In this case, we are asked to review a nationwide class certification, accompanied by an injunction against all competing class actions. The plaintiff class is made up of landowners whose property is subject to railroad rights of way, along which defendant telecommunications companies have installed fiber-optic cables without the landowners' permission. See generally Jeffery M. Heftman, Railroad Right-of-Way Easements, Utility Apportionments, and Shifting Technological Realities, 2002 U. Ill. L.Rev. 1401. We reversed certification of a virtually identical class in Isaacs v. Sprint Corp., 261 F.3d 679 (7th Cir. 2001), observing that differences in state law concerning the scope of the railroads' easements, along with differences in the various deeds themselves, would result in "a nightmare of a class action." This time, though, the class has been certified for settlement only, which the settling parties insist eliminates the complications that made the class uncertifiable in Isaacs. The intervening parties—who represent statewide plaintiff classes already certified in Tennessee and Kansas—argue that the class still fails to meet the certification requirements of FED. R. CIV. P. 23(a) and (b).

Before getting to the matter at hand, we note that this case has had a troubled history. The original complaint was filed in 1999 by certain representative plaintiffs against Sprint Communications and the Union Pacific Railroad, claiming damages for the wrongful installation of fiber-optics cables across their land and seeking class-action status in the district court for the Northern District of Illinois. In 2001 the parties announced that a nationwide settlement was in the works in which all similar claims against Sprint and four other companies not yet named as defendants would be settled. Thereafter, representative class-action plaintiffs in other cases around the country got wind of the deal and intervened in order to object.

After a half a dozen hearings in Chicago, engaging the time of a district judge, a magistrate judge, and a Special Master, the settling parties, apparently not pleased with how things were going in the Windy City because the court seemed to be disinclined to approve the settlement, migrated to the United States district court in Oregon and submitted it there for preliminary approval. In doing so, plaintiff's counsel sent a letter to the judge in Chicago informing him that the settling parties would no longer seek approval of the settlement agreement in the Northern District of Illinois. After one hearing, the Oregon district judge (Judge Ann Aiken), in a decision that hit the nail squarely on its head, dismissed the case on the grounds of "judge shopping." Zografos v. Qwest Communications Corp., 225 F.Supp.2d 1217, 1223 (D. Or. 2002). The settling parties then returned to Chicago for another stab at making their deal stick.

The fact that a settlement has been reached is, of course, relevant. "Confronted with a request for settlement-only class certification, a district court need not inquire whether the case, if tried, would present intractable management problems, for the proposal is that there be no trial." Amchem Prods., Inc. v. Windsor, 521 U.S. 591, 620, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997) (citation omitted). But settlement is not a cure-all: "[The] other specifications of [Rule 23]—those designed to protect absentees by blocking unwarranted or overbroad class definitions—demand undiluted, even heightened, attention in the settlement context." Id. These include the requirement that the class representatives' claims be typical of those of the class and that the representatives will adequately protect the class's interests. FED. R. CIV. P. 23(a)(3), (4). And not just the class as a whole: where there are significant differences among subgroups within the class, "the members of each subgroup cannot be bound to a settlement except by consents given by those who understand that their role is to represent solely the members of their respective subgroups." Amchem, 521 U.S. at 627, 117 S.Ct. 2231 (quoting In re Joint Eastern and Southern Dist. Asbestos Litig., 982 F.2d 721, 743 (2nd Cir. 1992)).

The intervening parties identify several ways in which the settling plaintiffs do not adequately represent the interests of landowners in Tennessee and in Kansas. Those two groups have already been certified as litigation classes in their respective states, and each was on the eve of trial when the district court in Chicago issued its injunction. Additionally, the Tennessee class members have established liability in state court for the taking of their property, see Buhl v. U.S. Sprint Communications Co., 840 S.W.2d 904, 912 (Tenn.1992), and estimate compensatory damages at approximately ten times greater than the upper limit provided by the proposed nationwide settlement. They have also shown that punitive damages may be available for trespass to their property, see Meighan v. U.S. Sprint Communications Co., 924 S.W.2d 632, 641-42 (Tenn.1996), subject to proof at trial.

The nationwide class, in contrast, has not been and cannot be certified for trial— see Isaacs, 261 F.3d at 681-82. The nationwide class plaintiffs thus entered negotiations in what the Amchem court describes as a "disarmed" state, unable to "use the threat of litigation to press for a better offer," Amchem, 521 U.S. at 621, 117 S.Ct. 2231—not a good position from which to represent the interests of parties that do wield such a threat.

