Substantive Equality and Procedural Justice

I.     Introduction

Modern federal civil procedure has an equality problem. The procedural rules governing civil disputes originate from a commitment to formal equality, such that the same rules apply to all types of civil cases. This formal equality, however, masks a secret: formal equality is not the same thing as substantive equality. In the 75 years since the Federal Rules of Civil Procedure went into effect, the composition of the civil litigation docket has changed to the point that the arguably once-justifiable formal equality approach now generates predictably unjust outcomes in many categories of cases.

Other areas of law recognize that formal equality is not the same as substantive equality. For instance, critical race theorists have long challenged the “ideology of equal opportunity” because “racism has contributed to all contemporary manifestations of group advantage and disadvantage along racial lines.”1 Consequently, they suggest substantive equality interventions to end racial oppression.2 Richard Delgado, for instance, argues that instead of being subject to First Amendment protection under formal equality principles, racist insults should instead constitute actionable torts.3 Other commentators criticize the Supreme Court’s 1992 opinion in R.A.V. v. City of St. Paul4 on similar grounds, arguing that the First Amendment should be subject to an “antisubordination” principle that would combat the pernicious effects of privilege and coercion inherent in society.5 Robin West makes a similar argument in the context of feminist legal theory. Drawing upon insights from Catharine MacKinnon and others, West argues that “formal equality . . . targets the wrong evil” because any presumption of “sameness between men and women . . . will be simply irrelevant to the true causes and nature of women’s inequality or will backfire and harm rather than help women.”6

Those concerned with ensuring justice within the civil litigation system face a similar challenge. To most procedural theorists, a system is procedurally just if it produces accurate results at a reasonable cost while ensuring meaningful participation rights for those with business before the civil courts.7 A system is accurate if it reliably and correctly applies the relevant law to the relevant facts.8 Of course, accuracy must be balanced to a certain degree against both time and cost—a perfectly accurate procedural regime would hardly be just if it routinely bankrupted litigants and took ten years to resolve the average case.9 Similarly, there is something inherently and intuitively unjust about a system—however accurate—that denies litigants the opportunity to participate meaningfully in the litigation process.10

Unfortunately, civil procedure has failed to take appropriate cues from substantive equality theorists. As a result, the system is structured in ways that work at cross-purposes to a procedural justice goal. In particular, the Federal Rules of Civil Procedure are broadly “transsubstantive.” That is, with a few exceptions, the same procedural rules apply to all different types of claims and cases.11 Under this transsubstantive regime, then, the Federal Rules of Civil Procedure implicitly embrace a formal equality norm. In simplest terms, formal equality norms all derive from the Aristotelian norm to “treat like cases as like.”12 Consequently, formal equality systems, such as the transsubstantive rules, understandably fare best when either the affected parties are functionally similar along the relevant dimensions, or when it would be particularly difficult for a real-world regulator to identify and remedy the dissimilarities that might otherwise call for a more targeted approach.

This transsubstantive procedural regime, and the formal equality it necessarily creates and lionizes, however, often produces outcomes that cannot be regarded as “just” under any serious theory of procedural justice.13 Put a different way, the current incarnation of transsubstantive procedure actually predictably does exactly what the Rules Enabling Act prohibits: it “abridge[s], enlarge[s], [and] modif[ies]” a variety of substantive rights through procedure.14 Indeed, in order to effectuate the core principle of Federal Rule of Civil Procedure 1—ensuring “the just, speedy, and inexpensive determination of every action and proceeding”—one must seriously consider what procedural justice even means, including how an outcome can be just if the procedural system that produced it predictably yields substantively inequitable results.15

The time has come for civil procedure to move beyond rigid formalism and instead begin focusing on substantive equality.16 Substantive equality norms generally allow regulators to peek behind the curtain to varying degrees to assess the pitch and camber of various dimensions of the playing field and to respond accordingly. Proponents of substantive equality approaches note that facial similarities in status and condition often mask significant and highly relevant dissimilarities so that nominally equal opportunities become functionally unequal. Consequently, a formally equal system, for example, may afford identical rights and treatment to all “persons,” yet it may still be functionally discriminatory when the people invoking those rights face very different paths and obstacles due to their race, gender, or other characteristics. The law might prescribe or proscribe “equally,” but the law’s effects will be felt differentially. Substantive equality theorists thus embrace targeted interventions designed to equalize opportunity when they conclude that nominal formal equality cannot overcome institutional or other impediments to a truly fair system.17 Substantive equality requires that regulators confront this unjust reality head on in ways that formal equality simply cannot.

In short, this Article makes a novel contribution to the literature by showing that, just as in other contexts, equality norms and conceptions of procedural justice cannot be divorced from each other when it comes to the federal civil litigation system. Instead, one must consider how the two interact and inform one another. As Larry Solum has explained, a procedural regime is only just when it strikes the appropriate balance between accurate application of law to facts, cost minimization, and meaningful participation rights.18 In the language of equality theory, accordingly, meaningful procedural justice equates to equality of opportunity. Equality of opportunity need not always be the same thing as equality of result, but it cannot lose sight of the fact that opportunity, in fact, must be meaningfully equal. After all, just as we would hardly regard the procedural system as just if it selected the winners and losers of civil disputes by coin toss rather than on the legal merits of participants’ claims and defenses,19 we also should reject as unjust any system that ignores the reality that defendants in certain types of cases have massive incentives.

Of course, the decision to embrace an equal opportunity norm does not end the inquiry, because equality of opportunity is itself challenging to define. We might choose to describe it in terms of negative freedoms that only protect participants from external obstacles to their goals.20 Under such a view, a jurisdiction that discriminates against racial minorities with respect to property ownership could be converted into an equal opportunity regime simply by repealing all legal barriers to minority property ownership.21 But we might also adopt a means-regarding form of equality that accounts for the fact that different participants have differing levels of resources.22 In that same discriminatory environment, means-regarding equal opportunity might require both the repeal of all relevant legal barriers and steps toward correcting resource disparities that may tend to prevent minority property ownership regardless of the legal regime. We might instead even choose to define “equality of opportunity” in prospect-regarding terms, attempting to level the playing field such that every participant faces equal prospects of attaining the relevant goal.23

Thankfully, the civil litigation context itself offers significant guidance as to the type of “equality of opportunity” to which the rules of civil procedure should aspire. Specifically, the Rules Enabling Act explicitly provides that the Supreme Court (typically acting through its committee rulemaking process) can prescribe only “general rules of practice and procedure” for the federal courts.24 And it expressly prohibits the promulgation of rules that “abridge, enlarge or modify any substantive right.”25 Accordingly, the committee rulemaking process is, or at least should be, confined primarily to the enactment of rules that further a particularly limited form of equality of opportunity—the equal and meaningful opportunity to litigate claims on the merits. Thus, the best path to substantive equality requires committee rulemakers to abandon their commitment to formally equal transsubstantive procedure and instead embrace a careful, realistic appraisal of various categories of federal civil claims. The aim of this appraisal should be the eventual promulgation of procedural rules that balance those incentives such that merit, and not unjust factors like cost asymmetry or procedural gamesmanship, drives substantive outcomes. For instance, committee rulemakers should reconsider the transsubstantive pleading rules announced in Bell Atlantic Corp. v. Twombly and Ashcroft v. Iqbal.26 As explained below, while the “plausibility” standard embraced by the Supreme Court in those decisions might promote just outcomes for certain categories of cases involving asymmetric incentives for the parties, it fosters significant injustice in the categories of cases in which the relevant incentives are more closely aligned.27

In short, modern civil procedure has an equality problem, and continued adherence to a transsubstantivity procedural norm actually undermines the goals of the Federal Rules and the Rules Enabling Act. Effective reform, moreover, is both necessary and possible. But to achieve substantive equality, and thus meaningful procedural justice, it is necessary to look beyond the stagnant model of formal equality that has dominated the discussion of procedure since the 1930s.

