Elizabeth Chamblee Burch
• Featured Article •
2009 Cardozo L. Rev. de novo 168
Download this Article
Mass tort litigation is rife with trade-offs. For instance, plaintiffs’ attorneys need to amass clients to achieve economies of scale and bring effective litigation, but an inventory of clients creates an attenuated attorney-client relationship. That nontraditional relationship tends to make clients ineffective monitors because they have little substantive input over how their attorney handles their case. Plus, attorneys must focus on achieving the best result for their clients in the aggregate. This breeds standard collective action problems including conflicts between the self-interest of the group members and the group as a whole.
When large-scale litigation proceeds outside of formal Rule 23 class certification—as is frequently the case after the Class Action Fairness Act—it lacks the judicial quality control measures that class certification affords. Those measures include appointing class counsel, ensuring a fair settlement, and authorizing attorneys’ fees. Without such measures, the trade-offs, tensions, and problems multiply. Broadly conceived, these problems fall into three categories: agency problems between attorneys and their clients, group problems between plaintiffs and other plaintiffs, and competition problems between plaintiffs’ attorneys and other plaintiffs’ attorneys. At the core of these problems lies the principal tension, the tension between efficiently resolving mass litigation to maximize social welfare on one hand, and aiming to afford individual justice to litigants on the other. Put differently, where we draw the line and what we emphasize in debating these trade-offs largely depends on whether we value individual justice or welfare maximization.
Judge Jack Weinstein highlights this tension in his Cardozo De Novo article, Preliminary Reflections on Administration of Complex Litigations. In reading his Article, two things struck me: (1) the extent to which we perceive welfare maximization and individual justice as a dichotomy and rely on one perspective to explain and justify aggregation procedures, and (2) the need for judicial flexibility and creativity in approaching large-scale litigation, particularly in nonclass aggregation. As to the first point, entrenchment in these two perspectives has led scholars down what are now well-trodden paths to familiar debates. But the very persistence of these debates indicates somewhat of a stalemate rather than a solution. Accordingly, in my most recent article, Litigating Groups, I argued for a third approach derived from moral and political philosophy, as well as social psychology, that focuses on inclusion within the relevant community. This alternative theoretical framework identifies an important potential source for group unity: group members’ beliefs regarding their obligations to other group members. It is through implementing this new approach that I address the second aspect of Judge Weinstein’s article, the need for creative handling.
This brief Response thus divides into two parts. Part I addresses the tension that Judge Weinstein observes in his opening paragraphs between “the somewhat academic search for perfection in achieving due process, development of substantive rules of law, and the court’s decision to meet the guideline of Rule One of the Federal Rules of Civil Procedure,” which requires “just, speedy, and inexpensive determination of every action.” Rather than addressing this tension through either traditional lens, Part I sets forth the theory and rationale behind an alternative “group responsibility” approach to nonclass aggregation. It provides a condensed, easily accessible excerpt of the more extensive theoretical framework developed in Litigating Groups.
Part II responds to Judge Weinstein’s use of alternative judicial procedures, which he highlights with examples from Agent Orange, Asbestos, Diethylstilbestrol (DES), Tobacco, Breast Implants, Guns, Zyprexa, and the New York Staten Island Ferry case. By employing creative approaches, advancing equitable concepts used by medieval institutions, and tailoring his methods to fit the unique aspects of each case, Judge Weinstein worked both with and without Rule 23. Drawing on similar equitable concepts and, in many ways, coming full circle to the medieval picture of the “community of the vil” painted by Stephen Yeazell, Part II provides preliminary details on implementing the theory described in Part I.
I PLAINTIFFS’ GROUPS: MEMBERSHIP AND OBLIGATIONS
In addressing the group problems, competition problems, and agency problems produced by large-scale nonclass litigation, scholars and judges typically emphasize either the need to regulate conduct efficiently and deter wrongdoing to maximize social welfare or the need for individual autonomy and consent. Judge Weinstein, for example, tends toward the former group, with his opinions and scholarship reflecting both his pragmatic and his humanist side. In Litigating Groups, I suggested a third approach based on group responsibility. It contends that groups of plaintiffs may have (or could be encouraged to develop) organic or indigenous origins such that social norms and moral obligations provide an internally coercive force keeping litigants together. It thereby mitigates the group problems that arise when some plaintiffs want to withhold consent—or “holdout”—and thus derail a settlement agreement that is in the group’s best interest.
