[47] ESCHBACH, Circuit Judge, concurring in part and dissenting in part.

For the reasons expressed in the majority opinion, I agree that the discovery order constituted an abuse of discretion. I therefore concur in the judgment in No. 81-2671. Because I cannot agree that res judicata bars the federal antitrust claims, however, I respectfully dissent in No. 83-8046.

Judge Cudahy recites, as Justice Stewart did before him, an impressive array of authorities contrary to the majority's holding that res judicata bars the plaintiffs' case. See, e.g. Hayes v. Solomon, 597 F.2d 958, 984 n. 19 (5th Cir. 1979), cert. denied, 444 U.S. 1078, 100 S.Ct. 1028, 62 L.Ed.2d 761 (1980); Kurek v. Pleasure Driveway and Park District of Peoria, 583 F.2d 378, 379 (7th Cir. 1978), cert. denied, 439 U.S. 1090, 99 S.Ct. 873, 59 L.Ed.2d 57 (1979); Restatement (Second) of Judgments § 26(1)(c) comment c, illustration 2 (1982). Moreover, the majority's primary source of support, Nash County Board of Education v. Biltmore Co., 640 F.2d 484 (4th Cir. 1981), is clearly distinguishable in that it involved an attempt to relitigate an antitrust claim. Notwithstanding the force of Judge Posner's opinion, therefore, I would adhere to the precedents and affirm the order in No. 83-8046.

[50] FLAUM, Circuit Judge, concurring in part and dissenting in part.

I agree with Judge Posner that this suit should be dismissed under the doctrine of res judicata, but I arrive at this position through a different analysis. Judge Posner fits his result into the existing strictures of res judicata doctrine by finding that the plaintiffs could have brought their entire cause of action in one court. I would hold instead that in finding this suit barred by res judicata, it is unnecessary to find that the plaintiffs could have brought all their claims in one court, as long as they could have brought all their claims contemporaneously in different courts. We should treat this matter as we would any other case in which the plaintiff litigates a dispute under one theory, loses, and then brings a second suit to litigate the same dispute under a different theory. In such a situation, absent equitable circumstances not present here, fairness to a defendant requires that the second suit be barred by res judicata.

The basic principles of res judicata were stated recently by this court in the case of Mandarino v. Pollard, 718 F.2d 845 (7th Cir. 1983):

"The doctrine of res judicata is that a final judgment on the merits in a court of competent jurisdiction bars the same parties or their privies from relitigating not only the issues which were in fact raised and decided but also all other issues which could have been raised in the prior action. . . . The essential elements of the doctrine are generally stated to be: (1) a final judgment on the merits in an earlier action; (2) an identity of the cause of action in both the earlier and the later suit; and (3) an identity of parties or their privies in the two suits."

At 849 (quoting Lee v. City of Peoria, 685 F.2d 196, 199 (7th Cir. 1982). In the present case, the res judicata controversy revolves around whether there is an identity of the cause of action and whether the issues in the present suit should and could have been raised earlier.

As a threshold matter, there is a question about which body of res judicata law should be applied here. As Judge Cudahy points out, there is language in the Supreme Court case of Kremer v. Chemical Construction Corp., 456 U.S. 461, 102 S.Ct. 1883, 72 L.Ed.2d 262 (1982), that suggests that when a federal court determines the preclusive effect of an earlier state judgment, it always must follow the res judicata rules of the state from which the judgment is taken. ("It has long been established that § 1738 does not allow federal courts to employ their own rules of res judicata in determining the effect of state judgments." Id. at 481-82, 102 S.Ct. at 1897-98.) However, the Supreme Court in Kremer did not hold that a federal court may not give a state court judgment greater preclusive effect than the courts of that state would give it. As Judge Posner notes, there is no reason for such a holding. In Kremer itself, the Court stated that Congress's purpose in enacting section 1738 was "specifically to insure that federal courts . . . would be bound by state judgments." Id. at 484 n. 24, 102 S.Ct. at 1889 n. 24. See also Davis v. Davis, 305 U.S. 32, 40, 59 S.Ct. 3, 6, 83 L.Ed. 26 (1938). The requirement that the federal courts give the same credit to state court judgments as would the courts of that state thus can be interpreted as a convenient means to ensure that state judgments are given full credit; it is not an end in itself. Therefore, I agree with Judge Posner that we are not constrained by section 1738 as long as we do not give a state court judgment less preclusive effect than would the courts of the state from which the judgment is taken.

Though we are not bound by Illinois law, the Illinois courts have stated traditional res judicata principles that are applicable to the facts of this case. Under Illinois law, a cause of action for purposes of res judicata is defined as follows:

[A] cause of action consists of a single core of operative facts which give the plaintiff a right to seek redress for the wrong concerned. Even though one group of facts may give rise to different claims for relief upon different theories of recovery, there remains a single cause of action.

Morris v. Union Oil Co. of California, 96 Ill.App.3d 148, 157, 51 Ill.Dec. 770, 777, 421 N.E.2d 278, 285 (1981), quoted in Lee v. City of Peoria, 685 F.2d at 200. See also Gasbarra v. Park-Ohio Industries, 655 F.2d 119, 121 (7th Cir. 1981) (courts have sought to discover "whether the entire amount claimed to be due plaintiff arises out of one and the same act or contract") (quoting Freeman Co. v. Regan Co., 332 Ill.App. 637, 645, 76 N.E.2d 514, 517-18 (1947)). The Illinois courts also have stated that plaintiffs are required to set out in their pleadings all grounds for recovery that they may have. They may not maintain a subsequent suit after the loss of an earlier one based upon the same facts by the simple expediency of limiting the theories of recovery advanced in the pleadings of the first. Baird Warner v. Addison Industrial Park, 70 Ill. App.3d 59, 64, 26 Ill.Dec. 1, 7, 387 N.E.2d 831, 837 (1979); Prochotsky v. Union Central Life Insurance, 2 Ill.App.3d 354, 356, 276 N.E.2d 388, 390 (1971). These rules are founded upon "the plainest and most substantial justice, that litigation should have an end and that no person should be unnecessarily harassed with a multiplicity of suits." Baird Warner v. Addison Industrial Park, 70 Ill.App.3d at 64, 26 Ill.Dec. at 7, 387 N.E.2d at 837. See also Dorland v. Steinbrecher, 50 Ill.App.2d 344, 347, 200 N.E.2d 424, 425 (1964).

As a matter of federal law, the plaintiffs' earlier state court suits and their current federal suit should be considered to involve the same cause of action for purposes of res judicata. In each suit, the plaintiffs essentially were challenging the legality of the defendant's admissions procedures. In the state suits, their position was based on the theory that the procedures were in violation of state common law and due process. In their federal suit, they have claimed that the procedures amount to a group boycott, in violation of federal antitrust law. The state and federal suits revolve around the same set of operative facts: the facts relating to the manner in which the defendant has chosen to limit its membership. In short, the suits are so closely related that logic dictates that the principles of res judicata should be applied to them.

This pragmatic approach to identifying a cause of action has significant support in both the law and legal literature. The Restatement (Second) of Judgments, which uses the terms "cause of action" and "claim" interchangeably, states:

[T]he concept of a transaction . . . connotes a natural grouping or common nucleus of operative facts. Among the factors relevant to a determination whether the facts are so woven together as to constitute a single claim are their relatedness in time, space, origin, or motivation, and whether, taken together, they form a convenient unit for trial purposes.