This Article thus proceeds as follows: Part II briefly frames modern debates regarding the nature of equality and defines “procedural justice.” Part III traces the history of transsubstantive procedure as a solution to the problems engendered by its patchwork quilt predecessors. Part IV further describes civil procedure’s equality problem, explaining that the Rules framers’ commitment to transsubstantivity was sensible given the economic dynamics of the 1930s “paradigm case” in federal court, but that an increasingly heterogeneous docket undermined the norm substantially over the decades after the Rules went into effect. Part V argues that, by adopting a substantive equality approach predicated upon balancing the intra-case economic incentives facing litigating parties, the committee rulemaking process can resolve much of the equality problem inherent in modern procedure. Furthermore, such an approach can be achieved without exceeding the statutory authority delegated to the courts in the Rules Enabling Act.

II.     Defining “Equality” and “Procedural Justice”

A.     The Meaning of “Equality”

A visitor climbing the 44 marble steps to the west entrance of the United States Supreme Court Building pauses halfway and turns her gaze upward. Above the Corinthian columns bracketing the entrance and below the symbolic sculptural representations of Liberty, Order, and Authority, she reads this solemn promise: “Equal Justice Under Law.”28 The inscription greeting those who enter the Supreme Court has no known source, but it nonetheless captures an essential reality of the American experiment: the rule of law and conceptions of justice are inextricably intertwined with the notion of equality.29

That said, it is extraordinarily difficult to forge consensus regarding the proper definition of equality in any legal context. Most subscribe to some version of moral equality, or what Michael Rosenfeld calls the “postulate of equality.”30 In Rosenfeld’s terms, the postulate of equality is essentially the idea that all persons have equal moral worth.31 But egalitarians of different stripes embrace radically different prescriptive permutations of moral equality,32 and some even question whether the concept of equality has any independent meaning at all.33

Though “equality” is difficult or impossible to define precisely in prescriptive terms, it remains enormously important as a framing device for critically important debates about how best to order society. In particular, the debate between proponents of formal equality—under which the goal is to ensure that all are treated identically—and those embracing substantive equality—under which participants are treated differentially to correct for underlying structural or characteristic differences—continues to animate much discussion within legal scholarship as a whole.34

Even within that broad discussion, however, procedural equality has received far less scholarly attention than broader notions of the concept, even though procedure is a key component of law.35 For example, most would reflexively reject a procedural system that afforded special treatment to some litigants but not others, especially if that special treatment affected litigation outcomes in ways unrelated to the merits of a claim. Equality, justice, and procedure are thus similarly interrelated, and notions of equality lie at the very heart of the procedural enterprise. It may well be beyond our means to define and implement consensus prescriptive equality principles across all substantive areas of law.  But it may be somewhat easier to reach consensus in the context of procedural system design, at least when operating under the constraints imposed by Congress in the rulemaking authority delegated to the Supreme Court under the Rules Enabling Act. To do so, we need to understand the boundaries of the debate.

1.     Aristotle, Formal Equality, and Substantive Equality

In technical philosophical terms, “equality” is defined as “correspondence between a group of different objects, persons, processes or circumstances that have the same qualities in at least one respect, but not all respects . . . .”36 But as commentators from antiquity forward have noted,37 this definition does little to clarify the meaning of “equality” as a prescriptive concept intimately tied to notions of justice.38 To gain purchase on how to think about procedural equality, we must first understand the ways in which notions of equality have evolved since Aristotle first attempted to formalize the concept.

i.     The Aristotelian Origins of Equality Theory

Most current equality debates owe an enormous debt to Aristotle, who may have been the first to formulate a relatively comprehensive theory of equality and whose ideas continue to frame debates over equality even today.39 The semantics have changed in the intervening two-plus millennia, but most modern commentators ultimately continue to accept Aristotle’s proposed relationship between equality and justice—that a just system must treat like cases as like.40 Egalitarians of all varieties typically agree that true justice requires such a commitment; they simply disagree (often quite vehemently) about which cases are truly “like,” and what it means to provide like treatment.41

Aristotle identified two distinct potential forms of equality, and explored the conditions under which each might satisfy the critical “treat like cases as like” precondition for true justice. Under Aristotle’s formulation, an “arithmetic equality” system is one that treats all cases identically.42 As we shall see, Aristotelian arithmetic equality is functionally identical to what modern legal commentators have termed “formal equality.”

As George Abernethy explains, Aristotle distinguished arithmetic equality from what Abernethy terms “proportional equality.” In contrast to the identical treatment afforded by arithmetic equality, Aristotelian proportional equality approaches instead treat persons by reference to desert or merit.43 Thus, while proportional equality still satisfies Aristotle’s justice condition that like cases be treated as like, some form of merit or desert is the characteristic that determines likeness.44 Aristotelian proportional equality thus calls for the distribution of like portions to all persons of like merit or desert.

For the most part, Aristotle endorsed a proportional equality approach to the distributional challenges facing societies.45 And while his views on the characteristics determining merit or desert are decidedly out of step with modern sensibilities,46 by emphasizing merit over type, Aristotelian proportional equality is the direct lineal ancestor of today’s substantive equality.

ii.     Modern Formal Equality

As stated above, what we today describe as “formal equality” is really just Aristotelian numerical equality by another name. For example, Catharine MacKinnon and other feminist legal theorists use the term “formal equality” as shorthand for the contention—with which they vehemently disagree—that all persons are fundamentally equal and should thus be treated identically.47 Critical race theorists like Anthony E. Cook and Kimberlé Crenshaw use the term in the same way.48

As MacKinnon points out, formal equality norms undergird much of U.S. law, especially in the area of constitutional equal protection doctrine.49 This is at least in part because we struggle to implement our descriptive belief in Michel Rosenfeld’s postulate of equality—the generally accepted idea that all persons have equal moral worth and are entitled to equal moral dignity50—in a complex, diverse, and often fractured society.51 While formal equality under law may fall far short of the ideals embodied in our abstract notions of moral equality, it may be at least a reasonable first step in certain contexts. As a general rule, formal equality is both less contestable and more easily and cleanly implemented than its more interventionist cousins. That said, formal equality only actually achieves socially desirable results when the relevant population is truly “like” with respect to the relevant characteristics.52 For instance, a formal equality regime in which no person under the age of eight has the capacity to contract is defensible to the extent we believe that:             (1) mental and emotional maturity are characteristics relevant to the question of whether one has the capacity to enter into binding agreements; and            (2) persons under the age of eight are collectively unlikely to possess the relevant maturity. But simply declaring the entire population “like” does not make it so. We thus might take issue with a law declaring that no person under the age of 25 has the legal capacity to enter contracts; there is simply too much variation among the relevant population, and a formally equal regime that limits contracting to parties 25 and older would be unlikely to capture the proper characteristics. In this fashion, formal equality norms often fail to promote true justice.

iii.     Modern Substantive Equality

Critics of formal equality norms have embraced varying forms of what feminist and critical race theorists describe as substantive equality.53 In most of its forms, modern substantive equality is just a subset of Aristotelian proportional equality. Like Aristotle, substantive equality theorists accept the core equality–justice postulate that like should be treated as like.54 And similar to Aristotle, substantive equality theorists believe that justice requires that we look beyond superficial similarity (e.g., “personhood”) to evaluate some form of desert or merit as the characteristic relevant to the treat-like-as-like inquiry.55 But whereas Aristotle tended to define desert positively in terms of merit or virtue, modern substantive equality theorists instead focus primarily upon characteristics like race, gender, or socioeconomic status that can place certain participants at a disadvantage relative to those possessing different characteristics.56

In broadest terms, then, substantive equality theorists embrace an Aristotelian proportional equality approach, but usually in ways directly contrary to Aristotle’s own intuitions. Aristotle, for example, would not have extended significant rights or distributional bounty to slaves and women, because he believed that they were inferior to free men of noble birth.57 By contrast, modern substantive equality proponents—deeply influenced by the post-Aristotelian idea that all persons are morally equal—would instead put a thumb on the scale in favor of racial minorities, women, and other structurally disadvantaged groups.58 But substantive equality is still Aristotelian proportional equality, since substantive equality theorists would distribute additional resources to minorities and women because they believe those classes of people deserve such benefits.