By conceiving plaintiffs within large-scale litigation as a community of sorts, we can draw upon an alternative source of obligations: group members’ obligations to one another. Assuming that plaintiffs actually form or could form a social group—and there is ample evidence that they already do this to some extent—then social psychology empirically shows that they will demonstrate positive “other-regarding preferences.” They will change their views about distributive and procedural justice such that they are no longer principally concerned about achieving the best result for themselves, but for the collective group. Moreover, group members develop and adhere to social norms such as promise-keeping, compatibility, social agglomeration, and the desire for means-end coherence. These norms provide the social glue keeping the group together and fostering cooperation. But if we are inclined toward this alternative idea or find it worth considering further, then we must determine what constitutes a litigation community, what obligations follow from membership in that community, and how to foster group cohesion.
To determine what constitutes a litigation community, I use a flexible umbrella term: “plural subject.” Put simply, a plural subject is an instance where multiple individuals—a set of “I’s”—become a single, plural subject—a “we.” As an umbrella term, what makes plaintiffs a plural subject can vary greatly: litigants might share the same desires, interests, or commitments that certain plans should come to fruition; they might collectively participate in a jofint activity; or they might decide to develop a group policy concerning the litigation. For instance, in corresponding with various plaintiffs involved with the Vioxx litigation, I came across the Merck Settlement Group, an online community committed to public education and to understanding Merck’s settlement offer. The group’s founder and moderator, Al Pennington, describes the group’s activities in plural subject—“we”—terms:
In the first months, we were all united in our efforts to study and understand the settlement. As we began to see the inequities in the settlement, we all agreed that we needed to bring these inequities to the attention of the public in order to get [its] support to stop the settlement and get other plaintiffs to reject the settlement. We all agreed that we had to hit all the blogs we could find and talk with anyone in the media who would listen.
This group thus shared a commitment to jointly understanding the terms of Merck’s offer. After understanding that offer and interpreting it to be contrary to their best interests, the members then dedicated themselves to preventing the settlement offer from becoming the settlement terms.
The Merck Settlement Group is not an anomaly. Plaintiffs regularly band together to create a credible threat to the defendant. Groups may predate the litigation—such as labor unions, veterans’ organizations, community organizations, and even homeowner’s associations. Or, the litigation itself might bring people together as illustrated by the Merck Settlement Group, Asbestos Victims of America, and the Dalkon Shield victims’ organizations. As one might imagine from these examples, plural subjects vary greatly in their degree of cohesiveness. Thus, I use the labels “shared cooperative activity” and “shared goals or policies” to describe stronger, more cohesive, subsets. The more cohesive the group, the more likely its members are to incur moral obligations to one another not to opt out of their shared endeavor. Thus, voluntary commitments form the basis for obligations to other group members. Those commitments and intentions similarly define group membership.
Once plaintiffs decide to do something together such as collaborating on discovery requests or sharing scientific research on causation, they might decide to work together on other things as well. They might bargain or negotiate about how to best accomplish a desired end or even develop an overarching goal or policy to guide their deliberations. But they might not. The group may fall prey to disruptive forces and disagree on key issues. Discussing issues during group formation might turn minor variances into major rifts. Or, some individual outliers may have no interest in becoming group members at all—either because of ignorance or deliberate choice.
Thus, the key questions become (1) what to do with the holdouts, the dissenters, those who join the group but then want to exit; (2) what to do with the outliers, those who have never joined the group or considered themselves group members; and (3) how to mitigate competition among the inevitable subgroups that form within large-scale litigation. Answering these questions necessitates an understanding of how and when members are obligated to one another. Isolating these questions also requires that we initially make a few significant assumptions: that the proposed settlement is objectively fair (a slippery term with multiple meanings), that the litigation has reached a point where collaboration and unity among plaintiffs is desirable, and that one can bracket the nature and purpose of the tort system and use a pluralist perspective.
