Watson v. Shell Oil Co.

UNITED STATES COURT OF APPEALS

FOR THE FIFTH CIRCUIT

No. 91-3449

TROY WATSON, ET AL., Plaintiffs-Appellees,

versus

SHELL OIL COMPANY and BROWN & ROOT, U.S.A., INC., Defendants-Appellants.

• • * * * *

ROBERT ADAMS, SR., ET AL., Plaintiffs-Appellees,

versus

SHELL OIL COMPANY and BROWN & ROOT, U.S.A., INC., Defendants-Appellants.

Appeal from the United States District Court for the Eastern District of Louisiana

(December 7, 1992)

Before POLITZ, Chief Judge, REYNALDO G. GARZA and WIENER, Circuit Judges.

POLITZ, Chief Judge:

Shell Oil Company and Brown and Root, U.S.A., Inc., defendants in this mass-tort class action, have permissibly appealed *2 interlocutory orders in this diversity suit. The orders at issue define the class and class issues, designate class representatives, and set a trial plan. Finding neither error nor abuse of discretion, for the reasons assigned we affirm the proposed trial plan.

I. Background

This litigation arises out of an explosion at Shell's manufacturing facility in Norco, Louisiana. At approximately 3:30 a.m. on May 5, 1988, failure of a pipe elbow, allegedly fabricated and installed by Brown & Root, permitted the escape of a vapor cloud of combustible gases. The vapor ignited and a massive explosion ripped through the plant, causing extensive damage both on the plant site and in the surrounding communities. That same morning the instant federal class action suit was filed. During the next week class action suits were filed in Louisiana state courts and were removed to federal court. The claims against Shell are founded on Louisiana law theories of negligence, strict liability and intentional tort. Plaintiffs assert claims in negligence and strict liability against Brown & Root.1 Plaintiffs also seek punitive damages against both defendants.2

1 See La. Civ. Code Ann. arts. 2315, 2316, 2317, 2322 (West 1979 & Supp. 1992). 2 See La. Civ. Code Ann. art. 2315.3 (West Supp. 1992). After certifying the orders on appeal pursuant to 28 U.S.C. § 1292(b), the district court granted summary judgment in favor of Brown & Root on plaintiffs' strict liability and punitive damages claims. See In re Shell Oil Refinery, 769 F. Supp. 214 (E.D. La. 1991) (punitive damages); In re Shell Oil Refinery, 765 F. Supp. 324 (E.D. La. 1991) (strict liability). As the district court did not certify those rulings for interlocutory appeal, they are

The actions were consolidated and referred to a magistrate judge with instructions to conduct an evidentiary hearing and to submit a report and recommendation regarding designation of class representatives and subclass definitions. The district court substantially adopted the magistrate judge's recommendations, certified the litigation as a class action under Fed. R. Civ. P. 23(b)(3), defined the plaintiff class,3 and, pursuant to Fed. R. Civ. P. 23(c)(4), defined the "outside the gate" and "inside the gate" subclasses ("Subclass A" and "Subclass "B", respectively).4

not now before us. 3 The district court defined the plaintiff class as:

All persons or entities who were physically present or owned property within the Parishes of St. Charles, St. John the Baptist, St. James, Orleans, or Jefferson on May 5, 1988, and who sustained injuries or damages as a result of the explosion at the Shell Oil Refinery in Norco, Louisana.

See In re Shell Oil Refinery, 136 F.R.D. 588, 590 & n.1 (E.D. La. 1991). 4 See id. Subclass A is defined as:

Those persons or entities having claims for damages or injuries caused by the explosion on the premises of the Shell Oil Company Refinery at Norco, Louisiana, on May 5, 1988, and who or which own property, or operated businesses, or were physically present within the area encompassed by the jurisdictional limits of the United States District Court for the Eastern District of Louisiana, at the time of the explosion,

Subclass B is defined as:

Those persons having claims for injuries to or death of employees at Shell Oil Company, or contractors thereof, sustained in the

Subclass A includes in excess of 18,000 claimants.5 Subclass B has sixteen Shell employee claimants.6 The district court established notification and opt-out procedures and approved a Plaintiffs' Legal Committee to represent the class.

The district court identified as liability issues common to both subclasses the determination of fault: (1) as it relates to compensatory damage claims, and (2) whether it is sufficient to warrant imposition of punitive damages. As to Subclass B only, the court identified as additional issues: (1) whether the fault of Shell Oil or any other person claiming benefit of workers compensation immunity was intentional thus obviating the immunity, and (2) whether punitive damages are available if workers compensation is the exclusive remedy.7 The district court thereafter established a procedure for identifying absent class members and obtaining information relating to their claims.