As a result, substantive equality theorists propose legal and policy approaches that move beyond formally equal legal solutions in favor of affirmative interventions genuinely intended to level the playing field.59 If an African-American woman has the deck stacked against her by virtue of both her race and her gender, it is not enough to simply guarantee her nominally equal rights; the law should instead actively promote her interests in ways that ultimately yield true equality.

iv.     The Advantages and Limitations of Substantive Equality

Admittedly, formal equality has its place in law. If nothing else, near universal Western acceptance of the idea that all persons have equal moral worth regardless of their inherent characteristics or social status suggests that formal equality is often a good starting point for legal regimes. Compared to predecessor theories like hereditary aristocracy, a commitment to formal equality has much to recommend it.

But formal equality is still tragically insufficient in many cases. Substantive equality proponents argue persuasively that we cannot treat an entire population as like when significant and persistent structural obstacles effectively prevent entire segments of that population from competing for success on even remotely equal terms.60 When characteristics below the level of the greatest common denominator, such as race, gender, and socioeconomic class, predictably and persistently drive outcomes, it is only reasonable to sort on the basis of those characteristics. And when the outcomes in question functionally reject the core postulate of equality, we cannot make all equal by simply declaring it is so.

Substantive equality approaches thus offer significant benefits over formal equality approaches, provided they retain a commitment to Rosenfeld’s “postulate of equality”—the idea that all persons are inherently of equal moral worth. Properly designed and executed, a system based upon substantive equality principles can remedy the effects of the structural imbalances that tilt so much of the playing field in favor of smaller subclasses, rather than guaranteeing everyone the promises of a just society.

To be sure, substantive equality comes with its own set of risks and limitations. For example, while substantive equality proponents uniformly reject a formal equality vision of equal opportunity,61 they sometimes disagree on the ultimate goal of their proposed interventions—should they pursue equality of outcomes/results or equality of opportunity?62 Those who promote equality of opportunity have to navigate thickets of their own, of course. Specifically, they disagree with one another as to whether substantive equal opportunity should allow for different outcomes because of differences in ability and native talent. Some contend that equal opportunity requires only that participants possess equal means for achieving their goals.63 Others instead argue that equal opportunity demands equal prospects for success, such that interventions must provide those with lesser endowments in ability and native talent proportionally more resources to offset more talented participants’ advantages.64

Along with these more theoretical differences, underlying all of this are the administrative problems inherent in most substantive equality approaches. Unlike many formal equality norms, substantive equality norms often require extensive, potentially subjective evaluation of both the existence and measure of relevant characteristics.65 It can sometimes be difficult to determine who should receive the benefits of substantive equality intervention, and can be harder still to identify the proper extent of that intervention in a complex and resource-constrained world.66

Still, the persistence of structurally-induced injustice militates in favor of exploring the viability of substantive equality interventions in a variety of contexts, including civil procedure.

B.     Procedural Justice and Substantive Equality

While the importance of considering a substantive equality approach to civil procedure is clear, the potential scope of this approach is significantly constrained by the nature of the committee rulemaking process, which is overseen by the Supreme Court under the auspices of the Rules Enabling Act. In earlier work, I described the dynamics of committee rulemaking in detail.67 Although the precise workings of the process are mostly irrelevant to the analysis in this Article, in order to understand the interplay between procedural justice, substantive equality, and the federal Rules, four key features of the rulemaking process bear special mention.

First, committee rulemakers are the primary architects of the procedural system, generating the vast majority of the rules that govern the administration of civil lawsuits.68 Second, the scope of their charge is deliberately limited for separation of powers reasons. Specifically, the Rules Enabling Act authorizes the Supreme Court and its rulemaking apparatus to make only “general rules of practice and procedure,”69 and expressly prohibits the promulgation of rules that “abridge, enlarge or modify any substantive right.”70

Third, the Supreme Court is a critically important player in the process at two levels. The Supreme Court sits atop the committee rulemaking process itself as the final “positive option” gate before proposed rules are submitted to Congress.71 That is, the Supreme Court must affirmatively approve proposed rules. By contrast, Congressional approval of rules is generally a negative option proposition—rules submitted to Congress as a result of the committee rulemaking process generally become law after a certain period of time unless Congress affirmatively rejects the rulemakers’ proposals.72 But the Court also fulfills an ostensibly interstitial interpretive role, providing guidance and clarification as to the meaning and application of the Rule.73 It is therefore doubly appropriate to include the Court in the list of rulemakers involved in the process.

Fourth, although rulemakers are at least nominally subject to congressional override when they overstep their authority, that override must be exercised by way of a negative option. Specifically, unless Congress affirmatively votes otherwise, committee rulemaking has the force of law.74 Given the transaction costs associated with obtaining congressional action in any area, therefore, we should not assume that congressional failure to act equates to tacit approval of the results of the rulemaking process.75 These Congressional transaction costs include both collective action costs and information cost components. That is, Congress must overcome both: (1) all of the difficulties associated with getting any consensus-based group to act; and (2) all of the difficulties associated with obtaining information sufficient to make an informed decision.76 This in turn suggests that Congress will fail to invest resources necessary to fully understand the implications of committee rulemakers’ work, and will have difficulty acting even when it does.77 We must thus evaluate the existing approach to rulemaking on its own terms, rather than assuming that congressional silence indicates congressional agreement that the current framework is working well enough to satisfy Congress that separation of powers principles are being observed.

These four limitations on committee rulemaking authority have significant implications for a substantive equality approach to civil procedure. While Congress may have the right to engage with higher-order substantive equality debates by way of statutory procedural lawmaking, committee rulemakers do not. Rather, the Rules Enabling Act explicitly prohibits committee rulemakers from doing anything that would “abridge, enlarge or modify any substantive right.”78 And if committee rulemakers did possess meaningful substantive policy authority, even wholly altruistic exercise of that authority on a case-by-case basis would do incalculable damage to broader norms of procedural justice and would harm the institutional legitimacy of the entire civil dispute resolution system.79

Specifically, although a single deviation from formal equality—for example, one intended to increase fidelity and fit between the civil litigation environment and the substantive intentions of the legislature—might do more good than harm, we cannot view each litigated case in isolation. The institutional harm from substance-specific committee rulemaking interventions motivated by goals other than maximizing procedural justice would be substantial. Committee rulemakers are generally selected by the Chief Justice of the Supreme Court, and are thus not democratically accountable to the electorate.80 Even assuming committee rulemakers are capable of divining legislative intentions, they lack both the authority and the ability to prioritize costs and benefits that may cut across multiple policymaking dimensions. As a result, substantive equality at the level of the committee rulemaking process must be tethered to notions of procedural justice, rather than a higher-order conception of what constitutes a just outcome.

Importantly, even with these distinctions, the approach this Article adopts is still a substantive equality intervention.81 Under the substantive equality approach this Article recommends, committee rulemakers would still look beyond the demonstrably untrue assertion that all civil cases are created equal. However, separation of powers, the limitations imposed by the Rules Enabling Act, and the importance of retaining and strengthening the institutional legitimacy of the civil justice system dictate that the committee rulemaking apparatus confine itself to interventions that foster genuine procedural justice rather than some broader notion of social justice.

1.     A Working Definition of Procedural Justice

In order to foster substantive equality within civil litigation, a procedurally just system must strike the appropriate balance among three potentially competing ends: accuracy, cost, and meaningful participation rights.82 These foundations of procedural justice are inconsistent with the formal equality norm embodied in the current transsubstantive rules.

i.     Accuracy

The primary, but not exclusive, goal of a just procedural system is accuracy.83 That is, for a procedural regime to be just, it must accurately apply the relevant law to the disputed facts an acceptably high percentage of the time. A procedural system that decides disputes by coin flip might be terribly efficient, low cost, and even fair in the sense that it favors neither party, but it would hardly be just—a coin flip would be accurate only accidentally, if then.84 While perfect accuracy is never attainable, any system that routinely yields inaccurate results is inherently unjust.