Beginning with the first question—what to do with the strategic holdouts—requires that we assume that settlement is objectively fair to all of the plaintiffs. Although it is possible that holdouts will experience a change of heart after strong encouragement or pressure from the group, the more controversial question is when both holdouts and outliers must become or remain group members. Put differently, when are litigants morally or legally obligated to participate in and assent to the settlement?
The legal answer is easier in the class action context. In Rule 23(b)(1) and 23(b)(2) class actions, class members cannot opt out of their “shared” endeavor whether they would like to or not. Allowing opt out rights in those kinds of class actions would create inconsistent results or deplete a limited fund, leaving some claimants with nothing. The Fifth Circuit has gone so far as to say that a “presumption of cohesiveness” applies in Rule 23(b)(2) class actions. Less obligatorily, Rule 23(b)(3) class members have a chance to exit from class membership by opting out. They thus incur obligations to one another only by remaining in the class. Even this is a fiction of sorts—remaining in the class is the default. It is unlikely that the class forms a genuine community. Instead, the class action is a legally coercive force to bind people together when the circumstances so require.
Determining when litigants are obligated to one another is even more difficult in nonclass aggregation, like the Zyprexa and Vioxx cases. Without Rule 23 to clarify the group’s boundaries, Judge Weinstein in the Zyprexa litigation and later Judge Fallon in the Vioxx case dubbed them “quasi-class actions” and relied on equitable doctrines to shape the litigation. Quasi class actions help address the agency problems through judicial oversight, but they, like aggregation and consolidation, are externally coercive. The resulting conglomerate is not an organic group; it is judicially and procedurally constructed. Yet, bringing plaintiffs together procedurally fosters opportunities for communities to develop.
Once plaintiffs’ cases are pending in the same forum, it may be possible to avoid further externally coercive measures by stimulating the development of an actual group that evolves organically from relationships, promises, assurances, social networks, commitments, and commonalities. Social and personal norms underlying these connections work hand-in-hand with moral obligations to address the question of when plaintiffs in nonclass litigation are obligated to one another as group members. Although one might adopt requirements for incurring obligations that range from those used in the class action context to the contractual context, this group responsibility approach relies on a stronger requirement formed by promises and mutual assurances.
Under this approach, plaintiffs involved in nonclass aggregation who jointly and voluntarily intend X, who commit to one another through promises or assurances, are morally obligated to act in accordance with that intention provided that no exit conditions to the contrary exist. Obligations in this sense are initially voluntary and evolve from social relationships within the group. Because intentions demand means-end coherence and are thereby semi-resistant to irrational change, group members sharing intentions routinely develop norms of consistency and stability.
The intention, however, is not the binding force giving rise to the obligation. Instead, the promises and assurances—the commitment itself—morally obligate. Promises need not be as explicit as saying “I promise;” rather, as evidence law illustrates, agreements might be tacit or implicit depending on the context. These promises and assurances work together with relevant norms to bind litigants. Once the group establishes its norms—or invokes preexisting personal or social norms—then other members may appeal to the relevant norm in their bargaining and reasoning, particularly if the norm or promise is violated.
This defines membership and explains when and how plural subjects are morally obligated to one another to carry out their joint intentional activity, but does not propose how substantive or procedural laws should reinforce those obligations. It is possible that once joinder mechanisms bring people and cases together, group development will occur and social and group norms will prove sufficiently cohesive such that legal coercion is unnecessary. Although this Response focuses on this possibility, we should begin to think about when and under what circumstances the law should reinforce moral obligations once a certain level of moral interconnectedness is present.
Discussing when litigants are morally obligated to one another partially answers the question of what to do with holdouts—those dissenters who are initially part of the group but then want to exit or to withhold their consent to a settlement agreement. Moral conditions suggest that holdouts who have made promises and assurances to other group members are obligated to keep their promises. If they defect, then other members can appeal to the norm of promise-keeping or alternative relevant norms to encourage compliance. Granted, problems of accurately identifying promises, of reading litigants’ fluctuating intentions and mindsets, and of enforcing amorphous commitments arise. As mentioned shortly, one possibility is to make those obligations explicit and legally binding through an intra-claimant governance agreement.