After extensive briefing by the parties, the district court

course of their employment and caused by the explosion on the premises of the Shell Oil Company Refinery at Norco, Louisiana, on May 5, 1988, to the extent that such claims may be subject to the exclusion of the remedy of the Louisiana Workman's Compensation Act. 5 Shell has conceded negligence liability under La. Civ. Code art. 2316 to any member of subclass A who proves damages legally caused by the May 5 explosion. 6 See In re Shell, 136 F.R.D. at 590 n.3. Shell informs that persons within Subclass B have brought fourteen personal injury claims and six wrongful death claims. 7 See In re Shell, 136 F.R.D. at 590.

issued orders detailing a four-phase plan for trial.8 In Phase 1 a jury would determine common issues of liability.9 If the jury found punitive damage liability it would then perform the Phase 2 function and determine compensatory damages in 20 fully-tried sample plaintiff cases.10 Based on the findings in these cases, the jury would then establish the ratio of punitive damages to compensatory damages for each class member. If the jury finds no punitive damage liability in Phase 1, Phase 2 is to be omitted.

In Phase 3, a different jury is to resolve issues unique to each plaintiff's compensatory damage claims, e.g. injury, causation, and quantum. Phase 3 calls for trials in waves of five, scheduled according to a format based upon factors,11 including location of the injured person or property at the time of the explosion and extent and nature of the damages. The district court anticipates that "after several waves are tried, a reasonable judgment value for each category of claims would emerge so as to

8 See id. at 593-96. 9 We previously approved this mass tort case procedure in Jenkins v. Raymark Inds., Inc, 782 F.2d 468 (5th Cir. 1986). 10 Under the Plan the district court would select a group of 100 claimants at random. The three parties would then designate claimants they would accept as having "representative claims." The first 20 three-way matches would serve as plaintiffs in the punitive damages trials. Should the parties fail to agree to twenty from the first group, the district court would select at random additional groups of 100 from the remaining claimants. 11 The district court apparently intends to employ a court selected statistician to analyze the damage claims for the purpose of establishing these groupings.

facilitate settlements."12 In Phase 4 the district court is to compute, review, and award punitive damages, if any are established in Phase 1, for the plaintiffs awarded compensatory damages.

Based on the district court's certification under 28 U.S.C. § 1292(b), Shell and Brown & Root timely sought leave for an interlocutory appeal which we granted.

II. Analysis

We revisit the problem of mass tort litigation recently addressed.13 The instant litigation, involving claims by more than 18,000 plaintiffs, starkly presents the nearly insurmountable problems of balancing procedural fairness with judicial efficiency in the management of mass tort litigation. At the threshold we must note that in many respects this appeal presents only the broad outlines of the district court's trial plan and, to a large extent, appellate review must await its implementation. Keenly mindful of the magnitude of the mass litigation problem, its increasing frequency, and the need for innovative solutions, we review the present challenges to the district court's orders. A. The Trial Plan: Punitive Damage Concerns

1. Applicability of Fibreboard

Shell and Brown & Root first argue that Phase 2 violates principles enunciated in In re Fibreboard Corp. In that case the

12 In re Shell, 136 F.R.D. at 596. 13 See, e.g., In re Fibreboard Corp., 893 F.2d 706 (5th Cir. 1990); Jenkins, supra.

panel reluctantly vacated a trial plan in mass tort litigation involving the claims of 3,031 plaintiffs asserting asbestos-related injuries. The dispute in Fibreboard centered on the aspect of the plan that called for a jury to ascertain damages for the entire class on the basis of a trial of the specific claims of eleven class representatives, together with such evidence as the parties presented about the claims of thirty illustrative plaintiffs, and the testimony of experts about damages to the entire class. We found the Fibreboard scheme infirm for two reasons. First, the proposed plan failed to require each claimant to prove both causation and damages, as required by Texas law. Second, because the proceeding was to ascertain damages for a group of claimants who suffered widely divergent injuries essentially on the basis of a statistical profile, the plan failed to qualify as a "trial" in the sense contemplated by Article III of the Constitution, and was thus beyond the authority of an Article III court. We find the instant case distinguishable from Fibreboard because the Phase 2 jury is to make a determination about punitive damages in a mass-disaster context, rather than compensatory damages in products liability litigation.