It is important to note that Solum’s conception of accuracy is explicitly decoupled from any of the broader social aims that might be associated with the substantive law. Solum persuasively contends that the only type of accuracy relevant to procedural justice is essentially an intra-case accuracy.85 For instance, imagine a procedural system that successfully promotes the underlying substantive goals of, say, antidiscrimination legislation, but does so by favoring discrimination plaintiffs without regard for the legal merits of their individual claims or the merits of defendants’ positions. Such a procedural system might be normatively desirable by certain lights, but it cannot be said to be accurate in any meaningful sense. Nor can its faithful service to some purportedly higher norm be characterized as contributing to procedural justice, whatever its social justice effects.86

As a corollary, to promote the right kind of accuracy, a procedurally just system necessarily must take litigants as it finds them in many ways. Implicit in the concept of accurate application of law to facts is the notion that the facts matter. It simply is not procedure’s role to ignore or bias interpretation of those facts, even in pursuit of some greater truth.

At first blush, the accuracy integral to a procedurally just system might seem to be at odds with some of the concepts underlying the idea of substantive equality. However, that tension is more apparent than real, especially when it comes to the statutorily limited role played by the Supreme Court and the committee rulemaking apparatus. As discussed below, while Congress has the authority to deploy procedure to substantive ends if it so desires, the Supreme Court and the rulemakers acting under its direction cannot legitimately pursue any form of “accuracy” other than “correct application of the law to the facts.”87

ii.     Cost-Balancing

A procedurally just system must also balance its commitment to accuracy against the costs of that accuracy. Like almost everything else, civil litigation is subject to the law of diminishing returns. At some point, the costs of additional expenditures—for example, additional discovery or more elaborate pretrial proceedings—will begin to exceed the benefits. It is certainly plausible that doubling or tripling national expenditures on civil litigation would generate some accuracy improvements; all else equal, a system that expended twice as much on civil dispute resolution than our current system would likely uncover additional relevant facts, help clarify and crystallize the relevant legal issues, and lead to better analysis by both the court and the ultimate trier of fact. In other words, it would be more accurate.

At the same time, however, it is unlikely that those accuracy benefits would be worth the additional costs. As Solum notes, “[d]ispute resolution systems impose costs on the parties . . . and on society at large.”88 Observing that even a procedural system that guarantees perfect accuracy might nonetheless be unjust if the costs of that accuracy exceed the value of the judgment a party will obtain, Solum concludes that the accuracy of the system must be weighed against the costs of obtaining that accuracy. As Solum explains, “[a] dispute resolution system that achieved 100% accuracy would be viewed as monstrously unfair if it required each disputant to devote her entire life to a painstaking process of fact-finding and consumed the great bulk of the social product to finance the enterprise.”89

Solum’s exploration of cost-balancing identifies two possible theories: (1) the intuitively attractive consequentialist approach concerned with the costs and benefits of the procedural regime;90 and (2) a somewhat less obvious deontological approach that will accept less-than-complete accuracy in return for a fairer and rights-sensitive distribution of costs.91 While Solum at least tentatively suggests that a deontological cost-balancing approach might attempt to allocate costs such that “[n]either plaintiffs nor defendants should enjoy an advantage in any particular category of cases” (a position with which this author agrees), he expressly reserves for future work any serious consideration of how deontological cost-balancing would attempt to account for the effects that differential resource allocations between the parties might have upon litigation outcomes.92

The incredibly broad scope of Solum’s project—formulation of a general, abstract theory of procedural justice—entirely justifies his reluctance to engage in resource-balancing cost considerations. This Article does not face the same constraints. Whatever the merits and drawbacks of resource-balancing deontological cost-shifting might be in the abstract, this move is not available to rulemakers under the limited authority Congress delegated to them under the Rules Enabling Act.93 Congress may well be able to take a position on that potential component of procedural justice, but committee rulemakers cannot without running afoul of the spirit—and potentially the letter—of the limitations imposed upon them by statute and broader separation of powers doctrine.

Even with these limitations, however, any coherent theory of procedural justice must temper its taste for accuracy with an appreciation for the various types of costs the search for accuracy inevitably entails.

iii.     Meaningful Participation

A procedurally just regime also guarantees meaningful participation to all disputants, and not simply because that participation might improve accuracy or lower costs.94 A full recounting of Solum’s “Participatory Legitimacy Thesis” lies well beyond the scope of this Article, and is not necessary to the analysis undertaken here. But the idea that participation matters is important, at least in part for reasons unrelated to increasing accuracy of result.

Solum cycles through a number of different potential participation theories in his treatment of procedural justice, but ultimately comes to rest on the proposition that meaningful participation is necessary because it supports the legitimacy of our dispute resolution mechanisms.95 Importantly, this legitimacy-enhancing component of meaningful participation is independent of the merits of one’s claims or defenses. Solum emphasizes that the real litmus test for participation as an independent, legitimacy-enhancing good is not whether one would feel less bound by a “good” decision if it came without the opportunity to participate meaningfully.96 Rather it is whether one would feel more bound by a “bad” decision if it came after notice and a meaningful opportunity to be heard.97

It is critical to remember that procedural justice does not require only participation, but rather meaningful participation. That distinction matters substantially when we begin to consider the intersection of procedural justice and substantive equality.

2.     Is Substantive Equality Compatible with Procedural Justice?

As explained above, a procedurally just system will have as its primary goal accurate application of law to facts, balanced by certain cost considerations, and with an overarching commitment to ensuring legitimacy-enhancing meaningful participation rights. To the extent a substantive equality approach can further those goals more effectively than a formal equality approach, we should pursue substantive equality solutions.

But in doing so, we must take care to avoid the potential excesses of overly enthusiastic substantive equality. In particular, if substantive equality within the procedural system is based upon fidelity to rulemakers’ perceptions of the underlying social goals embodied in substantive legislation, such fidelity likely comes at too high a cost. Most lawyers and commentators would intuitively be uncomfortable with a rulemaking regime in which the Chief Justice’s appointed committee rulemakers attempted to implement their own vision of the legislative purpose behind the Civil Rights Act of 1964 or the federal antitrust laws by crafting custom procedural rules intended to put a thumb on the scale in favor of their preferred litigants. A substantive equality solution must do a better job of sorting like from unlike than the formal equality-driven system it replaces, without undermining the legitimacy of the civil justice system as a whole.

As explained below, this sort of limited substantive equality intervention is possible.98 Substantive equality can remedy a number of structural imbalances inherent in a formally equal approach to civil procedure without running afoul of the limitations inherent in the Rules Enabling Act. And those Rules Enabling Act limitations are themselves a feature rather than a bug, since they allow the Supreme Court and the committee rulemaking process to avoid the primary risks—subjectivity and administrability concerns—associated with adoption of substantive equality approaches in other legal environments.

III.     Transsubstantivity and the Substance/Procedure Dichotomy

A.     Defining “Transsubstantivity”

As discussed above, United States civil procedure regimes are generally crafted along formal equality lines.99 In the parlance of procedure scholarship, formally equal U.S. civil procedure is transsubstantive because most procedural rules apply across all different types of substantive legal claims. In a recent article, David Marcus offered a description of transsubstantivity and its historical roots.100 In simplest terms, “[a] . . . rule is transsubstantive if it applies equally to all cases regardless of substance” or claim type.101

As Marcus notes, there are several traps for the unwary lurking in this seemingly straightforward definitional issue. First and foremost, uniformity and transsubstantivity are not synonymous.102 A rule is transsubstantive if it applies to all claim types; uniformity by contrast can refer to several different concepts. For example, geographic uniformity means that the same rules apply in all jurisdictions.103 But geographically uniform rules may be either transsubstantive or substance specific.104 Marcus also identifies another form of uniformity: a system is procedurally uniform when the same rules apply regardless of the size or complexity of a case.105 A system may thus be non-uniform if it applies one set of deposition rules to class actions, and another to individual suits, but that system would still be transsubstantive if the substance of claims within those size and complexity categories did not affect the applicable rules.106

Marcus makes two other important definitional contributions. First, he rebuts critics who contend that the federal procedural system is not transsubstantive because it vests significant discretion in judges.107 Though Marcus acknowledges that this discretion may permit a biased judge to advance her own outcome preferences under cover of ostensibly procedural rulings, he distinguishes this risk from “systemic approval or disapproval of a particular substantive area of litigation.”108 The fact that an individual judge may be able to use her procedural discretion to advance her own substantive agenda by claim type does not eliminate the transsubstantive character of the Federal Rules as a whole.