II FOSTERING GROUP DEVELOPMENT AND COOPERATION
In some ways, relying on group and social norms to enforce moral obligations puts the cart before the horse. To answer more fully the question of what to do with holdouts and outliers (those plaintiffs who do not consider themselves group members) and to address competition between plaintiffs’ groups, we must first determine how the judicial system should foster group development. This Part thus suggests: (1) using special officers to promote goal identification and mediate differences among subgroup members, and (2) using intra-claimant governance agreements to memorialize commitments and establish group decision-making procedures. While these remain preliminary observations subject to further development and revision, they retain ample flexibility for tailoring process to the unique circumstances presented by various mass torts. As illustrated by Judge Weinstein’s array of cases, there is no effective one-size-fits-all approach. Thus, any theoretical framework must be elastic enough to accomodate judicial tailoring. Accordingly, this Part highlights the nuts and bolts of cultivating group development and group cohesion, thereby tilling the soil for the resulting group to develop obligations to one another.
Findings from social psychology and even evolutionary biology suggest that once people view one another as group members, their principal concern is no longer the self, but the group’s collective welfare. Thus, once we realize that plural subjects incur moral obligations to one another under certain circumstances and that group members exhibit prosocial behaviors, then we should explore ways that the judicial system can facilitate (or at least not inhibit) organic group development.
As noted, when the group’s existence does not predate the litigation, procedural aggregation frequently brings litigants’ cases to the same forum before the same judge. Although that clustering effect does not, in and of itself, translate into a community, the proximity might catalyze group formation. Variables stimulating group formation and cohesion include physical and social immediacy; recognized homogeneity of goals, norms, values, or intentions; and shared experiences. Destabilizing influences include competition, heterogeneous claims and damages, few shared life-defining experiences, few overlapping intentions, disparate and incompatible litigation goals, and greater geographic dispersion. Consequently, bringing litigants’ cases together in the same forum can create the physical or social proximity that dispersed mass torts—such as product liability suits—typically lack. For instance, the Merck Settlement Group, the Asbestos Victims of America, and the Dalkon Shield victims’ organizations all coalesced after coordinated pretrial handling.
Still, many litigants—even those within groups that predate the litigation—sue the same defendant without intending to do so cooperatively. Take, for example, the Vietnam veterans’ organizations (Agent Orange Victims International, Citizen Soldier, and Vietnam Veterans of America), which divided ideologically over leadership, politics, and views about the Vietnam War. Competition with one another for veterans’ allegiance resulted in fractured organizational and litigation efforts. Large-scale nonclass litigation is particularly likely to experience this kind of subgroup formation and subsequent competition over resources, litigants, strategies, and goals. Accordingly, as Judge Weinstein recognizes, bringing litigants together procedurally is only the beginning.
Once their cases are aggregated, plaintiffs might: (1) realize that they share litigation goals and that collaborating results in cost-savings and more credible threats, or (2) divide or stay divided as did the veterans’ groups involved in Agent Orange. The strength of this first proposition is that it capitalizes on the benefits of other-regarding preferences and thereby makes holdouts less likely. Giving litigants an opportunity to cooperate with one another and to recognize agreement on certain litigation activities can strengthen group cohesion. In fact, one of the most robust findings in the social psychology literature is that when a group discusses a dilemma, they are substantially more likely to cooperate with one another. Once plaintiffs cooperate on one activity, they might decide to collaborate on other litigation-related matters; they might find that group cohesiveness and their normative story go hand in hand. Accordingly, facilitating opportunities for litigants to determine and discuss their litigation ends and move from amorphous intentions to practical goals may help them acheive a better result than striking out on their own.