The present trend is to see claim in factual terms and to make it coterminous with the transaction regardless of the number of substantive theories, or variant forms of relief flowing from those theories, that may be available to the plaintiff; regardless of the number of primary rights that may have been invaded; and regardless of the variations in the evidence needed to support the theories or rights.

RESTATEMENT (SECOND) OF JUDGMENTS § 24 comments a and b (1980). Professor Currie, cited by Judge Posner, has also suggested a pragmatic approach to the application of res judicata, one in which the "cause of action" terminology would be unnecessary: "Perhaps the time has come . . . to ask simply whether the party to be precluded had adequate opportunity to litigate the matter in the earlier proceeding and whether the matter is closely enough related to the original controversy so that judicial economy would be served by confining litigation to one proceeding." Currie, Res Judicata: The Neglected Defense, 45 U.Chi.L. Rev. 317, 342 (1978). Finally, and most importantly, it should be noted that this circuit has previously adopted a pragmatic approach to determining what constitutes a cause of action. See Himel v. Continental Illinois National Bank, 596 F.2d 205, 209 (7th Cir. 1979) ("primary test for determining if two suits are premised on the same cause of action is whether both suits arise under the same factual situation"), quoted in Gasbarra v. Park-Ohio, Inc., 655 F.2d at 121. See also Lambert v. Conrad, 536 F.2d 1183, 1186 (7th Cir. 1976) ("[s]ince both suits pertain to the same disputed facts and arise out of the same operative facts, they clearly are the same cause of action").

Having determined that the suits at issue here involved identical causes of action for purposes of the res judicata doctrine, the question now becomes whether the substance of the present suit could have been raised earlier. A basic tenet of the res judicata doctrine is that a party may not litigate a dispute and then, upon an unsuccessful disposition, revive the same cause of action with a new theory. This principle has been stated repeatedly by this and other circuits. See, e.g., Mandarino v. Pollard, 718 F.2d 845 at 849 (7th Cir. 1983); Harper Plastics, Inc. v. Amoco Chemicals Corp., 657 F.2d 939, 945 (7th Cir. 1981); Roach v. Teamsters Local Union No. 688, 595 F.2d 446, 450 (8th Cir. 1979); Astron Industrial Associates v. Chrysler Motors, 405 F.2d 958, 961-62 (5th Cir. 1968); Grace v. Grace, 394 F.2d 127, 128 (2d Cir. 1968); Brotherhood of Railroad Trainmen v. Chicago, Milwaukee, St. Paul Pacific Railroad, 380 F.2d 605, 608 (D.C. Cir. 1967). The instance where a party does attempt to revive an unsuccessful cause of action with a new theory is a particularly compelling case for the application of res judicata principles because it is almost always avoidable and can involve significant inconvenience to a defendant. One of the primary purposes of the res judicata doctrine is to protect defendants from the expense and vexation attending multiple lawsuits. Montana v. United States, 440 U.S. 147, 153, 99 S.Ct. 970, 973, 59 L.Ed.2d 210 (1979); Mandarino v. Pollard, at 848. When a defendant is charged with wrongdoing, expends considerable resources to defend him or herself in court, and is awarded a final judgment, it is unfair to make that defendant then go through the whole process anew to defend the same actions that were attacked in the first suit. Yet this is a description of what the plaintiffs would have us sanction in this case.

If this is not a classic case of losing and trying again on a different theory, it is only because, as Judge Cudahy argues, the plaintiffs could not have brought their federal antitrust claim in state court. I agree with Judge Cudahy that this jurisdictional obstacle excuses the plaintiffs from bringing their federal claim in state court with their state claims, but it cannot excuse their waiting until their state suits were lost — a period of more than three years — before bringing their federal action. Insofar as the jurisdictional competency rule relied on by Judge Cudahy requires only that plaintiffs raise those theories of recovery available to them in a particular court — a court of their own choosing — and allows them to raise other theories at any other time, it is a technical rule that is not well designed to serve the policies behind the doctrine of res judicata. I suggest that it is a better rule to require plaintiffs who are unable to bring their entire cause of action in the forum of their choice, such as the plaintiffs in this case, at least to bring their entire cause of action relatively contemporaneously, in order to avoid their having two successive "bites of the apple" at the expense of the defendant. In some cases plaintiffs may choose to bring their entire cause of action in federal court and attempt to have their state claims adjudicated under that court's pendent jurisdiction. In other cases plaintiffs may choose to split their cause of action between state and federal courts. Either course would be preferable to what the plaintiffs chose to do in the present case.

I do not suggest that requiring plaintiffs to bring their entire cause of action within a reasonable period of time would necessarily advance the interests of judicial economy, though I am satisfied that such a requirement would not increase the total volume of litigation. This equitable requirement is simply one that is fair to defendants and at the same time poses no undue hardship on plaintiffs. While this position represents an extension of the more confined and dated interpretations of res judicata, it is in greater compliance with the spirit of the res judicata doctrine than are the more rote approaches. As the Supreme Court has stated, "[T]he doctrine of res judicata is not a mere matter of practice or procedure inherited from a more technical time than ours. It is a rule of fundamental and substantial justice, 'of public policy and private peace,' which should be cordially regarded and enforced by the courts. . . ." Federal Department Stores v. Moitie, 452 U.S. 394, 401, 101 S.Ct. 2424, 2429, 69 L.Ed.2d 103 (1981) (quoting Hart Steel Co. v. Railroad Supply Co., 244 U.S. 294, 299, 37 S.Ct. 506, 508, 61 L.Ed. 1148 (1917)). In this case, because the plaintiffs had an opportunity to litigate their federal antitrust claim in a court of competent jurisdiction and failed to do so until their other legal theories failed, they now should be equitably barred from bringing a federal antitrust suit.

Because I find the plaintiffs' suit barred by res judicata, I deem it unnecessary and inappropriate to reach the discovery question raised in this case. It is essential, however, that the contempt issue be addressed. On that question, I agree with Judge Wood that the contempt citation should remain intact apart from the fact that the plaintiffs' suit has been barred by res judicata, and I join that portion of Judge Wood's opinion dealing with the viability of the contempt citation.

[64] HARLINGTON WOOD, Jr., Circuit Judge, with whom CUMMINGS, Chief Judge, and CUDAHY, Circuit Judge, join, dissenting.

Judge Posner's opinion has eliminated some points which appeared in the two prior opinions in this case, now vacated, and has added some new points. In certain respects the changes make this latest version less objectionable, but I respectfully must dissent. Those two prior opinions created the impression that this court was trying the case on its antitrust merits before the case had much of a chance to get started in the trial court. That prejudgment pall still hangs heavy over Judge Posner's latest opinion. Judge Cudahy's opinion, with which I fully agree, addresses the res judicata issue. I discuss the discovery and contempt issue in this dissent to the discovery portion of Judge Posner's opinion. Perhaps we could have simply adopted Justice Stewart's dissents from the vacated panel decisions, but as this latest opinion has developed, we believe some elaboration is necessary.