Second, Marcus contends that transsubstantive rules are “value-neutral” in an important sense.109 Although he recognizes the myriad ways in which even purportedly transsubstantive rules are inherently value laden, he finds some laudable value neutrality in the fact that generically applicable procedural rules will not affect the primary behavior of regulated parties in the way that substance-specific rules would.110 The across-the-board adoption of a heightened pleading standard,111 for example, might inspire generally lower compliance with the entirety of substantive civil law as potential defendants recognize that the bar just got higher for plaintiffs everywhere. But that stricter standard at least arguably would not yield disproportionate, substance-specific effects in the way that a substance-specific pleading standard would.112

B.     Substance and Procedure: Historical Background

The transition from common law pleading and traditional equitable practice, to code pleading, and finally to the procedural approach embodied in the Federal Rules is well rehearsed.113 But modern scholars have paid relatively less attention to the evolution and development of the transsubstantivity norm throughout that transition. It is important to undergo this historical examination, because it tells us something important about the assumptions undergirding modern transsubstantive procedure, and the extent to which those assumptions remain reasonable in today’s litigation environment.

Before the 18th century, procedure at common law was remarkably substance specific, in large part because the substantive law itself had no coherent theoretical design.114 As a result, the imported common law backbone of the American legal system was writ specific and thus procedurally byzantine.115

But commitments to transsubstantivity could not have developed without the predicate conceptual disengagement of substance and procedure. Under the common law, each individual writ encompassed “a wide range of procedural, remedial, and evidentiary incidents.”116 Professor Subrin points to the publication of Blackstone’s Commentaries (which conceptually separated rights, wrongs, and methods of enforcement) as a key development in the American evolution toward separated spheres of substance and procedure.117

The transsubstantivity movement arguably arose not long after Blackstone published his seminal treatises. In many ways, modern commitments to transsubstantive procedure trace their roots to Jeremy Bentham and the reforms he spearheaded in 19th century English legal practice.118 Bentham explicitly identified “adjective law” as a category separate from substantive law so that adjective law was directed toward “the maximization of the execution and effect given to the substantive branch of the law.”119 In addition to his focus on accuracy of result,120 Bentham also promoted efficiency as a subordinate value.121 Because Bentham’s adjective law was pointed at accuracy and efficiency, its ideal form was necessarily transsubstantive.122 Bentham’s focus upon “maximization of happiness through the substantive law” precluded any substantive role for procedure itself.123 His sharp substance/procedure dichotomy was bookended by his similarly sharp distinction between legislative and judicial functions.124

The idea of transsubstantive procedure began to make serious inroads in United States procedure with the rise of formalized legal codes in the 19th century.125 For example, the crisp demarcation between substance and procedure was a hallmark of early efforts to reform common law practice as well. David Dudley Field emphasized “nothing less than a uniform course of proceeding, in all cases” in his New York Code of Procedure that became the model for widespread reform throughout the United States in the latter half of the 19th century.126 As with Bentham before him and the Federal Rules architects after him, Field assumed that procedural rules were intended to ensure the efficient and accurate enforcement of substantive law.127 It was to this end that he employed his complexity-reducing transsubstantive approach.128

The transsubstantivity principle’s “final triumph” came with the adoption of the Federal Rules of Civil Procedure in 1938.129 The framers of the Rules were deeply concerned about the lawyers of their day “mak[ing] the rules an end in themselves and not the means to an end.”130 In recognition of this concern, they rejected formalistic, substance-specific procedural requirements in favor of court-made transsubstantive rules that would apply to all federal civil cases.131

Throughout the early 20th century, proponents of procedural reform generally embraced a transsubstantivity norm and typically supported their position with Benthamite arguments regarding accuracy and efficiency.132 Rules pioneer Charles Clark proclaimed that procedural rules exist only “to aid in the efficient application of the substantive law,”133 while Harvard Law’s Edmond Morgan went even further, defining procedure as pointing toward only one goal: “[T]he attainment of a just and speedy decision upon the merits, according to the principles of substantive law, at the lowest practicable cost . . . .”134 Those pronouncements accorded well with the Langdellian positivism of the late nineteenth century, when “[l]aw as [s]cience” was still a reputable opinion to hold.135

But even as these commentators wrote, the once sharp distinction between substance and procedure had begun to blur.136 Depression-era legal realists increasingly challenged the notion that substance and procedure could be separated, just as they challenged the “black-robed infallibility” of the all-too-human judges tasked with employing procedure to effectuate substantive goals.137

The essence of the realist attack on the substance/procedure dichotomy is perhaps best expressed by lawyer Thurman Arnold, who criticized the dichotomy as the byproduct of an indefensible “‘attitude’ about the sanctity of law.”138 In Arnold’s characterization, that attitude deified substantive law as “sacred” and unchangeable, relegating procedural law to a flexible, results-oriented “practical” role.139 But Arnold questioned the attitude, noting that either of the two ostensibly distinct categories could be employed to solve a particular legal problem.140 He thus described substantive law as “canonized procedure” and procedural law as “unfrocked substantive law.”141

It is somewhat unclear how the architects of the Federal Rules themselves understood the problem. Although Charles Clark acknowledged that “the line between [substantive and procedural law] is shadowy at best,”142 he seemed simultaneously to embrace the notion that procedure is “normatively distinct from and subordinate to substantive law.”143 In 1923, Edson Sunderland, one of the key drafters of the Federal Rules, described the distinction between the two as follows:

Now the law consists of two distinct and almost independent sets of rules or principles, one making up the field of so-called substantive law, the other the field of procedure. The first group is primary and constitutes an essential part of the structure of society; the second is secondary and derivative, and merely serves to make the first operative. The first group is relatively fixed, and only changes with the slow evolution of social relations; the second is relatively flexible, having no universal quality, but being the mere manifestations of opportunist ingenuity. To radically revise the first would mean a social revolution, but the second could be totally reorganized at a moment’s notice without causing a tremor in the social structure.144

Robert Bone further explains the mentality of the day:

This view of procedural rulemaking as an exercise in instrumental rationality divorced from substantive value may seem alien to many proceduralists today, accustomed as we are to focusing on the substantive effects of procedural rules. The distinction made sense to early 20th century reformers partly because they never seriously questioned the basic features of adversarial adjudication . . . .145

Although substantial variation in the numerous Rules architects’ viewpoints makes generalization a risky proposition,146 the following summary might come close. To begin with, the framers of the Rules generally acknowledged that procedural rules affected outcomes and, thus, had an impact upon substance, or at least upon substantive values. Notwithstanding this understanding, however, the framers of the Rules genuinely believed in process values and found instrumental utility in maintaining the distinction between procedural law and substantive law. In keeping with this, they ultimately saw the primary goal of procedural law in essentially Benthamite terms, promoting transsubstantive rules for efficiency reasons. Accordingly, in their view, the primary purpose of a transsubstantive procedural regime was to streamline litigation, while the adversarial process was responsible for generating “correct” results in litigated cases. Effective proceduralists were thus best thought of as a species of engineers, capable of designing procedural systems that generate efficient outcomes.

Transsubstantivity has thus clearly been the dominant paradigm for procedural rulemaking throughout the history of the Federal Rules. The approach the Federal Rules embraces is thus at least nominally formally equal, treating each civil case as “like” by applying the same rules to virtually all litigated claims. Today’s committee rulemakers still see themselves as bound to follow a strong transsubstantive procedure norm, and the idea of a sharp distinction between substance and procedure continues to resonate in broader society.147 

IV.     Civil Procedure’s Evolving Equality Problem

Formally equal systems treat all participants identically, often with disastrous results. This is true whether we are talking about the ways formally equal protection doctrine systematically discriminates against women and persons of color, or instead, as detailed below, about the ways in which transsubstantive procedure discriminates against claim types that do not fit the committee rulemakers’ rule design paradigm.