The law permits punitive damage awards primarily to punish the defendant guilty of egregious misconduct and to deter such conduct in the future.14 It need hardly be emphasized that the punitive

14 See, e.g., Pacific Mut. Life Ins. Co. v. Haslip, ___ U.S. ___, ___, 111 S. Ct. 1032, 1042, 113 L. Ed. 2d. 1, 21 (1991); Creamer v. Porter, 754 F.2d 1311, 1319 (5th Cir. 1985) (federal law); Karavokiros v. Indiana Motor Bus Co., 524 F. Supp. 385, 387 (E.D. La. 1981) (citing Restatement (Second) of Torts § 908

damages inquiry -- unlike that for compensatory damages -- focuses primarily on the egregiousness of the defendant's conduct.15 As the trial court aptly noted, the degree of culpability underlying a single act -- and hence the propriety of imposing punitive damages as a result of that act -- should not markedly vary in a setting such as is here presented, when considered with respect to different plaintiffs. Because of this minimal variance, assessing the propriety of punitive damages on the basis of the claims of a cross-section of the plaintiff class should not, in the words of Fibreboard, require "lift[ing] the description of the claims to a level of generality that tears them from their substantively required moorings."16 That the Phase 2 jury will consider only punitive damages in a mass tort case materially distinguishes this case from Fibreboard.17

(1965)); Sharp v. Daigre, 564 So. 2d 303, 303 (La. 1990) (dissenting opinion); Creech v. Aetna Cas. & Sur. Co., 516 So.2d 1168, 1173 (La. App. 1987), writ denied, 519 So.2d 128 (La. 1988); W. Page Keeton, et al., Prosser & Keeton on Torts, § 2, at 9 (5th ed. 1984). 15 See Jenkins, 782 F.2d at 474. 16 Cf. Sterling v. Velsicol Chemical Corp., 855 F.2d 1188, 1197 (6th Cir. 1988) (issue of defendant's liability properly resolved on class-wide basis where single course of conduct identical for each plaintiff caused disaster). 17 Shell accurately notes that the Phase 2 jury will, in making its punitive damages determination, consider a "statistical profile" of the claims asserted by the entire class. However, we have previously recognized that ascertainment of punitive damage liability in a class action on the basis of evidence concerning the claims of representative plaintiffs and statistical information about the entire class, before litigation of individual damage claims, presents no constitutional infirmity where the court adequately apprises the jury of the nature of the information before it. See Jenkins, 782 F.2d at 474. Because

More importantly, the Phase 2 jury is not to extrapolate punitive damages but, rather, is to determine a basis for assessment of punitive damages in the form of a ratio. One might argue that the logic of Fibreboard, if not its narrow holding, prohibits use of the Phase 2 procedure to determine quantitatively the amount of actual punitive damages. But Phase 2 purports to do no such thing.18 Unlike the plan in Fibreboard, Phases 2 and 3 appropriately enforce the Louisiana law requirement that a claimant must prove both causation and damage to recover compensatory and punitive damages.19

2. Applicability of Haslip

Shell and Brown & Root also claim that Phase 2 runs afoul of the latest Supreme Court teaching on punitive damages, Pacific Mutual Life Insurance Co. v. Haslip.20 Essentially reiterating

Fibreboard involved use of statistical profiles for quantitative assessment of compensatory damages rather than for determination of a basis on which to assess punitive damages, we cannot conclude, as Shell urges, that Fibreboard overruled Jenkins. 18 Notably, the district court's plan, in compliance with Fibreboard, 893 F.2d at 711-12, permits no extrapolation of individual compensatory damage claims: the parties must try in Phase 3 any claims not settled. 19 The similarity of each plaintiff's claim for punitive damages further underscores this point, because the proof offered in Phase 2 should apply to the punitive damage claims of all plaintiffs. To the extent that class members' punitive damage claims differ, we note the generality of the plan, and the potential for further refinement when the district court implements it. Such refinement might, for example, take the form of setting different ratios for different types of claims. 20 Supra, n.14.

their Fibreboard arguments, Shell and Brown & Root claim that because the Phase 2 plan determines damages on the basis of class representation and extrapolation it violates the Haslip due process requirements. Shell also argues that the Phase 2 plan violates the rule that punitive damages must bear a reasonable relationship to compensatory damages.