First, a review of some basis discovery principles is in order. A criminal contempt citation is a final judgment subject to full rights of appeal. United States v. Ryan, 402 U.S. 530, 532, 91 S.Ct. 1580, 1581, 29 L.Ed.2d 85 (1971); Union Tool Co. v. Wilson, 259 U.S. 107, 111, 42 S.Ct. 427, 428, 66 L.Ed. 848 (1922). Appellate review of a criminal contempt judgment for failure to comply with a discovery order may be used to test the propriety of the underlying discovery order, which is not otherwise appealable. Ryan, 402 U.S. at 532, 91 S.Ct. at 1581; Hanley v. James McHugh Construction Co., 419 F.2d 955, 957 (7th Cir. 1969). The trial judge has wide discretion in shaping pretrial discovery. Matter of Rassi, 701 F.2d 627, 631 (7th Cir. 1983). Reversal of a discovery order is appropriate only upon a clear showing of abuse of discretion. Voegeli v. Lewis, 568 F.2d 89, 96 (8th Cir. 1977); Keyes v. Lenoir Rhyne College, 552 F.2d 579, 581 (4th Cir. 1977), cert. denied, 434 U.S. 904, 98 S.Ct. 300, 54 L.Ed.2d 190 (1977).

When a party objects to discovery of confidential information, the trial court must weigh the interests of the parties and of public policy to arrive at an appropriate discovery compromise. Memorial Hospital for McHenry County v. Shadur, 664 F.2d 1058, 1061-62 (7th Cir. 1981) (per curiam); Keyes, 552 F.2d at 581. Generally, broad discovery is available under the Federal Rules of Civil Procedure. See Fed.R.Civ.P. 26(b)(1). Privileges excluding relevant evidence are construed narrowly because they constrict the fact-finding process. Ryan v. Commissioner of Internal Revenue, 568 F.2d 531, 542-43 (7th Cir. 1977), cert. denied, 439 U.S. 820 (1978). Further, in the antitrust area, there is a strong public policy interest in open competition and private enforcement of the antitrust laws. Memorial Hospital, 664 F.2d at 1062. With these principles forming the backdrop, we now turn to analysis of the discovery order in this case.

The Academy claims that the confidentiality of its files is necessary to preserve asserted first amendment interests in the candid review of applicants for membership. I agree with Judge Posner's opinion that a first amendment argument for complete immunity from discovery is untenable. The freedom of association is not a shield against antitrust regulation. See, e.g., NAACP v. Claiborne Hardware Co., 458 U.S. 886, 102 S.Ct. 3409, 3425, 73 L.Ed.2d 1215 (1982); National Society of Professional Engineers v. United States, 435 U.S. 679, 697, 98 S.Ct. 1355, 1368, 55 L.Ed.2d 637 (1978); American Medical Association v. FTC, 638 F.2d 443, 451-52 (2d Cir. 1980), aff'd by an equally divided court, 455 U.S. 676, 102 S.Ct. 1744, 71 L.Ed.2d 546 (1982). This court recently acknowledged that a privilege against any discovery of even partially privileged materials might serve as a shield to hide evidence of illegal behavior and impair the truth-seeking process. EEOC v. University of Notre Dame Du Lac, 715 F.2d 331, 337 (7th Cir. 1983).

In Notre Dame, which Judge Posner's opinion also cites, EEOC sought discovery of the University's tenure files to investigate an assistant professor's charge that he was denied tenure due to racial discrimination. This court recognized a qualified privilege against disclosure of the sources of documents in the University's tenure files, due to the critical role of confidential tenure review in the pursuit of academic excellence. Nevertheless, we sanctioned limited discovery using the technique of redaction of names and other identifying information. Judge Posner's opinion in the present case seems to accord the Academy as much or more protection than we gave the University in Notre Dame. The opinion, however, admits that the exact nature of the Academy is unclear. When we recognized the qualified academic freedom privilege in Notre Dame, we knew what we were protecting. Adlai Stevenson long ago put it well:

For the university is the archive of the Western mind; it is the keeper of the Western culture and the foundation of Western culture is freedom. Men may be born free; they cannot be born wise, and it is the duty of the university to make the free wise. The university is the guardian of our heritage, the teacher of our teachers. It is the dwelling place of the free mind.

Stevenson (G. D. Darling eds. 1977).

Nothing akin to the qualified academic freedom privilege applies to the Academy. The Academy, the only organization for this profession that counts as members nearly all the orthopaedic surgeons in the United States, does not claim to license, certify, or police the profession. Indeed, membership in the Academy appears to be most conspicuous by its mere absence. Yet the Academy itself fluctuates on the importance of membership. When an anti-competitive group boycott is the topic, the Academy gives the impression that exclusion from its ranks is of little professional consequence. In support of this characterization, the Academy argues that one of its principal functions is to sponsor educational seminars to which even non-members are welcome. However, when discovery of its admissions files is the issue, the Academy views itself as much too important to allow its confidentiality to be disturbed.

Judge Posner's opinion, by bestowing on the Academy some new kind of privilege in the nature of a first amendment interest, seems to agree with the Academy that its files are almost sacrosanct. All variety of trade associations and other types of organizations, from civic clubs to country clubs, should rejoice in their improved status and newly-fashioned protection from troublesome inquiries. Individuals and agencies, however, who have legitimate antitrust, discrimination, and other claims against such organizations have little reason to rejoice.

Rather than following the Notre Dame redaction model in a context not involving the academic freedom privilege, the district court chose to protect the confidentiality of Academy documents by ordering production of the unredacted files on rejected applicants for membership, but with restrictions on disclosure of the contents. Plaintiffs' attorneys were to be permitted to discuss with their clients only the general contents, without identifying sources, of the files on other disappointed applicants. The files were to be used solely for purposes of this litigation. In reaching the discovery compromise, the district court followed the path approved by this court in Memorial Hospital for McHenry County v. Shadur, 664 F.2d 1058 (7th Cir. 1981) (per curiam).

An asserted freedom of association does not raise a privilege against all discovery in the early stages of antitrust litigation, but requires a balance of the interests in confidentiality and disclosure to reach a discovery compromise. The trial judge endeavored to achieve that balance by working with the parties. I believe that the discovery order reasonably protected the Academy's confidentiality interests to at least a minimum extent while granting plaintiffs access to information without which, both Judge Posner's opinion and the district court agree, they probably would be foreclosed from proving their case.

Judge Posner's opinion now finds fault in the district court's failure to incorporate other protective devices for resolving the discovery dispute, such as an in camera inspection of the Academy's files or production of a redacted version of the files. I, too, believe that the trial judge could have improved on his discovery order. First, he could have been well-served by taking a look in camera at the Academy files on plaintiffs to see whether the files contained what plaintiffs claimed to need so badly, and whether the Academy had good reason to fight so desperately to keep the files out of sight.

The Supreme Court has commended an in camera view of documents in a discovery dispute to insure the striking of a proper balance between the interests in confidentiality and disclosure. Kerr v. United States District Court, 426 U.S. 394, 405, 96 S.Ct. 2119, 2125, 48 L.Ed.2d 725 (1976). Failure to make an in camera inspection, however, is not in itself an abuse of discretion. While the district court did not view the Academy's files, it apparently was aware that the state trial court had conducted an in camera inspection of the files in the earlier state court litigation between the same parties. After viewing the files, the state court ordered production over the Academy's objections. Treister v. American Academy of Orthopaedic Surgeons, 78 Ill. App.3d 746, 748, 33 Ill.Dec. 501, 503, 396 N.E.2d 1225, 1226 (1st Dist. 1979) (describing the trial court's action). Thus, one judge had viewed these materials and determined that the contents were relevant and disclosure was appropriate. We should not ignore the state court's first-hand determination. Further, the Academy has maintained that an in camera inspection would not improve the discovery order. In the Academy's view, as I understand it, an in camera inspection could lead to a discovery order which, even if including redaction of the files, would be an abuse of discretion at this stage of the litigation.