Designing opportunities for group deliberation fosters intra-group stability and increases cooperation by eliciting social norms such as promise-keeping, compatibility, non-abandonment, and means-end coherence. It also provides an occasion for a leader to interpret and translate the situation into a familiar schema, to ensure procedural fairness, and to invoke the appropriate social norm: cooperating is the “right thing to do,” one should “keep their promises,” or “we should defect and opt out.” Allowing the group to design and implement its own procedures—perhaps through an intra-claimant governance agreement—enhances cooperation, judgments about the procedure’s fairness, and judgments about the fairness of the substantive outcome. Giving participants an opportunity to voice their opinions about the decision-making and allocation process fosters belongingness and cooperation.
Discussing a dilemma facing the group in and of itself raises concern and awareness for fellow group members. Plus, the personal norms associated with promising regularly compel promisors to follow through with their commitments. Even though the default presumption—even in newly formed groups—is to cooperate and trust the others involved, binding pledges to cooperate are even more effective. Consequently, an intra-claimant governance agreement that not only memorializes the decision-making procedures but also the promises elicited as the result of communication, compromise, and assurances, may maximize prosocial cooperative behaviors. In short, deliberation and mutual assurances might entice the would-be holdout to develop other-regarding preferences such that her concerns extend to achieving equity for the collective group. Moreover, these opportunities may similarly afford outliers—those plaintiffs on the outskirts of the litigation who are not group members—the chance and incentive to join the group, gain common knowledge, collaborate on litigation goals, and form reciprocal obligations.
Turning now to the second proposition—the idea that subgroups could polarize and compete with one another—requires us to recognize that at some point competing groups in large-scale litigation need to connect and coalesce, the litigation clusters need to be smaller, or we need to design exit mechanisms. After all, in the class action context, objectors, opt-outs, and subclasses are common. Further, as the class context illustrates, it is possible for group dynamics to produce anti-social behaviors. In these instances, emergent group behavior may need to be managed based on our normative goals. Accordingly, I focus here on the possibility of connecting and coalescing competing subgroups, though I remain open to the idea of smaller litigation clusters—“polycentric litigation”—and to designing exit mechanisms.
By emphasizing the subgroups’ commonalities and minimizing their distinctions (called “salient social categorization”), a “special officer” appointed by the court to mediate between feuding plaintiffs’ factions could encourage plaintiffs to reconceptualize the aggregate as one superordinate group. In this way, special officers could serve as a go-between for competing plaintiffs. Because plaintiffs’ attorneys regularly (and appropriately) act as zealous advocates, most of the information that plaintiffs receive is filtered through that lens. Accordingly, plaintiffs often lack the information necessary to make fully informed judgments. Thus, a special officer could provide claimants with missing information, such as the strength of their claims vis-à-vis others. Moreover, the special officer could work with competing subgroups to design collective decision-making procedures that detail the conditions and circumstances for exit and voice.
Bringing litigants together in small discussion groups humanizes the process by giving them an opportunity to meet one another. For instance, Francis McGovern served as a special master in the Rhode Island nightclub fire that killed one hundred people and injured more than two hundred. By holding twenty-one group meetings with over three-hundred victims and their families, plaintiffs could see those who were severely disfigured, those who suffered from losing a loved one, and those who were less directly impacted. By asking each person about their preferred litigation outcome and presenting various compensation distribution models used in past disasters, he fostered group deliberation and cohesion.
Similarly, Judge Weinstein appointed Ken Feinberg as a “settlement master” in the DES cases and used Feinberg and David I. Shapiro as special masters in the Agent Orange litigation. Ken Feinberg also acted as the special master in the September 11 Compensation Fund, where he traveled to different jurisdictions across the country and conducted countless town hall meetings in schools, community centers, and hotels. This afforded victims some degree of procedural justice by allowing them to participate in an on-going discussion and to tell their story. Feinberg credits this aspect of the process as “the essential reason that the program was so successful.” The 97% participation rate tends to confirm his claim.