As the traditional equality literature demonstrates, there is substantial dissonance between the continued resonance of formal equality norms as an ideal and their unjust effects in practice.148 This dissonance is likely a product of the difficulties inherent in disentangling the pervasive and powerful descriptive belief in the essential moral equality of all persons from the pervasive and powerful descriptive reality that characteristics other than “personhood” (e.g., gender, race, socioeconomic background, etc.) have predictable effects on both opportunity and outcome. In effect, we end up legislating on the basis of belief and aspiration, rather than the far uglier reality.149

Formally equal transsubstantive procedure yields similarly unjust results, but for slightly different, largely evolutionary, reasons. In simplest terms, civil litigation dockets have changed substantially since transsubstantivity won the day, and those changes have largely destroyed the homogeneity necessary to justify the formally equal system we continue to employ.

When tracking the evolution of the federal civil litigation docket from the inception of the Federal Rules in 1938 to present, an interesting pattern emerges. In simplest terms, a critical mass of federal civil claims were in fact “like” along the characteristics most relevant to procedural justice in 1938. But within a few decades that was no longer the case. The 21st century civil docket is decidedly “unlike” along the claim type dimension, thus rendering formally equal transsubstantive procedure demonstrably unjust, all because we still insist on treating the federal civil claim as the relevant unit of analysis.

This Part therefore traces the development of transsubstantive procedure, describes the essential economic nature of litigation, and demonstrates that federal civil litigation as a whole has evolved substantially since 1938. As a result of that evolution, the essential economic “likeness” of civil claims has largely evaporated, taking with it the primary justification for a formally equal approach to procedure. Consequently, and as explained below, rather than treating the federal civil claim as the relevant unit of analysis, the rulemakers must instead consider the potentially differing economic incentives lurking behind certain categories of claims.

There is a credible argument that Rules drafters faced a relatively “like” docket as they worked to develop the Rules in the 1930s.150 But however homogeneous the federal civil docket of 1938 was, the modern federal docket is radically less so. Substantial economic heterogeneity between claim types is now the order of the day, and a transsubstantive procedural regime designed with the economic dynamics of some 1938 “paradigm case” in mind has disastrous effects upon the substantial subset of modern cases that do not fit the original model.

When the Rules framers first drafted the Rules, they made two interrelated mistakes. First, while they correctly sensed that 1930s rightsholders with business before the federal courts were fundamentally “like,” they did not fully consider precisely which characteristics were relevant to the “likeness” analysis. Second, by embracing transsubstantivity as something close to an article of faith, they effectively foreclosed consideration of the possibility that whatever those relevant characteristics were, the likeness of the 1930s might diverge over time.

To be clear, federal civil claims are not “unlike” simply because there are many different causes of action available to federal civil plaintiffs. There were many different federal civil causes of action in 1938 as well, and formal equality still worked reasonably well for a time.151 Rather, modern federal civil claims of different types are “unlike” because, unlike the relatively homogeneous docket of 1938, different modern claim types involve differing sets of intra-case litigation incentives.152 Those incentives are in turn a function of significant differences in the distribution of costs, information, and risk preferences for different classes of litigants—differences that did not really exist across claim types in 1938.153 The fundamental “likeness” of the 1938 federal civil docket was thus a function of the intra-case economic incentives that drive litigation behavior.

A.     The Economics of Civil Litigation

Civil litigation is an inherently and historically economic enterprise. Because, in part, even a highly efficient litigation system will be costly to invoke, both actual and potential litigants tend on balance to make litigation decisions on the basis of the expected costs and benefits of those decisions. This is true both at the initial file/don’t file or defend/settle junctures, and at every other point during litigation. In colloquial terms, it would be irrational to file a suit seeking only $25,000 in damages if it will cost $50,000 to obtain a judgment in that amount, even if victory is guaranteed.154

Other features of the system reinforce this tendency. Cases involving only damages claims are rarely filed and even less frequently pursued to judgment if the defendant will not be able to pay in the end—it makes no sense for even the most grievously injured plaintiff to file suit when the defendant lacks the resources to satisfy a judgment. Additionally, whatever the deadweight losses associated with the involvement of attorneys in litigation, lawyers tend to drive litigants toward economically rational decision-making. No matter how irrational a party’s original expectations, no matter how intensely a party believes in litigating for the principle of the thing, the vast majority of lawyers expect to be paid for their efforts. They also work hard to preserve their own reputational capital. Taken together, then, these incentives drive lawyers to encourage their clients to make economically rational decisions based upon costs and benefits rather than passion and emotion.

1.     Critiques of Economic Rationality

To be sure, economic rationality is hardly universal in litigation. For instance, some small fraction of litigants will not respond to economic incentives in the way most will. More troubling, as Jeff Rachlinski has argued, framing effects can produce systematic irrationality in civil litigants.155 Chris Guthrie has expanded on Rachlinski’s work, arguing that the “prospect theory” developed by Dan Kahneman and Amos Tversky casts serious doubt on the notion that litigants are in fact rational actors.156 In simplest terms, prospect theory explores human decision-making when considering a menu of different “gambles” or “prospects.”157 Following Kahneman and Tversky, both Rachlinski and Guthrie essentially contend that litigants tend to systematically deviate from risk-neutral economic rationality in one direction when considering possible gain (overvaluing their chances of success), and in the opposite direction when contemplating possible loss (overvaluing their chances of failure).158 An individual might thus regard a 50% chance of winning $10 as less desirable than a 100% chance of winning $5, despite the economic equality of those propositions from an “expected value” perspective. By contrast, that same person might prefer a 100% chance of losing “only” $5 to the economically equivalent 50% chance of losing $10.

While valid, the Rachlinski–Guthrie critique of economic rationality in litigation does not substantially affect the analysis of this Article for at least two reasons. First, actual civil litigation—rather than the stylized experiments upon which they base their conclusions—has a number of features that tend to mitigate biases of the sort they identify. As Rachlinski himself admits, large chunks of the civil litigation docket involve either or both large institutional defendants and insurance companies responsible for paying claims.159 Citizens United v. Federal Election Commission may declare these companies “persons” by First Amendment fiat, but it cannot make them “individuals.”160 Both institutional entities and the insurance companies responsible for payment of many claims have strong incentives to overcome any prospect theory biases inherent in the litigation enterprise.

Similarly, the presence of attorneys will—at the systemic level at least, if not always in each individual case—drive players toward economic rationality. It is not too much of a stretch to describe litigation lawyers as something like rationality engineers; if nothing else, a lawyer’s personal reputation is her stock-in-trade, and that in turn depends enormously on her ability to inject rationality into her conversations with her client. Moreover, litigation itself is more than a one-time gamble.161 It is instead a series of economic decisions large and small, therefore creating the possibility for learning—something noticeably absent in the single-iteration contexts studied by Rachlinski and others. As a result, the systemic effects of prospect theory are likely to be far smaller than simple extrapolation from economic experiments would suggest.

Second, even if these sorts of biases do represent a pervasive systemic problem, it is far from clear that they require us to discard economic analysis of litigation incentives as the basis for incorporating substantive equality into civil procedure. Even Rachlinski states: “I do not question the basic premise that litigants try to achieve the best possible outcome, but I do question their ability to identify the most favorable options when risk and uncertainty are involved.”162 Rachlinski’s approach thus might require someone considering the economic incentives of litigation to incorporate the specific types of irrationality Rachlinski identifies into her calculus. For example, if defendants do tend to prefer smaller certain losses to economically equivalent but uncertain larger losses, rulemakers attempting to balance intra-case economic incentives as this Article suggests may ultimately have to provide defendants with a bit of additional incentive to overcome that bias, or vice versa. In the end, however, the Rachlinski–Guthrie critique does not change the fact that economics drive litigation behavior.