The propriety of the district court's order to produce the unredacted files also must be considered in light of the surrounding circumstances. In the early stages of this litigation, plaintiffs expressed their willingness to accept a redacted version of the Academy's files, at least as the first phase of discovery. The Academy refused plaintiffs' early offer to enter a redaction agreement, and plaintiffs later rescinded the offer. Both sides submitted proposed protective orders to the district court; apparently neither provided for redaction.

The Academy's participation in later court proceedings suggested its general approval of the protective order as it was developing. However, the Academy meanwhile continued to press its effort to avoid any discovery of its files. At oral argument, counsel for Dr. Treister stated that his client now would accept a redaction order if this would not limit the right to seek sanctions against the Academy. The Academy's position at oral argument was that any disclosure of its files, even after excision of names and other identifying information, would enable plaintiffs to determine the source of comments, thereby chilling participation in the Academy's admissions process. The Academy steadfastly maintained that no modification of the district court's protective order would be acceptable to it, and any forced disclosure of confidential information prior to the establishment of a jury issue on antitrust injury would be an abuse of discretion. Although the Academy's stonewalling now has paid off, I do not find an abuse of discretion in the district court's failure to order redaction under these circumstances.

Judge Posner's opinion suggests that the district court should have used its Rule 26(d) powers to control the sequence of discovery in this case. Discovery usually may proceed in any sequence, but the court upon motion, for the convenience of parties and in the interests of justice, may schedule the order of discovery on different issues. Fed.R.Civ.P. 26(d). Judge Posner's opinion correctly notes that the district court's power to order sequential discovery may rise to a duty in some cases, citing Bruno Stillman, Inc. v. Globe Newspaper Co., 633 F.2d 583 (1st Cir. 1980), a reporter's privilege case. The present case does not begin to fit into the Bruno Stillman mold. The interests in the confidentiality of the Academy's files do not require such delicate treatment as do these involved in the case of a reporter seeking to protect confidential news sources from discovery. The district court considered and rejected the Academy's motion to bifurcate discovery. The court was satisfied that plaintiffs' claim was not frivolous, having withstood a motion to dismiss once, and again upon reconsideration. The Academy's proposal for sequential discovery would have required piecemeal discovery, which the district court found would burden plaintiffs unnecessarily since the Academy's confidentiality interests could be preserved adequately through a protective order. I would hold that the district court did not abuse its discretion by refusing to order sequential discovery.

Judge Posner's opinion discusses the misuse of discovery tools to coerce settlement, and apparently concludes that the district court's discovery order failed to prevent plaintiffs from abusing the discovery process. The opinion finds two "hints" of predatory discovery here: first, that plaintiffs sought access to federal court — and its liberal discovery rules — only after losing in the state court; and further, that plaintiffs planned to use information from the Academy's files as the basis for further discovery. The first hint merely returns us to the res judicata issue in the context of a dual court system. The second hint, plaintiffs' plan to follow up leads found in the files, is one of the conventional purposes of discovery.

Upon questioning at oral argument, counsel for the Academy labeled plaintiffs' suit a "fishing expedition," an attempt to gain access to otherwise unavailable information. Under the circumstances, however, plaintiffs' pursuit of the files and plan to seek further discovery using leads from the files were within the bounds of appropriate discovery. The discovery record in this case evidences not the slightest abuse, harassment, or coercion to pressure a settlement. Judicial concern about discovery abuse is always legitimate, but such arguments are gratuitous in the context of this case. The abuse of discovery here instead is the Academy's obstinate defiance of the trial court, which now is sanctioned by this court.

Although not condemning any one omission as an abuse of discretion, the discovery majority is left with the "firm conviction" that the district court's discovery order was erroneous. I am not. The district court sought and received proposed protective orders from the parties, and mediated negotiations on this issue. Bifurcation of discovery was not mandatory in this case, and the court reasonably provided for the Academy's confidentiality concerns through the protective order. The Academy should not now reap a windfall from reversal of the discovery order because the order did not incorporate certain provisions that the Academy still would refuse to accept. The Academy treated the trial judge's reasonable discovery order with contempt and its contempt should be recognized by this court.

The Academy further objects to the contempt penalty. It argues that the $10,000 fine invoked the right to a jury trial, and the court erred in not granting a jury trial on the contempt adjudication. The Academy also challenges the fine as excessive in view of the nature of the offense.

The right to a jury trial for a charge of criminal contempt arises when the penalty imposed exceeds what would be considered a petty offense under the statutory definition of federal criminal offenses. Frank v. United States, 395 U.S. 147, 89 S.Ct. 1503, 23 L.Ed.2d 162 (1969); Cheff v. Schnackenberg, 384 U.S. 373, 86 S.Ct. 1523, 16 L.Ed.2d 629 (1966). A petty offense is any misdemeanor for which the penalty does not exceed six months' imprisonment or a $500 fine. 18 U.S.C. § 1(3). The $10,000 penalty against the Academy obviously exceeds the $500 limit for petty offenses, and was imposed without a jury trial. Plaintiffs argue, however, that the Academy waived its right to a jury trial. The development of the contempt citation in this case convinces me that such a waiver did take place and that the amount of the fine was not an abuse of discretion. See Shibley v. United States, 236 F.2d 238, 240 (9th Cir. 1956), cert. denied, 352 U.S. 873, 77 S.Ct. 94, 1 L.Ed.2d 77 (1956) (abuse of discretion standard of review); Jewel Tea Co. v. Kraus, 204 F.2d 549, 551 (7th Cir. 1953) (same).

In the presence of all parties, the district court and the Academy discussed possible avenues for the Academy to procure appellate review of the discovery order, after this court had dismissed an earlier appeal attempt for lack of jurisdiction. The Academy suggested that the court hold it in criminal contempt with a fine of $500, and waived issuance of a rule to show cause. Plaintiffs, desiring to coerce production, proposed a civil contempt citation with a daily $5,000 fine. A few days before entering the contempt order, the district court informed the parties of its intention to impose, as a compromise, a criminal contempt citation with a one-time $10,000 fine. After the district court informed the Academy of its draft order, the Academy did not object or attempt to rescind its waiver of a rule to show cause; nor did it assert its right to a jury trial. Rather, the Academy appeared willing to accept the citation and fine as the basis for an appeal to this court, and risk paying the price. It is now too late for the Academy to become parsimonious.

Judge Posner's opinion declines review of the contempt citation under a belief that the citation falls with reversal of the district court either on res judicata grounds or for abuse of discretion in shaping the discovery order. The status of a contempt citation after reversal of the underlying order is an issue independent of the propriety of reviewing the underlying order on appeal of the contempt citation. On the latter issue, which I have discussed above, I agree with Judge Posner's opinion that review of the underlying discovery order is appropriate here. On the former issue, however, the opinion cites yet overlooks precedent which supports sustaining the contempt order despite reversal of the district court order on other grounds.

As Judge Posner's opinion notes, many decisions hold that a contempt citation remains intact even after invalidation of the underlying court order which was violated. See, e.g., Maness v. Meyers, 419 U.S. 449, 458, 95 S.Ct. 584, 590-91, 42 L.Ed.2d 574 (1975); United States v. United Mine Workers of America, 330 U.S. 258, 293, 67 S.Ct. 677, 695, 91 L.Ed. 884 (1947). See also Blocksom and Co. v. Marshall, 582 F.2d 1122, 1124 (7th Cir. 1978) ("it is no defense to a charge of criminal contempt that the order disobeyed was invalid"). The Court in United Mine Workers concluded:

[W]e find impressive authority for the proposition that an order issued by a court with jurisdiction over the subject matter and person must be obeyed by the parties until it is reversed by orderly proceedings. This is true without regard for the constitutionality of the Act under which the order is issued. . . . Violations of an order are punishable as criminal contempt even though the order is set aside on appeal.