These regional discussions make group formation, discussion, and cooperation feasible in both small-scale and large-scale litigation. The idea of using a special officer in this way simultaneously retains the necessary flexibility to accommodate the unique aspects of each mass tort. Of course, the idea of using a special officer, settlement master, or special master is not new. What is different, however, is using a special officer to mediate solely or at least principally between feuding plaintiffs’ groups and to facilitate collaboration and development of group decision-making procedures that might be embodied in an intra-claimant governance agreement.
Encouraging groups to discuss the dilemmas facing them fosters group cohesion. And cohesive groups provide a more durable solution to the challenges inherent in collective litigation, particularly client-client and attorney-client conflicts. This focus on group cohesion reallocates power to the plaintiffs themselves. Unlike most scholarship on this topic, this Response does not concede or assume that the attorney rightly acts as the fulcrum in aggregate litigation. The attorney’s role and the power imbalance it creates is the principal cause of many of the conflicts in nonclass aggregation. And yet, much of aggregate litigation would never begin without entrepreneurial plaintiffs’ attorneys acting as private attorneys general. Consequently, strengthening group cohesion post-aggregation can restore the tether between the group and its agent and better situate the group to monitor the litigation. Allowing the group itself to design and implement fair procedures for group deliberation, collective decision-making, and allocation provides procedural justice components that large-scale court-based litigation typically lacks.
It also suggests, at least in some ways, a return to the communal obligations from which the class action emerged. As developed by Stephen Yeazell, the modern-day class action can be traced back to medieval guilds, where community members shared collectively in the duties, obligations, and privileges of “villeinage” membership. All community members were jointly liable for any duty that was principally assigned to just one person. The courts regularly imposed collective liability on villages for these obligations regardless of who bore individual responsibility. Of course, with the increase of technology and mobility, our modern-day society often lacks the interpersonal relationships that form community bonds. But sociation still occurs—it ranges from simply walking together, to developing friendships, to building families, to less personal activities such as staying in the same hotel. Thus, it seems, at least in some ways, that the way forward calls us to modernize, strengthen, and reinvigorate certain community ideals inherent in medieval times.
In sum, even if we disagree about when obligations arise or how to enforce the content of those obligations, I urge us all to think about how communal and group obligations can and do impact litigants. This alternative way of thinking means that the content of litigants’ rights and duties may not depend solely on what maximizes the general welfare or what preserves litigants’ individual autonomy, but also on what follows from plaintiffs’ membership in a particular group.
CONCLUSION
I share Judge Weinstein’s pessimism that appellate courts are hostile to class actions and other devices for efficient administration of mass litigation, but I am less convinced than he that administrative agencies are the right way forward. As I have written in the past, administrative agencies introduce a new host of problems including the stigma of being less legitimate than their judicial counterparts. Moreover, so long as mass torts continue to be used as venue for social policy debates, they should not be relegated to the bowels of administrative processing. The system must retain the transparency necessary to serve both its public and private ends.
Although this Response takes issue with Judge Weinstein’s contention that much of complex litigation should “fall on the shoulders of administrative agencies” and instead suggests the alternative proposal of using special officers and moral obligations, difficult work remains in fully formulating that proposal. This account is currently incomplete; the devil is, as always, in the details. Future scholarship must craft feasible procedures to promote cooperation, decide when and how to impose sanctions when norms and moral obligations fail, determine when exiting the group is appropriate, and contemplate how to accomplish that departure in the best way possible. Moreover, the barebones structure as sketched here and more fully developed in Litigating Groups requires further development to take into account non-reciprocal, non-proportional obligations. If we think of a group as an interconnected set of individuals, some of those individuals will have obligations to others within the group that are inapplicable to the group as a whole. Only some obligations are reciprocal, and only some are proportional.
If we begin to think about nonclass aggregation as a system with a panoply of interactions between agents, rules, procedures, and institutions, then we can better evaluate the ripple effect of change. Grappling with what constitutes a litigation community and what obligations flow from membership within that community is just one small piece of the puzzle for institutional design. But it is a critical component. It has the potential to enable the litigants themselves to do some of the heavy lifting rather than requiring hefty doses of judicial coercion. And understanding the system and the group dynamics within it may enable us eventually to design and employ a process that more effectively equalizes litigating power.
de•novo Articles