Shell and Brown & Root at best present premature Haslip concerns. Haslip, while not a class action or a case purporting to address the concerns which might arise relative to punitive damages in a case involving more than 18,000 compensatory claims, does stand for the general proposition that a punitive damage award by a properly instructed jury, where there is adequate post-verdict review, will not violate due process.21 In addition to recognizing the fundamental purpose of punitive damage awards -- to punish the defendant and deter future misconduct -- Haslip appears to require that the award have a reasonable basis in the conduct and degree of fault of the defendant, and an understandable relationship to compensatory damages.22 We cannot, at this early stage, conclude that the plan at bar will not satisfy these criteria. The proposed procedure does not provide for the precise mechanisms of the Phase 2 punitive damage trial nor does it detail the Phase 4 judicial review. However, the absence in Louisiana law of a scheme for review of punitive damages awards such as that approved in Haslip

21 See id. at __, 111 S. Ct. at 1044-45, 113 L. Ed. 2d at 20- 22. 22 See id. at ___, 111 S. Ct. at 1045, 113 L. Ed. 2d at 22.

should not impede the district court's Erie-mandated effort to act as a dutiful Louisiana trial court mindful of the Supreme Court's teaching in Haslip. We hold that Phase 2 of the instant plan, on its face, adequately satisfies Haslip's command. B. Phase 3 Trial Rules and Procedures

Shell and Brown & Root maintain that the Plan is constitutionally unsound because the district court intends to limit traditional trial rules in Phase 3. The district court indicates that Phase 3 will "not necessarily [involve] full-blown trials," and that "traditional trial procedures, methods of proof, and evidentiary rules will be abbreviated and simplified to shorten trial time.23 Further quoting Newberg on Class Actions, the trial court states that, in class actions, "[p]leadings, discovery, and strict application of rules of evidence associated with normal adjudication processes for individual lawsuits are often replaced with greatly simplified, informal procedures, often summary in nature . . . ." Appellants insist that this language evinces an intent to limit unduly the application of the Federal Rules of Civil Procedure and Federal Rules of Evidence in Phase 3 proceedings.

At this point we can only speculate about how the district court will fill in the broad outlines of its plan in Phase 3. Such speculative concerns do not, however, present an issue ripe for

23 In re Shell, 136 F.R.D. at 596 (citing H. Newberg, Newberg on Class Actions, §§ 9.63, 9.64).

review at this time. While we do not read the plan, as a whole, as indicating the district court's intent to act impermissibly, we simply remind all that the federal rules have the force of law.24 The secondary source quoted by the district court offers no credible support for the proposition that our rules of evidence and procedure may be altered or diminished in any manner, in actions of this kind, other than those recognized to be within the sound discretion of the district court. We express our confidence that the district court will adhere to acceptable norms in the shaping of the rules to meet the judicial crisis presented by the instant litigation. C. Class Certification

Brown & Root vigorously opposes litigation of the claims as a class action. Relying on the district court's grant of summary judgment in its favor on the strict liability and punitive damage claims, Brown & Root argues that subject matter jurisdiction concerns militate against maintenance of a class action against it, and that such a class action would violate Fed. R. Civ. P. 23. Brown & Root thus suggests that we should sever it from this class action, and permit it to defend the negligence claims in separate proceedings. These contentions lack merit. We review district

24 Societe Nationale Industrielle Aerospatiale v. United States District Court, 482 U.S. 522, 533, 107 S. Ct. 2542, 2550, 96 L. Ed. 2d 461, 477 (1987); 4 C. Wright & A. Miller, Federal Practice & Procedure: Civil § 1030, at 125 (2d ed. 1987); See 28 U.S.C § 2072.

court class certification decisions only for abuse of discretion;25 we find no such abuse here. 1. Federal Subject Matter Jurisdiction -- Amount in Controversy

Brown & Root urges that the Supreme Court's holding in Zahn v. International Paper Co.26 counsels against class certification against it. Zahn teaches that each plaintiff in a class action under Fed. R. Civ. P. 23(b)(3), where subject matter jurisdiction is founded on diversity of citizenship, must independently meet the 28 U.S.C. § 1332 jurisdictional amount requirement.27 Brown & Root contends that the claims of a substantial number of Subclass A plaintiffs fall short of the required amount in controversy because they assert only claims for fright and minor property damage, and because the district court has granted summary judgment for Brown & Root on the punitive damages. Because the Zahn rule will require dismissal of the claims against it by an unknown number of Subclass A members, it argues that we should limit the class action proceedings for that subclass to claims against Shell.

