.]
Although the Supreme Court has not yet expressly applied this logic to civil proceedings, several United States Courts of Appeals have found that the same functional concerns that argue in favor of open criminal proceedings also argue in favor of open civil proceedings in general, Publicker Industries, Inc. v. Cohen, 733 F.2d 1059, 1070 (3d Cir. 1984), and to evidentiary exhibits in particular. Continental Illinois Securities, supra, 732 F.2d at 1308-09; Brown Williamson, supra, 710 F.2d at 1179.
Indeed, at least one circuit has suggested that these functional concerns are more compelling in civil cases because the criminal defendant's right to a fair trial, free of prejudicial publicity, is not implicated. Wilson v. American Motors Corp., 759 F.2d 1568, 1571 (11 Cir. 1985). And although some might argue that because the state is always a party to criminal cases there is a stronger presumption of publicity surrounding such proceedings, the Supreme Court has found such a distinction to be of dubious value. See Gannett Co. v. DePasquale, 443 U.S. 368, 386-87 n. 15, 99 S.Ct. 2898, 2908-09 n. 15, 61 L.Ed.2d 608 (1979) ("Indeed, many of the advantages of public criminal trials are equally applicable in the civil trial context. While the operation of the judicial process in civil cases is often of interest only to the parties in the litigation, this is not always the case. * * * Thus, in some civil cases the public interest in access, and the salutary effect of publicity, may be as strong as, or stronger than, in most criminal cases.") Only one Supreme Court Justice has suggested that the First Amendment right of access to trial proceedings does not apply to civil cases. See Globe Newspaper Co. v. Superior Court, 457 U.S. 596, 611, 102 S.Ct. 2613, 2622, 73 L.Ed.2d 248 (1982) (O'Connor, J., concurring in judgment).
I agree. Civil adjudication, no less than criminal trials, must maintain its public legitimacy. And to the degree that such legitimacy turns on public access to the grounds of judicial decisionmaking, it makes little sense to distinguish oral and written testimony. Either may provide the basis for a final determination of individual legal rights. As the Sixth Circuit observed in Brown Williamson: "These principles apply as well to the determination of whether to permit access to information contained in court documents because court records often provide important, sometimes the only, bases or explanations for a court's decisions." 710 F.2d at 1177. Consequently, I find little reason not to extend the presumptive right of access to trial exhibits in civil proceedings.
The majority, however, suggests three reasons why the records of civil proceedings ought not be presumptively open until a judgment has issued.
First, the majority speculates that prejudgment access is rarely requested. Maj. op. at 1337. Whether or not this is true as an empirical matter, it is simply irrelevant. It may be that the public often has little interest in private suits. But the functional concern for judicial legitimacy does not depend on ongoing, daily public monitoring of the court system. Rather it depends on the public's assurance that when an important case arises the public will have a presumptive right of access to the bases of judicial decisionmaking at the time when that case is newsworthy.
Second, the majority argues that denial of contemporaneous access to trial exhibits "does not provoke the kind of outcry which closing of a trial usually excites." Id. The volume of public outcry does not provide the measure of constitutional guarantee. This is particularly true where the value of that guarantee can only be appreciated after the documents in question have been publicized in a timely manner. Indeed, provisional seals, preventing timely release of critical materials, are probably the surest method for minimizing public concern over closed proceedings.
Finally, and perhaps more seriously, the majority argues that, unlike live proceedings, the public can learn of the subleties of trial exhibits through a post-judgment reading as readily as by a pre-judgment reading. Id. The majority, however, has completely misapprehended why courts have found that "the presumption of access normally involves a right of contemporaneous access[.]" Continental Illinois Securities, supra, 732 F.2d at 1310 (emphasis in original). As the Supreme Court observed over 40 years ago in Bridges v. California, 314 U.S. 252, 268-69, 62 S.Ct. 190, 196-97, 86 L.Ed. 192 (1941):
It must be recognized that public interest is much more likely to be kindled by a controversial event of the day than by a generalization, however penetrating, of the historian or scientist. Since they punish utterances made during the pendency of a case, the judgments below therefore produce their restrictive results at the precise time when public interest in the matters discussed would naturally be at its height. * * *
No suggestion can be found in the Constitution that the freedom there guaranteed for speech and the press bears an inverse ratio to the timeliness and importance of the ideas seeking expression. Yet, it would follow as a practical result of the decisions below that anyone who might wish to give public expression to his views on a pending case involving no matter what problem of public interest, just at the time his audience would be most receptive, would be as effectively discouraged as if a deliberate statutory scheme of censorship had been adopted. * * *
This unfocussed threat is, to be sure, limited in time, terminating as it does upon final disposition of the case. But this does not change its censorial quality. * * * And to assume that each [temporary moratorium on publicity] would be short is to overlook the fact that the "pendency" of a case is frequently a matter of months or even years rather than days or weeks.