2.     Implications of the Economic Frame

For committee rulemakers genuinely committed to the creation of a procedurally just system, the most important moving parts are the merits of the parties’ claims, the defenses, and the intra-case economic incentives that influence parties’ litigation behavior apart from the merits. In very general terms, a procedurally just system will maximize the effects of the parties’ merits positions upon the outcome of litigation while minimizing the effects of unrelated economic incentives upon the result.

But it is not quite that easy in practice, mostly because the relationship between the cost of litigation and the stakes of litigation163 matters, at least on the margins. For example, most would intuitively reject an adversary system that attempted to balance costs such that it becomes economically viable for an individual plaintiff to file a federal damages lawsuit in connection with even a completely valid ten dollar claim.164 This is especially true if doing so imposes thousands of dollars of costs on society.

Nonetheless, the primary procedural justice goal for committee rulemakers should be promulgation of rules that get intra-case economics out of the way, such that litigation outcomes will be driven by the merits of each party’s claims or defenses.165 Importantly, this is not quite the same thing as saying that economic imbalances automatically make substance-specific procedure necessary. If economic imbalances are uniformly distributed across all case types, we can still employ a transsubstantive approach. It would be procedurally just or unjust depending upon the ways in which it interacted with that uniform set of incentives, but the justice or injustice of the system would not be a function of its transsubstantivity per se.

However, because committee rulemakers should pursue only procedural justice in their rulemaking, it follows that they must not deploy transsubstantive procedure if these economic incentives do affect different case types differently.

B.     The Evolution of the Formal Equality Approach from Just to Unjust

When the Federal Rules went into effect in 1938, the broad, nominal claim type diversity of the federal civil docket masked a rather remarkable homogeneity in the ways cost, information, and risk preference profiles were distributed. Today’s federal civil litigation docket includes even more claim type diversity, occasioned by the explosion in federal private rights of action that began shortly after the Rules took effect.166 Far more important, the modern federal civil docket is far more economically heterogeneous than its 1938 analogue. The distribution of cost, information, and risk preference incentives today differ substantially from claim type to claim type, and this heterogeneity puts real pressure on the transsubstantivity norm. The economic incentives engendered in modern transsubstantive procedure can yield results largely independent of the merits of claims and defenses in certain categories of cases.

1.     Cost, Information, and Risk Preference Distribution

For a transsubstantive procedural regime to be successful, the vast majority of cases must have relatively similar pretrial cost, information, and risk preference profiles. When pretrial costs, information, and/or risk preferences are not uniformly distributed, a formally equal transsubstantive procedural regime creates opportunities for litigants to leverage asymmetries to their advantage without regard for the merits of the claims or defenses at issue in the case.167 Specifically, when any one of these three characteristics is distributed heterogeneously across the spectrum of civil claim types, that heterogeneous distribution ultimately yields heterogeneous incentives. For example, in a world in which all case types involve roughly identical distribution of information, a transsubstantive discovery regime will produce consistent incentives. But in a heterogeneous information environment, a discovery-related rule that might be perfectly appropriate for certain information distributions might actually create undesirable information cost-related leverage in others. And so on.

To be clear, this critique of the transsubstantivity norm depends primarily upon the distribution of different cost/information/risk preference profiles across the run of civil litigation, and not upon the specific shape of these profiles. Admittedly, this Article focuses substantial attention on the potentially perverse and unequal effects of cost, information, and risk preference asymmetries between plaintiffs and defendants upon civil litigation outcomes. Standing alone, those asymmetries and their effects on litigant behavior might justify changes to existing rules, but they do not necessarily justify abandoning the transsubstantivity norm. For instance, if a particular asymmetry is affecting virtually all cases, a transsubstantive amendment to the civil rules may well be in order to fix the problem. Transsubstantive procedure instead falters when some types of cases predictably face different cost/information/risk preference profiles than others; it is only then that the one-size-fits-all approach fails the test and is no longer just.

As explained in more detail below, this Article therefore focuses on asymmetries not because those asymmetries themselves undermine the transsubstantivity norm, but because the architects of the modern transsubstantive approach reasonably and understandably assumed that relative parity in costs, information, and risk preference characterized the civil litigation of their time.168 Importantly, at the dawn of the Rules era, that is exactly what Rules framers saw—economically relatively homogeneous cost profiles and information distribution.169 In fact, when the Rules went into effect, virtually all routinely litigated civil claims would have been characterized by relative parity in pretrial costs, and in consistent, generally predictable informational environments.170 Consequently, it is understandable that a set of transsubstantive rules designed with the belief that most cases are economically similar struggles to promote justice in the far more economically heterogeneous modern environment.

i.     Pretrial Cost Distribution at the Dawn of the Rules Era

When we begin considering the context in which the original Rules were drafted, it becomes clear that, in one sense, there was no “paradigm federal case” in 1938.171 Roughly contemporaneous studies of the federal dockets instead demonstrate that federal civil proceedings at the time involved a wide range of causes of action.172 Although the United States was itself a party in a large percentage of these cases,173 private litigation was a significant component of the federal civil docket as well.174

However, the variety of the nominal claim types in the 1930s federal docket does not equate to significant variety in litigation cost and information profiles. Rather, at the time the Rules were adopted, virtually every relevant federal case175 (and most state cases) would have been characterized by relative cost parity, at least compared to the benchmark that I have elsewhere suggested justifies intervention of some sort.176 This is in part because of the nature of the claims actually litigated in federal court at that time.177 Virtually none of the traditional common law causes of action—higher-value contract and tort injury suits which were the backbone of the federal diversity docket—involved significant pretrial cost disparity between plaintiffs and defendants.178 In addition, there were far fewer statutory causes of action available to private plaintiffs at the time.179 Finally, many of those that were technically available were not financially practicable because certain classes of plaintiffs were unable to aggregate small claims effectively.180

Antitrust law provides an example of this latter phenomenon. When Congress first enacted the Sherman Act in 1890, private enforcement was deemphasized; rather, the Attorney General had primary responsibility for enforcing the law.181 In 1903, the newly created Department of Justice Antitrust Division took over public enforcement of the law.182 Private plaintiffs were first encouraged to pursue antitrust claims seriously with the passage of the Clayton Act in 1914, which authorized treble damages suits by “any person who shall be injured in his business or property by reason of anything forbidden in the antitrust laws.”183 Consequently, before 1938 (and, indeed, before 1966, when the Federal Rules of Civil Procedure were amended to add the small claims class action under Rule 23(b)(3)), the typical private civil antitrust claim pitted one business against another.184

Although these sorts of suits may have involved some disparity in business size between plaintiffs and defendants,185 it is unlikely that these size disparities occasioned much in the way of pretrial cost disparities. Specifically, these cases typically involved a claim of illegal behavior directed at a competitor; as such, they would necessarily have involved searching inquiry into the transactions between the two companies and/or their customers or suppliers. Therefore, the plaintiff’s ability to impose costs on the defendant likely would have been matched by the defendant’s ability to do likewise.186

In contrast, consumers with small individual claims against a price fixing cartel had the nominal right to sue under the Clayton Act in 1938 (“any person”), but they had no true functional opportunity to do so. The massive costs of suit dwarfed their paltry potential individual recoveries.187

The antitrust story is hardly unique to the time; the federal dockets in the early years of the Federal Rules consisted largely of economically similar disputes, regardless of their substantive law contexts.

ii.     Information Distribution at the Dawn of the Rules Era

Along with largely economic consistent disputes, federal civil litigation at the dawn of the Rules Era typically involved a relatively small universe of relevant factual information.188 Moreover, in a world without Westlaw, Google, and even Wikipedia, search costs for relevant legal information (i.e., which procedural or substantive law governed a given claim) were quite high. Those high search costs likely fed into the Rules framers’ perception that transsubstantive rules would lower process costs associated with civil litigation. Data storage at the time was entirely hard copy,189 and even the production and duplication of hard-copy information was severely limited by available technology.190 This in turn meant that the vast majority of litigated claims, including those that would today involve massive pretrial cost disparity, simply could not generate hugely disproportionate discovery bills—a primary source of pretrial cost disparity. Importantly, the data explosion of the last forty years has not had uniformly distributed effects. Specifically, many civil cases continue to be “low data,” involving relatively limited information. Instead, it is the uneven distribution of the effects of the Information Age data revolution that weakens the transsubstantivity norm, not the fact that there is simply more information available.191 Although a plaintiff or defendant bent on conducting impositional discovery in 1938 might have been able to force her adversary to search its own haystacks for needles, they generally would have been pretty small haystacks.