The settling parties argue that the intervenors' interests are nevertheless protected. Specifically, the settlement agreement provided that adjustments will be made to the amount of recovery available to landowners in a given state, based on an analysis of that state's law by independent property-law experts. But although that may tend toward a more equitable division of funds, it does not provide the "structural assurance of fair and adequate representation" prior to the settlement itself that Rule 23 demands. Amchem, 521 U.S. at 627, 117 S.Ct. 2231. Law professors are no substitute for proper class representatives. Cf. Uhl v. Thoroughbred Tech. & Telecomms., Inc., 309 F.3d 978, 987 (7th Cir.2002); In re Agent Orange Prod. Liab. Litig., 818 F.2d 179, 185 (2nd Cir.1987) (disallowing the administration of class funds by independent foundation without judicial oversight).

We agree with the intervenors that they are inadequately represented by the settling plaintiffs.^[1] We therefore VACATE the nationwide class certification and the district court's injunction against competing class actions and REMAND the case to the district court for further proceedings. Costs are awarded to the Intervenors.

CUDAHY, Circuit Judge, dissenting.

It seems to me that the majority has entirely lost sight of the benefits of the federal court settlement that has been successfully negotiated here. The development involved here is the laying of a 36,000-mile network of transcontinental fiber-optic cables crossing many states to provide a national telecommunications grid. This installation of fiber-optic cables becomes part of the national communications infrastructure, having an important value for the national economy as well as for national security. Obviously, to the extent uniformity in treatment of affected landowners can be achieved, legal costs and costs of administration (ultimately charged to telecommunications users) can be reduced. The state-by-state treatment favored by the majority is likely to produce a nightmare of complexity, the inequitable treatment of landowners in different states and increased charges to telephone users everywhere. If a similar approach had been applied to the construction of the first transcontinental railroad, the Pony Express might still be galloping along.

The principal point made by the majority opinion is that the interveners have not been adequately represented, for two reasons: A) class counsel were "disarmed" because no federal class action could be certified for litigation; and B) having law professors make adjustments to the settlement amount based on a state law is no substitute for proper class representation. Both arguments are without merit. The majority's first argument hinges on three assumptions, each of which must be true for the majority's argument to succeed. Unfortunately, each of these three assumptions is at best, unfounded and, at worst, simply incorrect.

First, based on the Supreme Court's holding in Amchem, the majority assumes that if class counsel are "disarmed" during the settlement negotiation process, the resulting settlement cannot or should not be approved. Amchem, however, does not stand for that proposition and only raises the issue of "disarmed" counsel to explain one reason why courts must consider commonality and the other requirements of Rule 23(a) and Rule 23(b) even in a settlement class. See Amchem Prods. Inc. v. Windsor, 521 U.S. 591, 621, 117 S.Ct. 2231, 138 L.Ed.2d 689 (1997) ("[I]f a fairness inquiry under Rule 23(e) controlled certification, eclipsing rule 23(a) and (b), and permitting class designation despite the impossibility of litigation, both class counsel and the court would be disarmed.").

Second, even if Amchem did stand for the proposition that settlements by "disarmed" counsel should not be approved, the majority assumes that class counsel should be considered "disarmed" whenever their class could not be certified for litigation in federal court. In Amchem, however, class counsel were "disarmed" not simply because the class could not be certified for litigation in federal court, but because the settlement "exclusively involv[ed] persons outside the MDL Panel's province—plaintiffs without already pending lawsuits." Amchem Prods., 521 U.S. at 601, 117 S.Ct. 2231 (emphasis added); see also id. at 601 n. 3, 117 S.Ct. 2231 ("It is basic to comprehension of this proceeding to notice that no transferred case is included in the settlement at issue, and no case covered by the settlement existed as a civil action at the time of the MDL Panel transfer."). Thus, if the class in Amchem could not bring a federal class action, it was left with nothing except mere speculation about the possibility of filing future actions. In stark contrast, in the present case there have been 12 years of litigation pending in various state courts, some of which have already decided that punitive damages may be sought. See, e.g., Meighan v. Sprint, 924 S.W.2d 632 (Tenn.1996). Therefore, unlike Amchem, in this case, class counsel could exert enormous leverage over the defendants, since the alternative to settlement for the defendants is to spend years litigating the case in many different state court venues, which, according to appellants, might award as much as 3,000 percent more than is being offered in this settlement. See Buhl Br. at 23 ("[T]he plaintiffs' proof in Hord v. Quest is that compensatory damages will be $15-19 per foot, with punitive damages expected to multiply that figure several times."). Thus, defendants' next best alternative to settlement would not be the possibility of no litigation at all, as in Amchem or as the majority opinion seems to suggest in this case, but expensive and lengthy chaos in the various states that the fiber-optic cable has traversed. Far from being the classic disarmed counsel, these counsel have the fabled weapons of mass destruction.