In short, the harm to the public interest does not lie in the possible distortions that might result from untimely release of important documents. Instead, the harm to the public interest derives from the effective suppression of important news through delay which, as the Court noted in Bridges, can be as censorial as an overt scheme of restraints.
The facts of this case provide an object lesson in the importance of assuring a contemporaneous presumptive right of access to trial exhibits. Mobil initially swept thousands of pages under the November 5, 1981 umbrella protective order. When put to its proof, Mobil only specifically contested the confidentiality of three of these documents (and only contested two of them in a consistent manner). When the trial judge examined Mobil's brief justification for its claim of confidentiality for these documents it found it insufficient to sustain a permanent seal.
Mobil fully understood that it would some day have to justify its claims of confidentiality. Indeed, the terms of the July 20, 1982 order made plain that the court would determine the confidentiality of the documents at issue shortly after completion of the trial. Mobil was presumably aware that many of its allegations would ultimately fail to pass muster. Nonetheless, Mobil indiscriminately sought to take maximum advantage of the terms of the provisional seal, obtaining the one sure benefit the July 20, 1982 order could provide: delay. Once the trial was over, and the glare of publicity had subsided, Mobil voluntarily waived its claims to confidentiality. Such conduct, if nothing else, belies the majority's assertion that little turns on the timing of the public's presumptive right of access.
In its historical discussion the majority hints at another policy reason for supporting a post-judgment rule: vindication of private interests in reputation and privacy. Presumably, by delaying access until a judgment has been issued the court will be able to prevent publication of unfounded allegations absent a legal determination reflecting on the weight of such data. Maj. op. at 1335. But even if such a view would provide grounds for limiting the common law privilege in libel cases, it is a wholly one-sided resolution of the issue of public access. When facing the question of access the courts have sought to balance individual privacy interests and the public interest in learning of the grounds of judicial decisionmaking. By contrast, under the majority's view the public interest would be swept aside in the hope of a more complete vindication of individual privacy interests. Such a justification sweeps too far. Under such a rationale public access to the trial itself ought to be restricted until a judgment has been rendered. Unwilling to face such a patently unacceptable result, the majority never fully articulates the grounds for its preference for a post-judgment rule and instead rests its conclusion on its questionable reading of the view of a Michigan court at the turn of the century. Maj. Op. at 1334 n. 7.
In the end, the majority's argument rests on its ipse digit that there be "an historic practice of such clarity, generality and duration as to justify the pronouncement of a constitutional rule preventing federal courts and the states from treating the records of private civil actions as private matters until trial or judgment." Maj. op. at 1336 (emphasis in original). For the reasons stated above, I find this requirement to be unacceptably rigid. Few constitutional rights that we now take for granted would survive this brittle test. In light of the general common law presumption of access to civil trial exhibits, at least once the trial had begun, and in light of the compelling policy considerations identified by the Supreme Court, I would find that the First Amendment provides a presumptive right of contemporaneous access to trial exhibits in civil proceedings.
B. The Applicable Standard
Having found that a presumptive right of contemporaneous access applies to trial exhibits in civil proceedings, it remains for us to determine what standard the First Amendment imposes on a court before it can provisionally seal such exhibits until the completion of the trial. The Supreme Court has most recently spoken as if closure orders must meet the test of strict scrutiny. See, e.g., Press-Enterprise, supra, 104 S.Ct. at 824; Globe Newspaper, supra, 457 U.S. at 606, 102 S.Ct. at 2619. But the Court has also sometimes suggested that less exacting standards might apply. See, e.g., Gannett Co., supra, 443 U.S. at 393, 99 S.Ct. at 2912; Richmond Newspapers, supra, 448 U.S. at 600, 100 S.Ct. at 2840 (Stewart, J., concurring). Indeed, the Second Circuit has gone so far as to suggest that "uncertainty marks the nature — and strictness — of the standard of closure" adopted by the Supreme Court regarding the right of access to criminal trials. Application of the Herald Co., supra, 734 F.2d at 99, quoting Richmond Newspapers, 448 U.S. at 603, 100 S.Ct. at 2841 (Blackmun, J., concurring in judgment).