Id. 330 U.S. at 293-94, 67 S.Ct. at 696 (footnotes omitted).

The district court had both subject matter and personal jurisdiction in this action. Even if the action is barred by res judicata, this is not a jurisdictional deficiency; res judicata is an affirmative defense that may be pleaded and proved, or waived, by the party asserting it. See, e.g., Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation, 402 U.S. 313, 350, 91 S.Ct. 1434, 1453, 28 L.Ed.2d 788 (1971); Church of the New Song v. Establishment of Religion on Taxpayers' Money in the Federal Bureau of Prisons, 620 F.2d 648, 651 (7th Cir. 1980), cert. denied, 450 U.S. 929, 101 S.Ct. 1387, 67 L.Ed.2d 361 (1981); Fed.R.Civ.P. 8(c). See also 18 C. Wright, A. Miller E. Cooper, Federal Practice and Procedure § 4405 (1981). The criminal contempt citation against the Academy remains very much alive. The district court's jurisdiction was uncontested. The court entered a discovery order which the Academy chose to disobey. The Academy itself invited the contempt citation which has enabled it to perfect this appeal.

Although the discovery order could have been improved, the district court's fashioning of the terms was not an abuse of discretion under the circumstances of this case. What the merits of this case would have turned out to be, we now will never know; but we must not let a prejudgment on the merits cloud our review of the discovery order. Plaintiffs, their case not barred by res judicata, deserve the opportunity within reasonable limits to develop their case, and then the opportunity to try it before a judge and jury. I would affirm the district court's contempt holding, but on remand I would direct the court to view the Academy's files in camera and to consider possible redaction before actually enforcing the discovery order. This opportunity to reconsider the discovery order would be a fairer disposition of this case, giving the Academy another chance to review its position on discovery without depriving plaintiffs of their chance to see if they can make out a case.

Judge Posner's opinion reaches the discovery issue despite its conclusion that the suit should be dismissed on res judicata grounds. The opinion directs that plaintiffs' complaint be dismissed with prejudice. That is a hard blow to strike, particularly when paving new ground in the res judicata area. I have no reservations about joining Judge Cudahy's opinion on the res judicata issue, and I respectfully dissent on the discovery and contempt issue.

[92] CUDAHY, Circuit Judge, with whom CUMMINGS, Chief Judge, and HARLINGTON WOOD, Jr., Circuit Judge, join, dissenting.

I join enthusiastically in Judge Wood's able dissent on the discovery branch of this case while, at the same time, I write separately in dissent principally on the plurality's res judicata holding.

We are presented here with a Sherman Act claim against what may for present purposes be characterized as a dominant national trade association. As Illinois Supreme Court Justice Seymour Simon, then of the Illinois Appellate Court, has said,

[Dr. Treister] seeks admission to a prestigious medical society, not to a social, religious, or fraternal organization like an Elks Club . . . or the Order of the Eastern Star. . . . The Academy holds itself out as the sole legitimate organization of its kind in the profession, and is widely recognized as such. It is a monopoly; and it is affiliated with the A.M.A., itself a monopoly. . . . The Academy is more like a government than like a truly voluntary combination, and may justly be treated accordingly.

Treister v. American Academy of Orthopaedic Surgeons, 78 Ill. App.3d 746, 759-67, 33 Ill.Dec. 501, 600, 602, 605, 396 N.E.2d 1225, 1234, 1236, 1239 (1979) (Simon, J., dissenting).

When a large national organization with a membership including the vast majority of the board-certified orthopaedic surgeons of the country conducts its affairs, one may assume that its interests cover the gamut of economic, as well as medical, concerns of the membership. Whatever differences may otherwise exist, at least for antitrust purposes, the Academy seems comparable with other great national trade or professional associations such as the American Bar Association, the Cement Manufacturers Protective Association and the American Iron and Steel Institute.

I wonder whether the majority would extend the solicitude which it feels for the secret files of the Academy if one of these other trade associations were the defendant. I see no reason why the principles announced here would not apply equally to such a case. I believe the Academy is more like the American Iron and Steel Institute — both organizations are comprised of the practitioners of a particular trade, business or profession (who are potential competitors) — than the Academy is like the National Association for the Advancement of Colored People. Yet the majority discusses NAACP v. Alabama, 357 U.S. 449, 78 S.Ct. 1163, 2 L.Ed.2d 1488 (1958), as if it had significant bearing on the case at bar. Supra, at 1159. The NAACP, like the John Birch Society, is an advocacy organization, having occupationally diverse members committed to the same (relatively controversial) ideas and causes. First amendment concerns with respect to such organizations are obvious. The members of the Academy, on the other hand, have in common, like other trade groups, their occupation, business or profession. Like other such groups, the Academy presumably promotes the betterment and serves the occupational interests of its members.

The majority cites no case which recognizes any "interest" under the first amendment, or its penumbra, in secret proceedings of trade or professional associations. At oral argument, counsel for defendants indicated that the same secrecy interest belongs to all private organizations — including country clubs. I believe there are serious dangers in recognizing substantial rights to secrecy in organizations which bring together members of the same trade, occupation or profession, where no legal privilege exists. Cf. Memorial Hospital for McHenry County v. Shadur, 664 F.2d 1058 (7th Cir. 1981) (per curiam). When people who make their living in the same way join together — however exalted their purposes — it is the height of naivete to suggest that their association is without economic significance. Certainly the antitrust laws have never indulged in any such unrealistic assumption. As Justice Simon has said:

The secrecy in which . . . associations [like the Academy] have historically cloaked themselves is therefore little proof of its value for legitimate purposes, and is open to sustained abuse. . . . [T]he power of groups like the Academy is so great, so self-perpetuating, so little checked by private forces, and has such potential for abuse, that the balance of interests favors judicial oversight.

78 Ill.App.3d at 766, 33 Ill.Dec. at 605, 396 N.E.2d at 1239 (Simon, J., dissenting). In this connection I note that our recent decision in EEOC v. University of Notre Dame Du Lac, 715 F.2d 331, 340 (7th Cir. 1983), involving the confidentiality of peer review documents, specifically recognized a qualified academic freedom privilege.

The plurality, with Judge Flaum concurring, has, however, seen fit to dispose of this case not primarily on the basis of the right to secrecy but rather on grounds of res judicata. At least the plurality describes the grounds of its ruling as "res judicata," although the application here is hardly recognizable in terms of the well accepted principles of that doctrine. The plurality concedes that, as Justice Potter Stewart noted in his dissent from the first panel opinion, Marrese v. American Academy of Orthopaedic Surgeons, 692 F.2d 1083, 1099 (7th Cir. 1982) (Stewart, J. (Retired), dissenting), reh'g granted and vacated (1983), the overwhelming weight of authority is against it. See, e.g., LektroVend Corp. v. Vendo Co., 660 F.2d 255, 274 (7th Cir. 1981), cert. denied, 455 U.S. 921, 102 S.Ct. 1277, 71 L.Ed.2d 461 (1982); Hayes v. Solomon, 597 F.2d 958, 984 (5th Cir. 1979), cert. denied, 444 U.S. 1078, 100 S.Ct. 1028, 62 L.Ed.2d 761 (1980); Kurek v. Pleasure Driveway Park District of Peoria, 583 F.2d 378, 379 (7th Cir. 1978) (per curiam), cert. denied, 439 U.S. 1090, 99 S.Ct. 873, 59 L.Ed.2d 57 (1979); Abramson v. Pennwood Investment Corp., 392 F.2d 759, 762 (2d Cir. 1968). The plurality opinion is in fact an aggressive tour de force — going well beyond existing law — in the abdication of federal jurisdiction.