Brown & Root fails to consider four principles which guide application of the ad damnum requirement. The Supreme Court has

25 Jenkins, 782 F.2d at 472 (citing Horton v. Goose Creek Independent School District, 690 F.2d 470, 483 (5th Cir. 1982), cert. denied, 463 U.S. 1207, 103 S. Ct. 3536, 77 L. Ed. 2d 1387 (1983); Boggs v. Alto Trailer Sales, Inc., 511 F.2d 114, 117 (5th Cir. 1975)). 26 414 U.S. 291, 94 S.Ct. 505, 38 L.Ed.2d 511 (1973). 27 As this action was commenced prior to May 18, 1989, the $10,000 amount in controversy requirement in effect before the 1988 amendments to 28 U.S.C. § 1332 applies.

held the amount claimed in good faith by initial pleadings controls the question of amount in controversy.28 Further, it must appear to a legal certainty that the claim is for less than the jurisdictional amount before a court may dismiss for lack of quantum. Third, subsequent events generally will not deprive the federal court of its jurisdiction.29 Finally, when a claim includes compensatory and punitive damages, both must be considered in determining the amount in controversy.30

The complaints in this action seek over $32,750,000,000 in damages -- far in excess of $10,000 then required for each member of Subclass A.31 Further, because Louisiana law permits all plaintiffs proving actual damages to share in any punitive damages award,32 the claim for punitive damages increases the amount in controversy for each class member. There is no suggestion that plaintiffs made their damage claims other than in good faith. There is no record basis upon which such a finding can be made at

28 See St. Paul Mercury Indemnity Co. v. Red Cab Co., 303 U.S. 283, 58 S. Ct. 586, 82 L. Ed. 845 (1938). 29 See id.; Seafoam, Inc. v. Barrier Systems, Inc., 830 F.2d 62, 66 (5th Cir. 1987). 30 Bell v. Preferred Life Assur. Soc., 320 U.S. 238, 64 S.Ct. 5, 88 L.Ed. 15 (1943). 31 A review of the damages sought in the instant action finds one complaint seeking $30,000,000,000 (Watson), another seeking $2,750,000,000 (Cauley), and a third, the "Master Pleadings," seeking unspecified and unlimited punitive damages. The essence of Shell's argument is that 18,000 class members multiplied by $10,000 is only 180 million dollars, an amount well below the total of the good faith ad damnum prayers.

32 See La. Civ. Code Ann. art. 2315.3 (West Supp. 1992).

this time. The district court here has not, in contrast to the Zahn trial court, found to a legal certainty that any plaintiff had suffered damages of less than $10,000. Brown & Root's jurisdictional argument based on quantum fails.

The dismissal of the punitive damages claims against Brown & Root does not alter this conclusion. In Seafoam we found that dismissal by summary judgment of one of plaintiff's claims as time barred did not warrant dismissal of the other for lack of subject matter jurisdiction, even though the remaining claim was for less than the jurisdictional amount. The summary judgment in favor of Brown & Root on the punitive damages issue presents an analogous situation. We therefore conclude that consistent with Zahn, Red Cab and Seafoam, the voiced subject matter jurisdiction concerns do not militate against class certification of the claims against Brown & Root. 2. Numerosity of Subclass B

Pointing to the fact that Subclass B contains only 16 plaintiffs, Brown & Root argues that this subclass fails the numerosity requirement of Rule 23(a)(1). That requirement imposes no mechanical rules,33 turning instead on the practicability of joining all class members individually.34 We previously have noted

33 General Telephone Co. v. Equal Employment Opportunity Comm'n, 446 U.S. 318, 330, 100 S. Ct. 1698, 64 L. Ed. 2d 319 (1980). 34 Zeidman v. J. Ray McDermott & Co., 651 F.2d 1030, 1038 (5th Cir. July 1981) (citations omitted).

that while the number of claimants is relevant to this determination, a court also may consider other factors, including the nature of the action. In the instant case, the district court's class certification includes Subclass B in a larger class of more than 18,000 plaintiffs for the purposes of litigating liability issues with respect to Shell and Brown & Root. Considering the nature of this action, we cannot say that identifying Subclass B for the purpose of litigating the related issue of Shell's liability for intentional tort, amounted to an abuse of the trial court's broad discretion. 3. Predominant Common Issues

Brown & Root, citing Jenkins, urges that the absence of issues common to both defendants requires its dismissal from the class action. Brown & Root misperceives controlling law. The commonality requirement of Fed. R. Civ. P. 23(b)(3) is intended to ensure that the disallowance of individual trials is warranted by a sufficient gain in efficiency.35 Rule 23(b)(3) accordingly requires that "resolution of the common questions affect all or a substantial number of the class members."36 The commonality requirement focuses on the common issues relevant to claims by or against the class members; it does not require that all issues be common to all parties. In the litigation at bar the claims of all

35 Jenkins, 782 F.2d at 472; 7A C. Wright & A. Miller, supra, § 1777, at 519. 36 Jenkins, 782 F.2d at 472 (emphasis added).

plaintiffs require resolution of Shell's liability for punitive damages and of Brown & Root's liability for negligence, both arising out of the same event.37 That the plaintiffs assert different theories against Shell and Brown & Root does not obviate the commonality of issues.