Nor do the standards provided by the other Courts of Appeals that have considered this question yield a single standard. See, e.g., Continental Illinois Securities, supra note 3, 732 F.2d at 1313 ( ad hoc balancing); Brown Williamson Tobacco Corp. v. FTC, 710 F.2d 1165, 1179 (6th Cir. 1983) (assuming the right to access unless the seal could be shown to be (a) a mere time, place, and manner restriction, or (b) required because of the sensitive nature of particular documents). The Supreme Court has also suggested that time, place, and manner restrictions would include those rules necessary to maintain order or decorum in the courtroom. See, e.g., Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 581 n. 18, 100 S.Ct. 2814, 2830 n. 18, 65 L.Ed.2d 973 (1980) (plurality opinion). Because delay in access can result in such serious First Amendment harm, I do not find that such a seal is a mere time, place, and manner restriction, to be sustained if it is merely "reasonable" under the circumstances. See Richmond Newspapers, 448 U.S. at 600, 100 S.Ct. at 2840 (Stewart, J., concurring).
But even if the most recent Supreme Court pronouncements on the subject can be said to establish a rule of strict scrutiny, I would hesitate to apply that standard to issuance of a provisional seal. Although the strict scrutiny standard applied to closure of criminal trials might appropriately be extended to permanent closure of civil trial records, common sense suggests that a trial court ought to retain the flexibility to issue provisional seals without meeting such an exacting test.
As noted above, in choosing to enter a provisional seal the trial court infringes on the public's First Amendment rights by imposing a delay on access. Although a delay plainly restricts First Amendment rights, it need not be understood as a direct restriction on the right of access. In many cases the District Court will provisionally seal certain designated trial exhibits without a document-by-document determination because it is concerned about the administrative burden caused by the need to sift through large quantities of documents stamped "confidential." Neither the content of the contested material nor the consequences of publicity need animate the decision to issue a provisional seal. Under such circumstances, where the government has burdened First Amendment rights without regard to content and without seeking to stifle expression, courts have declined to employ strict scrutiny. Cf. United States v. O'Brien, 391 U.S. 367, 377, 88 S.Ct. 1673, 1679, 20 L.Ed.2d 672 (1968).
On the other hand, the merely provisional character of the seal does not, as the majority suggests, prevent such a seal from running afoul of the First Amendment. Indeed, provisional seals pose a particular danger to the public's First Amendment interests where they are subject to abuse because they have been entered on the basis of an affidavit that merely identifies the general harms that might flow from disclosure without suggesting why disclosure of particular documents will produce that harm.
See note 10 supra and accompanying text.
Consequently, I am inclined to require only that the trial court establish (a) that the provisional seal was justified by a substantial government interest, and (b) that there was no less restrictive means of achieving that interest. Cf. Application of the Herald Co., supra, 734 F.2d at 100-01 (applying a similar intermediate standard in the context of reviewing closure of a suppression hearing); Publicker Industries, supra, 733 F.2d at 1070 (applying similar standard in the context of reviewing closure of a preliminary injunction hearing and sealing of the record of that hearing). Moreover, in evaluating the government's interest I would require that it be unrelated both to the content of the material subject to the provisional seal and to the consequences of disclosure. Cf. generally O'Brien, supra, 391 U.S. at 377, 88 S.Ct. at 1679 (applying this standard to "incidental" infringements on First Amendment rights). Requiring a showing of a mere "substantial" government interest would enable the District Court to cite the need to avoid undue delay in the underlying substantive litigation as a good and sufficient reason for entering a provisional seal. But requiring the court to consider less restrictive means of attaining such legitimate ends will protect the public from the talismanic invocation of the need for efficient litigation as a ground for delaying public access.