Since the plurality opinion seems to take us so far from the terra firma of solid precedent and facts of record, let us first turn to fundamentals. An old and honored formulation of the res judicata defense in barring a subsequent suit is that "a right, question, or fact distinctly put in issue, and directly determined by a court of competent jurisdiction, as a ground of recovery, cannot be disputed in a subsequent suit between the same parties or their privies . . . ." Southern Pacific Ry. Co. v. United States, 168 U.S. 1, 48-49, 18 S.Ct. 18, 27, 42 L.Ed. 355 (1897). See also Montana v. United States, 440 U.S. 147, 153-54, 99 S.Ct. 970, 973-74, 59 L.Ed.2d 210 (1979). The plurality opinion, however, does away with the requirements that there be identity of causes of action and that the first judgment be rendered by a court jurisdictionally competent to hear the claim which is later held to be barred.

This is judicial activism in the service of judicial abdication. We are told, in effect, that state and federal courts are busy, and it will save everyone's time if the plaintiffs are refused the right to bring their Sherman Act claim in the only court which can hear it. It would surely save us even more time if the federal courts refused to hear all claims on the theory that the state courts could hear ones that are similar. With all respect, I am afraid we are simply refusing to do our job as federal judges in circumstances where the well-accepted rules plainly require us to do it.

In determining when to apply res judicata, the plurality opinion formulates a new "federal rule" that a federal court should give a prior state court judgment greater preclusive effect than would a state court. But the federal courts, in fact, are bound under the "full faith and credit" clause of the Constitution and 28 U.S.C. § 1738 to give the same res judicata effect to a prior state court judgment as would another court of that same state. In this situation, "more" is clearly not "better." The plurality does concede that a federal antitrust claim could not arise in state court and that therefore the outcome of such a scenario could not specifically decide the case before us. Nonetheless, because the plurality continues to advocate a "federal rule" which would be more preclusive than state practice, it is appropriate for me to address the reasons why a state law test should govern.

The plurality's new rule is contrary to section 1738 as well as to the purposes underlying "full faith and credit." As Justice White has written,

[i]t has long been established that § 1738 does not allow federal courts to employ their own rules of res judicata in determining the effect of state judgments. Rather, it goes beyond the common law and commands a federal court to accept the rules chosen by the State from which the judgment is taken.

Kremer v. Chemical Construction Corp., 456 U.S. 461, 481-82, 102 S.Ct. 1883, 1897, 72 L.Ed.2d 262 (1982). This is the Court's understanding of Congress' intent and the Kremer holding reflects the plain command of section 1738.

The plurality, however, argues that

section 1738 ought not prevent a federal court from giving a state judgment more effect in a federal suit than the courts of the state would give it in a state suit. A federal court does not undermine the authority of a state's courts when it holds that a state judgment bars a federal suit.

Supra, at 1154 (emphasis in original) (citations omitted). Most of the authority cited by the plurality for this proposition does not specifically discuss this issue and predates the two recent Supreme Court decisions, Kremer and Allen v. McCurry, 449 U.S. 90, 101 S.Ct. 411, 66 L.Ed.2d 308 (1980), which do address the issue. The plurality also cites Federal Deposit Insurance Corp. v. Eckhardt, 691 F.2d 245 (6th Cir. 1982), which in fact holds that the preclusive effect of a prior judgment is determined by the law of the rendering forum (in that case, the state of Ohio). The Sixth Circuit emphasized the new trend supporting this holding, in stating, "the emerging rule is that the preclusive effect of a valid judgment is to be determined by the law of the system which rendered the judgment." Id. at 247. See also General Foods Corp. v. Massachusetts Department of Public Health, 648 F.2d 784 (1st Cir. 1981). A perceptive commentator has recently endorsed this modern trend:

A different rule is emerging today, and an even stronger one will prevail in the future. It will read something like this:

A valid judgment rendered in any judicial system within the United States must be recognized by all other judicial systems within the United States, and the claims and issues precluded by that judgment, and the parties bound thereby, are determined by the law of the system which rendered the judgment.

Degnan, Federalized Res Judicata, 85 YALE L.J. 741, 773 (1976) (emphasis in original).

In addition to the recent authority contradicting the plurality's proposed rule, there are important policy considerations which also militate against it. Deference to state courts is not the only value to be served by the res judicata doctrine. Even the commentator whom the plurality opinion cites approvingly recognizes that there are other values at stake when making a res judicata determination:

[T]his reading of section 1738 [to allow greater preclusive effect] might offend state policies limiting the res judicata effect of judgments, and there may be due process problems in giving a judgment greater effect than the court entering it intended . . . .

Currie, Res Judicata: the Neglected Defense, 45 U. CHI.L.REV. 317, 327 (1978). The plurality has thus overlooked the policy interests of the rendering forum in predicting the preclusive effect of its own judgments. In addition, by rejecting the Illinois law of res judicata in the guise of improving upon it, the plurality has undermined the Illinois law to which it purports to defer.

The significance of section 1738 and its interpretation in Kremer is that a federal court, in determining the res judicata effect of a prior state court judgment, should apply the criteria of the state's res judicata law. It would be incorrect to limit the question strictly to the outcome of a federal antitrust claim in state court because, as the plurality concedes, such a suit would be dismissed for lack of jurisdiction. Kremer still dictates, however, that the appropriate issue is the application of the principles of Illinois res judicata law to the current case.

Illinois res judicata law thus requires that, in order to bar a subsequent claim which was not raised in the first suit, the first court must have had jurisdiction to hear the subsequent claim.

The doctrine of res judicata provides that a final judgment rendered by a court of competent jurisdiction on the merits is conclusive as to the rights of the parties and their privies, and as to them, constitutes an absolute bar to a subsequent action involving the same claim, demand, or cause of action. ( People v. Kidd (1947), 398 Ill. 405, 408, 75 N.E.2d 851).

Rotogravure Serv. v. R.W. Borrowdale Co., 77 Ill.App.3d 518, 524, 32 Ill.Dec. 762, 767, 395 N.E.2d 1143, 1148 (1978) (emphasis supplied). Apparently, no Illinois judge has questioned the jurisdictional competency rule; any departure from the rule has presumably been deemed so unfair and illogical as to require no discussion. Therefore, because the Illinois courts did not have jurisdiction to hear the plaintiffs' Sherman Act antitrust claim, the prior Illinois suit cannot bar the subsequent federal suit.

Turning from state law, I shall now examine the plurality's suggestion that the plaintiffs' federal antitrust claim, as a matter of federal law, should be barred in federal court because they should have brought either a federal or a state antitrust claim in state court simultaneously with their common law claims. However, a federal antitrust claim brought in an Illinois court would certainly have been dismissed for lack of jurisdiction. The plurality laments the fact that federal courts have long asserted exclusive jurisdiction under the federal antitrust statutes. Blumenstock Bros. Advertising Agency v. Curtis Publishing Co., 252 U.S. 436, 440, 40 S.Ct. 385, 386, 64 L.Ed. 649 (1920); Kurek v. Pleasure Driveway Park District, 583 F.2d 378 (7th Cir. 1978). Ultimately, however, the plurality acknowledges that this court, even in the course of its more creative efforts at narrowing its own powers, must accept this rule as a given.

