INTRODUCTION 552 I. THE MODERN (B)(2) CLASS ACTION 556 A. Shutts and Money Damages Class Actions 557 B. Amchem and Settlement Classes 559 C. Wal-Mart and (b)(2) Classes 561 II. DEFINING CLASS COHESIVENESS 564 A. Cohesive Remedies 567 B. Cohesive Claims 570 III. COHESIVENESS DOCTRINE'S CONSTITUTIONAL AND MANAGERIAL AMBITIONS 573 A. Preclusion and Due Process 573 1. Claim Preclusion 574 2. Issue Preclusion 583 a. Settlement Classes 586 b. Litigation Classes 589 3. Participatory Benefits of (b)(2) Litigation 592 B. Class Manageability 595 IV. COHESIVENESS ON ITS OWN TERMS 598 CONCLUSION 602
INTRODUCTION
On the evening of September 17, 2017, protestors clashed with police across downtown St. Louis following the acquittal of the officer who shot Anthony Lamar Smith, an unarmed black man. (1) Some protestors were maced and shot with rubber projectiles. (2) At least one protestor's cellphone, which they were using to record the events, was seized and the protestor arrested. (3) Others merely observed. (4) The night was the subject of extensive media coverage, (5) and two months later, a class action lawsuit. In Ahmad v. City of St. Louis, three protestors sought to represent others harmed by the police department's allegedly unconstitutional responses to protest, hoping to secure an injunction to change department practices. (6) Class actions in federal court are generally authorized by Federal Rule of Civil Procedure 23. The Rule permits plaintiffs to represent others in four types of cases, including when "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." (7) Rule 23(b)(2), often shortened to just (b)(2), governs class actions seeking primarily injunctive or declaratory relief. In Ahmad, the district court certified the (b)(2) class because the plaintiffs sought only a set of broad policy changes within the police department rather than money damages. (8)
The Eighth Circuit reversed, holding that the class lacked the "cohesiveness" required by Rule 23(b)(2). (9) First, the court recited the rule in the Eighth Circuit that "because unnamed class members are bound without an opportunity to opt out, a Rule 23(b)(2) class requires even greater cohesiveness than a Rule 23(b)(3) class seeking damages." (10) Next, the court considered the three lead plaintiffs, finding that they all had experienced different allegedly unconstitutional conduct on the night of September 17: "Ahmad intentionally violated state and local ordinances at the time she was maced. Mobley was allegedly recording a protest against police when police detained him and seized his phone. [And] Lewczuk was an 'observer' of, not a participant in, the protests when she was arrested." (11) Given the individual factual circumstances that each plaintiff alleged and finding that they sought "broad injunctive relief that [did] not 'relate only to a single policy regarding one particular injury,'" (12) the court decertified the class.
Like the Eighth Circuit in Ahmad, lower federal courts regularly prohibit (b)(2) certification for classes they find to be insufficiently "cohesive." (13) The cohesiveness requirement generally demands that (b)(2) classes have a degree of homogeneity beyond the "commonality" and "typicality" required of all federal class actions. (14) Commonality and typicality, included by name in the text of Rule 23(a), refer to issues that will be litigated in the class. Commonality, most recently interpreted by the Supreme Court in Wal-Mart Stores, Inc v. Dukes, refers to "the capacity of a class-wide proceeding to generate common answers apt to drive the resolution of the litigation." (15) Typicality requires that "the claims or defenses of the representative parties are typical of the claims or defenses of the class." (16) Cohesiveness goes beyond these requirements by demanding that (b)(2) classes share things in common beyond the ability to generate common answers to drive resolution of the litigation and typicality of claims or defenses. Sometimes courts go further to hold that cohesiveness requires that the common issues in the litigation "predominate" over individual ones, as if the class action was one for money damages. (17) And, like the Eighth Circuit in Ahmad, some courts require even more than that. (18)
The purported reason for a strict cohesiveness rule, which maybe striking to those not well-versed in class action procedure, is usually that it would be unfair to the plaintiffs to bind them to a class-wide judgment. (19) There is a kernel of truth in that explanation: existing doctrine permits some potential for issue preclusion--the binding effect of a judgment on the relitigation of issues actually litigated and essentially decided as part of that judgment--following a (b)(2) class action to affect class members' individual claims. (20) But on its own, that issue preclusion concern cannot shoulder the weight of the cohesiveness doctrine and its broader fears about preclusion and due process in class actions.