C. The Standard Applied
There is no doubt that the interest relied upon by the District Court in its order of July 20, 1982 constituted an adequate government interest. The court argued that it sought to postpone a final determination of the confidential status of the disputed documents so that it might concentrate its efforts on the substance of the underlying libel litigation. Order of July 20, 1982 at 4, Appellants' Record Excerpts at 163. When the District Court entered this order it was faced with several unpalatable choices. It could have (1) continued the terms of its protective order, (2) stopped the trial until it had determined the confidentiality of all of those designated trial exhibits for which Mobil still claimed confidentiality, or (3) issued collateral orders on the confidential status of various documents as they were admitted into evidence. It is hardly surprising that the court chose the first option.
The District Court, however, could have avoided this mid-trial dilemma by denying Mobil's motion of July 5, 1982. Had the court required Mobil to provide the sort of document-by-document justification for the continued confidentiality of those documents that the parties had designated as trial exhibits it would never have had to face such an unfortunate set of mid-trial choices. I am mindful that parties tend to designate many more documents as exhibits than are actually introduced as evidence at trial. But the document-by-document justification eventually offered by Mobil was hardly voluminous. The record reveals that Mobil only provided a paragraph or two for each of the documents whose confidentiality it ultimately contested. A comparison of Mobil's affidavit of June 12, 1981 supporting the initial protective order and the affidavit of May 31, 1983 suggests that it would not have been that difficult for the District Court to make a pre-trial determination of the confidentiality of Mobil's documents. For example, the District Court appears to have been able to dispose of at least one of Mobil's claims of confidentiality merely by noting that the document in question was too old to contain competitively sensitive information. Memorandum of June 21, 1983 at 4, Appellants' Record Excerpts at 183.
Thus my First Amendment analysis, like my analysis of appellants' federal common law rights, see note 11 supra, would not require disruption of ongoing trials. Rather the District Court would simply deny motions to extend provisional seals to trial exhibits absent a document-by-document justification. It is true that, under the particular circumstances of this case, Mobil's "pre-trial" motion was filed on the eve of trial. Consequently, it might be argued that there was no real opportunity for the court to require Mobil to file a document-by-document justification without disrupting the trial. Such an objection, however, passes over a critical fact: there is no reason why the District Court ought not to have made Mobil bear the burden of its eve-of-trial motion. Had the District Court determined that a full-blown analysis of the contested documents was impractical at this late date, it need merely have denied Mobil's motion and unsealed the record. As with the public's common law interest in access, there is no reason why the public's First Amendment interests should suffer because Mobil only acted at the eleventh hour to extend the terms of the original protective order.
Of course, it is wholly possible that Mobil would have contested a larger number of documents if it were faced with the prospect of mid-trial disclosure. It may also have expended greater resources in arguing for the confidentiality of particular documents. In that instance, I would defer to the judgment of the District Court in postponing a final determination until after completion of the trial. But by forcing Mobil to make some sort of document-specific showing before trial the District Court would have reduced the potential for abuse of its umbrella order of July 8, 1982. Such an order would have served the public's First Amendment interest in access to those materials as to which there was not even a colorable claim of confidentiality. Put simply, it would have been a "less restrictive alternative."
To the degree that the public interest is strongest as to those documents actually offered in evidence, Mobil would have to bear the burden of justifying the confidentiality or documents as to which there was little or no First Amendment value. The alternative, however, would be to require the District Court to engage in disruptive mid-trial motions. Consequently, the judicial interest in efficient litigation cut against the interests of the party seeking to seal the evidentiary exhibits in a civil hearing, forcing them to justify the confidentiality of all designated trial exhibits and not merely admitted or introduced trial exhibits.
Thus, although I would not require a District Court to make a document-by-document determination in every case before issuing a provisional seal, I believe the First Amendment minimally requires that the proponents of a provisional seal provide a pre-trial document-by-document justification for the continued confidentiality of designated trial exhibits. A provisional sealing order entered in the absence of such a showing would constitute a failure to consider reasonable alternatives. Consequently, I believe that the District Court's order of July 8, 1982 violated the First Amendment. Insofar as the order of July 20 simply sustained the July 8 order, it too ran afoul of the public's right of access.
I respectfully dissent from affirmance of the provisional sealing of designated trial exhibits.