In his dissent from the first panel decision in this case, Justice Stewart observed that the federal common law of res judicata bars the relitigation of all those claims — but only those claims — which were or could have been brought in a prior litigation between the same parties arising from the same cause of action. As the Supreme Court said in Federated Department Stores v. Moitie, 452 U.S. 394, 101 S.Ct. 2424, 69 L.Ed.2d 103 (1981):

There is little to be added to the doctrine of res judicata as developed in the case law of this Court. A final judgment on the merits of an action precludes the parties or their privies from relitigating issues that were or could have been raised in that action. Commissioner v. Sunnen, 333 U.S. 591, 597 [ 68 S.Ct. 715, 719, 92 L.Ed. 898] (1948); Cromwell v. County of Sac, 94 U.S. 351, 352-353 [ 24 L.Ed. 195] (1877).

Id. at 398, 101 S.Ct. at 2428.

By definition, the "could have been raised" standard necessarily entails the jurisdictional competency of the court that decided the prior litigation to rule on the merits of any such possible other claims. Sherman Act claims are within the exclusive jurisdiction of the federal courts; they simply cannot be brought in the courts of Illinois.

The plurality also raises the possibility that the plaintiffs, unable to bring their federal claim in state court, should have brought a state antitrust claim. The plurality argues that the alleged near identity of such a state antitrust claim with a Sherman Act claim should bring into play the application of res judicata to the present federal claim. Never in recent times — with only one exception — has a federal court held a federal antitrust claim barred by the availability or actual prosecution of a state antitrust action.

The plurality rejects the myriad precedents against a state action as a bar ostensibly because most of these decisions predate the Fourth Circuit's departure from this rule in Nash County Board of Education v. Biltmore Co., 640 F.2d 484 (4th Cir. 1981). In the one decision subsequent to Nash, the Ninth Circuit declined to follow the Fourth Circuit decision since it was easily distinguishable. Turf Paradise, Inc. v. Arizona Downs, 670 F.2d 813, 818 n. 1 (9th Cir. 1982), cert. denied, 456 U.S. 1011, 102 S.Ct. 2308, 73 L.Ed.2d 1308 (1982). However, the RESTATEMENT (SECOND) OF JUDGMENTS is explicit on the very problem presented in Nash and in the case before us.

A Co. brings an action against B Co. in a state court under a state antitrust law and loses on the merits. It then commences an action in a federal court upon the same facts, charging violations of the federal antitrust laws, of which the federal courts have exclusive jurisdiction. The second action is not barred.

RESTATEMENT (SECOND) OF JUDGMENTS § 26(1)(c) comment c, illustration 2 (1982).

In Nash, the single modern exception to the rule that a state action cannot bar a Sherman Act claim, the plaintiff school board attempted to bring a federal antitrust claim. This claim was held to be barred by res judicata because of an earlier state antitrust claim prosecuted by the state attorney general — a privy of the plaintiff. Very importantly for the equities of the matter, the earlier case had been settled with a consent decree accepted by all parties to the litigation. In holding that the federal suit was barred, the court also noted that the North Carolina antitrust statute was identical to the comparable federal statute except for the latter's requirement of involvement in interstate commerce. 640 F.2d at 490.

There are at least four persuasive reasons for refusing to follow Nash. First, the Nash court used language apparently recognizing that the court hearing the first claim must have had jurisdiction to hear the second claim before res judicata could be applied. Nonetheless, in holding that the state antitrust claim barred the federal claim, the Nash court in effect eliminated this jurisdictional competency requirement from the res judicata formulation which the Supreme Court had once again made explicit in its 1981 decision, Federated Department Stores, supra. The Nash court thus misconstrued the Supreme Court's res judicata formulation in Montana and Federated Department Stores and went on to base its analysis on the incorrect thesis that an exclusively federal action and a state court action could ever be the same "cause of action" and that the latter could thereby bar the former.

Second, the special and strong equities of Nash induced the Fourth Circuit to overlook overwhelming federal and other authority, as well as crucial and long-honored policies. Nash is a hard case that has made bad law. The plurality here fails to note that the first lawsuit in Nash had been terminated by a settlement and a consent decree involving the very antitrust allegations that formed the basis of the federal suit. This presented the strongest possible equitable case for claim preclusion against a subsequent federal treble damages antitrust action brought by a party in privity. The termination of the first lawsuit by a settlement and consent decree is the true rationale underlying the Nash decision, and Nash is, therefore, not in fact in conflict with the holding which I think appropriate here. In the case before us, Drs. Treister and Marrese never obtained a consent decree in their state common law actions. In fact, these plaintiffs never got their case beyond the pleading stage. Instead, the state court merely found (over Justice Simon's persuasive dissent) that the doctors had failed to allege an essential element, economic necessity, of their common law claim.

Third, there is in the case before us a lack of identity between the Illinois antitrust statute and the Sherman Act. The plurality here concedes that in Nash the state statute was a "mirror image" of the federal. Supra, at 1153. For one thing, the liability provisions of the Illinois statute (unlike that of North Carolina) are on their face different from the Sherman Act provisions. In addition, the federal statute grants treble damages to a successful plaintiff as a matter of right. The statute in Nash did the same. The Illinois statute, however, permits damages in a "rule of reason" case up to a treble amount only at the discretion of the trial court and when the violation was willful. ILL.REV.STAT. ch. 38, § 60-7(2) (1981). Although not considered a significant distinction by the plurality, the availability and structure of monetary damages may make every difference in the world in a plaintiff's decision to pursue a particular remedy. The Nash court itself relied on this factor to distinguish its decision from the long line of cases which hold that a federal antitrust claim is not barred by the availability of a state antitrust claim.

In this suit, however, under the state statute, modeled as it is after the federal statute and offering the same right to recover treble damages as the federal statute, there is not the ground for denying res judicata effect to the state court judgment stated in [ Hayes v. Solomon, supra] and [ Cream Top Creamery v. Dean Milk Co., supra, n. 12].

640 F.2d at 490.

There is a delicate balance in res judicata doctrine between the undesirability of relitigating the same claim and the desirability of having a chance to litigate a particular claim at least once. To ignore Nash's reliance upon absolute identity of claims and to move on to mere similarity of claims is to lose all coherence and predictability in the law of res judicata, as well as to visit manifest injustice upon litigants.

Fourth, the Fourth Circuit in Nash did not even suggest that a prior state court action, as different from the subsequent exclusively federal action as were the prior state court actions here, could constitute the same "cause of action." I should think a minimal concern for fairness here ought to prevent us from following the Nash approach. As the district court has said with respect to the present case:

Nash considers that a plaintiff who "choos[es] to file the state action on the same cause of action [has] voluntarily waived the benefit, if any, of a federal forum" ( 640 F.2d at 492) — and that simply does not apply here.

Marrese v. American Academy of Orthopaedic Surgeons, 524 F. Supp. 389, 394 (N.D.Ill. 1981).

Although the plurality displays an impressive conceptual virtuosity in pursuing a sort of free-floating theory of res judicata, it goes well beyond any precedent or well-accepted explanation. The plurality rationalizes its position by invoking certain policy goals which it purports to further, yet it both overlooks other policies, which are at least as important, and subverts the very policies which it professes to advance.