Indeed, beyond giving courts exceptionally wide latitude to declare certain classes simply not cohesive enough, (21) the due process justification for the rule (i.e., that significant homogeneity is required in (b)(2) classes in order to avoid unfairly precluding class members' individual claims) makes two key errors. (22) First, the due process rationale underlying the cohesiveness doctrine often conflates claims for injunctive relief and claims for money damages when only the latter requires affording absent class members notice and an opportunity to opt out. (23) Second, the cohesiveness doctrine mishandles what preclusion risks are actually created by (b)(2) litigation, imposing a sweeping limit at the certification stage that, after everything else, fails to fix the very due process deprivations that inspired its creation. (24) Both errors underscore that the cohesiveness requirement is a vestige of an era when (b)(2) class actions could be larger, more heterogeneous, and include money damages. The supreme Court has since curtailed (b)(2), eliminating many features that had raised legitimate concerns about due process for absent class members.
Nevertheless, cohesiveness remains a limit on the (b)(2) class action, which since 1966 has stood as one tool for redressing class-wide civil rights violations. Despite dozens of opportunities over five decades to formulate the requirements of (b)(2) class "cohesiveness"--including one penned by then-Judge Gorsuch of the Tenth Circuit--the lower courts have come to meaningfully different conclusions about the rule. (25) Many scholars disagree on basic premises of the doctrine--many nevertheless think it should not exist at all. (26) What follows is a comprehensive evaluation of cohesiveness as a limitation on the (b)(2) class action.
Part I lays out the context of the cohesiveness rule by way of three significant Supreme Court cases that shaped the scope of the modern (b)(2) class action, including the most recent, Wal-Mart Stores, Inc. v. Dukes. (27) Part II then introduces cohesiveness and the types of class heterogeneity that the lower federal courts have considered beyond the reach of Rule 23(b)(2). The remainder of this Comment elaborates two criticisms of the cohesiveness doctrine. First, Part Ill refutes the prevailing due process rationale for requiring cohesiveness of (b)(2) classes beyond that already required by commonality, typicality, and adequacy of representation. That Part subdivides the due process implications of (b)(2) classes by discussing both claim and issue preclusion before moving to the due process benefits of (b)(2) litigation and, finally, the non-constitutional class manageability doctrine. And second, Part IV argues that cohesiveness is not the proper device to limit (b)(2) classes, regardless of due process concerns.
Cohesiveness, especially the strains that approach or supersede the "predominance" required of (b)(3) class actions, is founded neither in relevant Supreme Court precedent nor in a sound theory of the due process and manageability stakes of injunctive relief class actions. (28) Even more problematic, the types of due process problems cohesiveness purports to address cannot be solved by ratcheting up the homogeneity of (b)(2) classes. Fortunately, there are other tools at courts' disposal to manage the due process risks attendant to (b)(2) litigation that do not require imposing a prohibitively high bar to class certification.
I. THE MODERN (B)(2) CLASS ACTION
Courts, like the Eighth Circuit in Ahmad, require cohesiveness of (b)(2) classes based on assumptions about how class actions for injunctive relief may jeopardize the ability of absent class members to subsequently litigate individual claims on their own. (29) I will reserve the discussion of those assumptions for Part III and focus for now on some of the constitutional and Rules-based limitations that together mark the metes and bounds of a contemporary (b)(2) class. Those limitations have undergone significant revision since the provision was added to the Rules as part of the 1966 amendments. (30) They are the context out of which the cohesiveness rule emerged.
The text of Rule 23(b)(2) provides for a class action where "the party opposing the class has acted or refused to act on grounds that apply generally to the class, so that final injunctive relief or corresponding declaratory relief is appropriate respecting the class as a whole." (31) Important to the modern scope of classes brought under that provision are the three significant Supreme Court cases: Phillips Petroleum Co. v. Shutts, Amchem Products, Inc. v. Windsor, and Wal-Mart Stores, Inc. v. Dukes. (32) Notably, Shutts and Amchem were not (b)(2) cases at all. Shutts instead elaborated a standard for due process in 23(b)(3) money...