Finally, even if the majority is correct in its first two assumptions (i.e., Amchem means both that (a) settlement by "disarmed" counsel should not be approved; and (b) counsel must be considered "disarmed" simply because their class could not be certified for litigation in federal court), the majority would still need to demonstrate that the present class action could not be certified for litigation. The majority assumes this to be so; however, it provides no analysis as to why and pays only lip service to the district court's findings. This is in stark contrast to Amchem, in which the Supreme Court went through a detailed analysis of why the class in that case failed to meet the commonality requirements of Rule 23(a) and (b). See Amchem Prods., 521 U.S. at 622-31, 117 S.Ct. 2231.

With respect to commonality, the problems presented in Amchem do not apply here. In Amchem, cohesion was missing, because the class included members who were exposed to different asbestos-containing products, for different amounts of time, in different ways and over different periods. Some class members suffered no physical injury, some had only asymptomatic pleural changes, others had lung cancer (some of whom were smokers), others disabling asbestosis, and still others mesothelioma — a disease with a latency period of 15 to 40 years. Indeed, as to some class members, it was unclear whether they were ever exposed, and whether they would ever contract an asbestos-related disease and, if so, which one.

In the present case, there are no disparate personal injuries. Plaintiffs' and all class members' claims arise from defendant's installation and maintenance of fiber-optic cable on railroad rights of way. Any harm rising from that installation has occurred and is capable of being ascertained. All class members also raise the same legal claims. Therefore, the class has sufficient unity for settlement class certification purposes.

The majority apparently believes that such rigorous analysis is unnecessary because of this court's one-paragraph disparagement of a related class action in Isaacs v. Sprint Corporation. See 261 F.3d 679, 682 (7th Cir.2001). However, in Isaacs, we merely found that the district court improperly "certified the case to proceed as a class action before making any of the determinations ... that Rule 23 makes prerequisite to certification." Id. at 682 (emphasis added). Although we suggested that it was "unlikely" that common issues could be found to predominate, we did not rule out this possibility nor did we rule out the possibility that a national class action could be bifurcated or otherwise made manageable. Id. In the present case, unlike Isaacs, the district court made the determination that the Rule 23 prerequisites were met prior to certification. Therefore, the dicta of Isaacs do not obviate the need to analyze whether the Rule 23 prerequisites are properly met in this case or require rejection of the district court's conclusions.

The majority's position has the unfortunate effect of insuring that no settlement can ever be entered in federal court in this case, because any proposed class counsel will be considered "disarmed" despite the obvious leverage counsel may wield in practice. The majority thus forfeits the manifest advantages of a national settlement for a national undertaking.

With respect to the second argument (about adjustments to the settlement by experts), the fact that the settlement agreement may require that law professors make adjustments in the settlement amount based on state law has practically no relevance to the issue whether class counsel is adequate. At worst, the adjustment procedure injects some uncertainty into the settlement agreement. If, for instance, the settling parties had the law professors derive the adjustments ex ante and included these adjustments in the settlement agreement (thus eliminating the uncertainty), there would be nothing for the majority to complain about. But the mere fact that the settlement agreement contains some uncertainty does not make class counsel inadequate. It cannot even be known at this point to whose benefit the uncertainty will accrue.

It is tempting to look for holes in this settlement agreement, because it appears that the intervenors may now be significantly worse off than they might have been in state court. However, because we are ill-equipped to determine the ultimate fairness of any settlement from a substantive standpoint, our job must be to insure that the process under which it was negotiated and approved was a fair one. See Amchem Prods., 521 U.S. at 621, 117 S.Ct. 2231 ("[T]he standards set for the protection of absent class members serve to inhibit appraisals of the chancellor's foot kind—class certifications dependent upon the court's gestalt judgment or overarching impression of the settlement's fairness."). Neither the intervenors nor the majority has presented any evidence to challenge the conclusion that the agreement did not arise from a fair bargaining process. To the contrary, after significant study, the magistrate judge found no evidence of collusive negotiations in this case. Supp.App. at A-36. And, of course, the discretion of the district court is entitled to something more than lip service. Where, as here, the district court has addressed fairness in depth, we should be reluctant to disagree.

In any event, this settlement agreement is not unfair on its face. No judgment has been awarded in any state court, so although plaintiffs may find themselves in a strong position, as Yogi Berra once said, the game's not over until its over. While the land use rights in question no doubt have some theoretical value, the only real value comes from the owner's ability to sell his right to the telecommunication companies to lay such cable. Realistically, these landowners were not going to build a retirement cottage lying three feet below the railroad tracks on their property—and, if they were, it is unlikely that this settlement agreement will prevent them from doing so. The point is that it is not for this court to second-guess the value of these rights and then vacate a settlement agreement as unfair.

Hence, I believe that to cast aside a national settlement agreement fairly arrived at, and not substantively unfair on its face, is to imprudently reject the preferred treatment of this nationwide infrastructure, where it seems class counsel were anything but "disarmed" but could constantly wield the threat of a disorderly recourse to state litigation. It is crucially important that the procedures followed here be suited to the emphatically interstate and national character of this important infrastructure development. I therefore respectfully dissent.