Brown & Root further suggests that the class issues thus far identified will not "predominate" as required by Fed. R. Civ. P. 23(b)(3). In the context of mass tort litigation, we have held that a class issue predominates if it constitutes a significant part of the individual cases.38 The class issues to be determined by the Phase 1 jury form integral elements of the claims asserted by each of the more than 18,000 plaintiffs.39 There can be no serious contention that the district court abused its discretion in determining that these issues predominate for the purpose of class certification.

37 Brown & Root also points to Yandle v. PPG Industries, Inc., 65 F.R.D. 566 (E.D. Tex. 1974) and the Advisory Committee note to Fed. R. Civ. P. 23(b)(3) in support of its proposition that, because mass tort cases often present disparate issues, they are generally inappropriate for class action litigation. These authorities have no application to the instant litigation in which many people suffered injury resulting from a common disaster and seek recovery on identical theories. 38 Jenkins, 782 F.2d at 472. 39 To the extent that Brown & Root argues that the main issue in this litigation is Shell's liability for punitive damages and that the issue of its liability does not predominate, Brown & Root's argument fails to persuade. The plaintiff class has asserted claims for negligence against Brown & Root. The plaintiffs must therefore prove Brown & Root's negligence and even though they may be more interested in punitive damages from Shell than in recovering compensatory damages from Brown & Root, issues with respect to Brown & Root predominate for the purposes of Fed. R. Civ. P. 23(b)(3).

4. Superiority

Brown & Root finally contends that class proceedings are not a "superior" means of litigating its negligence liability, as required by Fed. R. Civ. P. 23(b)(3). Pointing to In re Tetracycline Cases,40 Brown & Root argues that the variety of class issues will confuse the Phase 1 jury. It further suggests that because it will seek contribution from other contractors, the class action is not a superior means for litigating its negligence liability. Brown & Root insists that class litigation will not reduce complexity and will not substantially reduce the number of issues left for decision in the Phase 3 trials. These arguments fail to persuade.

The proposed Phase 1 should not unduly confuse the jury. This litigation differs markedly from toxic tort cases such as Jenkins, Fibreboard, and Tetracycline, in which numerous plaintiffs suffer varying types of injury at different times and through different causal mechanisms, thereby creating many separate issues. The case at bar actually will present fewer and simpler issues to the Phase 1 jury. Further, we cannot find that the trial court abused its discretion in opting to utilize the class action in this case simply because Brown & Root may seek contribution from other contractors. Finally, because of the great import of the class issues to the claims of each plaintiff, we cannot agree with defendants' contention that class litigation will not reduce the number of issues or complexity in the Phase 3 trials. To the

40 107 F.R.D. 719 (W.D. Mo. 1985).

contrary, after the Phase 1 resolution, only causation and damages will remain in each plaintiff's claim against Brown & Root. In light of the massive proportions of this litigation, and the need to reduce the systemic burden it will impose, we cannot conclude that the district court abused its discretion in fashioning this class-litigation format.

III. Conclusion

In Fibreboard we reluctantly issued a writ of mandamus, vacating a portion of the trial plan in that case. In so doing, however, we closed with a salute to the trial judge:

We admire the work of our colleague, Judge Robert Parker, and are sympathetic with the difficulties he faces. This grant of the petition for writ of mandamus should not be taken as a rebuke of an able judge, but rather as another chapter in an ongoing struggle with the problems presented by the phenomenon of mass torts.41

Judge Parker had 3,031 cases consolidated in one action. Judge Henry Mentz has more than 18,000 plaintiffs in the case now before him. We express our admiration for the manner in which Judge Mentz, aided by a very able magistrate judge and equally able trial counsel, has woven our mass tort case law into an acceptable and workable trial plan. We AFFIRM the district court's orders establishing that trial plan and return this case to the district court for further proceedings.

41 Fibreboard, 893 F.2d at 712.