The first and perhaps most crucial policy at issue here is the one underlying the Sherman Antitrust Act. 15 U.S.C. § 1 et seq. The Sherman Act is our nation's "comprehensive charter of economic liberty" and embodies, together with its judicial interpretations, the fundamental rules of competitive conduct in our national market economy. Northern Pacific Ry. Co. v. United States, 356 U.S. 1, 4, 78 S.Ct. 514, 517, 2 L.Ed.2d 545 (1958).

It is not difficult to see . . . that a congressional intent that the Sherman Act reach as far as the Constitution permits would embrace virtually any economic act in our infinitely interconnected national economy . . . . Ours is a unitary national economy. With a few exceptions for contrary state and federal law, . . . the Sherman Act commands a competitive regime for which state borders are irrelevant . . . . [T]he economy is a seamless web in which everything affects everything. What may appear to be relatively local will, in fact, ultimately affect similar or related goods or services moving into or out of a market.

1 P. Areeda and D. Turner, ANTITRUST LAW 229 (1978). If the national economy is a "seamless web," the possible bias of state courts toward local industries and toward local trade customs makes it undesirable that they administer the Sherman Act. In addition, a uniform interpretation of the antitrust laws in the context of uniform rules of federal practice and of discovery is necessary to a consistent and predictable application of national policy toward competition.

Federal judges are relied upon to achieve uniform and expert interpretation of the law in a way that cannot be achieved by Supreme Court review of state decisions. Federal procedure is relied upon to achieve accurate resolution of disputed factual matters, with particular emphasis on the availability of jury trial and searching discovery.

18 C. Wright, A. Miller, E. Cooper, FEDERAL PRACTICE AND PROCEDURE § 4470, at 676 (1981) (citations omitted).

The plurality suggests that the reason for exclusive federal jurisdiction must be "a judgment that only federal judges can decide antitrust cases intelligently . . . ." But the plurality's protestations of respect for the mental prowess of state judges are beside the point and prove too much. If the intelligence of state judges were the only limitation on the abdication of federal jurisdiction, all federal claims could be tried in state court and the federal courts could go out of business.

Allowing the different state courts to adjudicate federal antitrust issues would tend to result in 50 different interpretations of the antitrust statutes. Such a state of affairs is inconsistent with the goal of a national regime of competition. Without uniform enforcement in the federal courts, a corporation selling in a national market could, for example, find its prices deemed predatory in one state and not in another. The gain which the plurality may perceive in judicial economy (which is entirely speculative) could be lost many times over through a loss of uniformity and certainty in the application of the antitrust laws, as well as through a need to relitigate the legitimacy of the same practice several times in different state courts.

Second, an important policy goal which the plurality ascribes to Federated Department Stores and Nash — the lightening and efficient handling of the judicial caseload — is not furthered by the plurality decision here. In fact, the plurality decision may produce a result directly opposite to its professed intentions. Under the views of res judicata espoused both by the plurality and by Judge Flaum, plaintiffs with both state and federal claims will be compelled to file claims in both courts at the same time; furthermore, the cautious plaintiff will be inclined to file all state claims as pendent to a federal claim — the strategy which the plurality recommends. Not only will this require federal district courts to adjudicate many state claims and thus further burden the federal system, but it will specifically require federal judges to decide numerous questions of state law. This result is surely not consistent with the plurality's presumed concern for deference to state courts. In sum, this approach will backfire by moving litigation into federal court, by strengthening the federal role in state matters and by doing nothing to diminish the overall volume of litigation.

Again, as we have noted, this is distinctly not a question of relative intelligence of state and federal judges, but it may be a question of relative expertise. Just as federal judges may have greater experience in the adjudication of federal antitrust claims, so state judges develop more extensive experience in the law of their own state. From this standpoint, there would be no practical advantage in trying Sherman Act claims in state court or state common law claims in federal court. Further, plaintiffs should have the right to choose their forum, particularly when issues of relative expertise are at stake, without their choice being arbitrarily dictated under a rigid doctrine of federal abstention. As Justice Stewart wrote in his dissent to the first panel opinion:

[t]his expansion of the doctrine of res judicata contradicts the principle, fundamental to our legal system, that parties, not judges, choose the forum in which to resolve their differences and frame the claims advanced in their pleadings.

692 F.2d at 1100.

The plurality approach ignores the possibility that the plaintiff has identified one claim which he believes to be easier to prove than the others. Here the plaintiffs brought a common law claim in state court, which certainly seems plausible to me but which bears no relation to the antitrust laws. Plaintiffs pursued this course in the expectation, I presume, that they would have an opportunity to bring the federal antitrust claim at some other time, if that proved necessary. I do not see how requiring them to bring the federal claim at the outset — or risk losing it — conserves the resources of the federal court.

Third, I think we should consider seriously the ramifications of the kind of freewheeling res judicata analysis employed by the plurality. In considering whether the liability and damages standards are the same under the Illinois antitrust statutes as under the federal antitrust statutes, the plurality says:

The question should be answered not in gross but with reference to the specific provisions of the state and federal statutes, read in light of the specific allegations of the complaint. If, applied to the particular case, the state and federal standards are the same, it should not matter that applied to some other case they might be different.

Supra, at 1155.

The plurality then conducts a quite speculative analysis of the plaintiffs' claims under the state statute and their claims under the federal statute. The plurality tracks possible outcomes with respect to liability under both, as well as the respective remedies available under the state and federal statutory schemes. At last, the plurality concludes that the standards of liability would have been the same in a state suit as in a federal suit and that plaintiffs were probably not much interested in the more generous damages provisions of the federal scheme. Supra, at n. 15.

This sort of speculation (or second-guessing) about the plaintiffs and their thoughts and intentions is so conjectural as to resemble more a flight of fancy than an exploration of fact. It would be manifestly impractical for us to guess in every case, for example, whether or not the plaintiffs "really" needed or "really" sought damages as a basis for determining whether their state action should preclude their federal action. This is surely not the path to certainty in the law, and lack of certainty is a major contributor to unnecessary lawsuits. Nor should we retroactively dictate a party's litigation strategy by determining when and in what court he should have brought his claims.

Fourth and finally, there is a basic problem of fairness in the approach advocated by the plurality. The plurality may have concluded in advance of trial (and even of discovery) that the defendant is a wholly innocent vehicle of professional enlightenment, while the plaintiffs are vengeful malcontents seeking to harass their betters. On this basis, the plurality thinks it appropriate to construct a new law of res judicata to employ the plaintiffs' abortive invocation of state law as a barrier to entry into the federal courts. Dr. Treister has alleged that he was blackballed because he had testified for plaintiffs in medical malpractice suits. If true, this is a claim that federal courts should presumably hear; see Williams v. St. Joseph Hospital, 629 F.2d 448 (7th Cir. 1980) (allegation that doctors conspired to refuse to treat patients who instigated malpractice suits stated a cause of action under the federal antitrust statutes). We, of course, will never know whether Dr. Treister's allegation is even partly substantiated. Discovery has been blocked in both state court and federal court, and, if Dr. Treister or Dr. Marrese has any just claims, they will be heard by no one.

The plurality is concerned about the possible hardship to those who vetoed the plaintiffs' membership applications (or their informants) if the plaintiffs should identify them and subsequently take their depositions. Nothing is said about the hardship to Drs. Marrese and Treister of being excluded without stated reason from this powerful association of orthopaedic surgeons, which bears a coveted seal of approval. Harassment by discovery is to be deplored but not more so than arbitrary and unjust denial of access to the federal courts.

I therefore respectfully dissent.