iii.     Cost and Information Disparities in Modern Civil Litigation

In contrast to these more limited and homogenous disparities, the modern civil litigation environment is characterized by dynamics that inevitably lead to significant heterogeneously distributed pretrial cost disparity across the federal docket. Moreover, although information costs are still high for those without substantial experience in the civil litigation arena, ready access to information in the digital age has, on the margins at least, reduced the need for formally equal transsubstantive procedure.192

Since the adoption of the Federal Rules in 1938, lawmakers have added countless new private rights of action to the law. At the federal level, the most notable developments are the dramatic expansion of civil rights-related claims and a revolutionary turn toward private enforcement of environmental claims.193 At the state level, the development of strict liability in tort stands out.194 Moreover, these new causes of action tend to involve transactions that are disproportionately focused upon the actions and/or motivations of the defendant.195 As explicit tools of social policy, new causes of action have thus typically been significantly less focused on the actions of plaintiffs than the paradigm case envisioned by early proponents of transsubstantivity.

For example, the Civil Rights Act of 1964 prohibits employment discrimination on the basis of race.196 As interpreted by the Supreme Court, Title VII plaintiffs can prove illegal discrimination not only by direct evidence of discrimination but also by indirect evidence of disparate treatment.197 Under the Supreme Court’s approach to indirect evidence discrimination claims, any plaintiff that demonstrates a prima facie case of discrimination198 will ultimately have to prove that the legitimate justification proffered by the defendant is pretextual.199 Pretrial cost burdens are likely to fall more heavily on defendants in such situations, because the plaintiff can force the defendant to search for evidence of pretext in the defendant’s own files under the prevailing liberal discovery rules; the defendant’s reciprocal ability to impose costs upon the plaintiff will be limited by comparison.

Similarly, in citizen-suit environmental litigation, individual plaintiffs act essentially as private attorneys general. The only inquiry in such cases is whether the defendant violated the environmental law in question—the plaintiffs’ activities are completely irrelevant.200 The same is true to a lesser extent in strict liability cases, such as product liability litigation. In those cases, although the plaintiff must demonstrate an injury and will likely retain experts to testify on the issue of defect, the bulk of pretrial proceedings will focus upon information uniquely within the defendant’s possession, custody, or control—design specifications and history, safety evaluations, etc. To the extent new causes of action create a cost dynamic favoring plaintiffs over defendants, they may well call for a new approach to transsubstantivity.

An even bigger story in modern litigation is the law’s relatively recent embrace of devices that mitigate a different type of cost disparity problem in certain types of litigation. Specifically, claim-aggregating devices like Rule 23(b)(3)201 allow the owners of small claims to band together, thus making it economically feasible for small-claim plaintiffs to pursue legal remedies against defendants whose misdeeds might otherwise go unpunished, as the cost of pursuing such claims would be far too high for any single plaintiff.202               As an example, return again to the history of antitrust law enforcement. Once the Rules were amended to allow small claim class actions in 1966, consumer suits became a viable alternative.203 Without the ability to aggregate claims, however, an individual purchaser of price-fixed goods would not likely find filing a suit attractive since she would likely pay many times more in fees and time expended than her expected recovery could possibly justify, even with treble damages.204 However, once her claim can be aggregated with other similar claims under Rule 23(b)(3), her suit becomes viable. And indeed, modern antitrust litigation reflects this change.205 The same is true for other types of claims as well. For example, many shareholder securities fraud claims only became viable with the adoption of Rule 23(b)(3).

For claim types characterized by significant information disparities, the arrival of the digital age has exacerbated the problem a thousand-fold or more. While the ready availability of terabyte-level storage has undoubtedly provided a prospective plaintiff with myriad additional avenues to pursue in her search for proof of wrongdoing, it has also increased the costs she can impose on the defendant by several orders of magnitude.206 Moreover, for several post-1938 claim types, growth in the quantum of potentially relevant and discoverable information has not been symmetrical between plaintiffs and defendants. For example, the typical class action securities litigation plaintiff will still possess relatively little discoverable information, as will her class action antitrust counterpart. Even a Title VII plaintiff—whose Facebook posts and email accounts likely will be scrutinized closely by the defendant—likely has not seen her relevant digital file room expand in proportion to her defendant employer’s.

Taken together, unanticipated changes in the distribution of cost asymmetries in the civil litigation environment since 1938 severely undermined the strongest argument in favor of formally equal transsubstantive procedure. The strength of this critique is only amplified when we consider similar changes in the distribution of informational and risk preference asymmetries in the modern civil docket, as discussed below.

The foregoing discussion of heterogeneously distributed cost asymmetries often implies a mirror-image change in the distribution of information asymmetries. Changes in the litigation environment since 1938 have produced substantially greater variance in pretrial cost asymmetries that disproportionately disadvantage defendants in many types of cases.207 Those same changes have also increased variance in the distribution of informationally asymmetric cases, largely in ways that disadvantage plaintiffs.

As discussed above, the rise of statutory state-of-mind requirements in strict liability tort causes of action and private-attorneys-general suits have together created a class of justiciable rights largely disconnected from the individual plaintiff’s own actions.208 In the Rules architects’ paradigm cases, claim-relevant information was likely relatively evenly distributed among the parties.209 In the modern environment, those paradigm cases now coexist with a set of new or evolved claim types in which relevant information resides primarily with one party.

This Article assumes the primacy of substantive lawmakers’ preferences regarding the suite of legal rights and obligations society wishes to enforce. Given that assumption, it is unjust under Rule 1’s terms of “just, speedy, and inexpensive” to apply transsubstantive rules that are perfectly well suited for the informationally symmetric 1938 paradigm case still present in large quantities on civil dockets when those same transsubstantive rules make it significantly more difficult for parties facing an informationally asymmetric modern claim to vindicate the rights assigned to them by substantive lawmakers.

A formally equal transsubstantive approach also makes sense only to the extent that the distribution of parties’ risk profiles is essentially uniform across all claim types. Risk consists of two components: degree of risk and risk preference.210 In the most general terms, “degree of risk” is simply some measure of the quantum of risk associated with a given course of action or outcome. By contrast, “risk preference” refers to one’s tolerance for a given degree of risk. One may be risk-neutral such that she would be indifferent between all options producing identical expected values. One may also be risk-averse or risk-seeking. A risk-averse individual values certainty, and would be likely to choose a course of action with lower risk, whereas a risk-seeking person derives utility from the very act of taking chances. Both degree of risk and risk preference may be asymmetrically and heterogeneously distributed among the range of civil litigants. As an example of degree of risk asymmetries, a consumer class action plaintiff likely faces a relatively low degree of risk relative to the bet-the-company risk faced by the defendant she has sued. If she loses, she suffers few consequences. In contrast, a plaintiff in a garden-variety commercial dispute may face a degree of risk roughly equivalent to or even greater than that of her adversary.

The same variance is likely present with respect to risk preference as well. The class action plaintiffs’ attorneys who are the driving force behind most of the consumer class action docket may be risk-neutral or even risk-seeking. However, an individual plaintiff involved in even relatively low stakes litigation—for example, a simple automotive negligence case involving relatively modest damages—may have quite a different risk preference profile. Depending upon the severity and type of injuries suffered and that plaintiff’s own resources, she might be expected to fall anywhere on the spectrum from risk-seeking to risk-averse, even without considering her individual psychology.211 This same variance is likely present across the range of civil defendants as well.212

Risk preference differs from party to party; it is almost certainly distributed heterogeneously across different claim types as well. If this heterogeneous distribution were simply random, then risk preference would likely be irrelevant to the transsubstantivity question. But if certain claim types systematically tend to involve particular risk profiles among plaintiffs and/or defendants, the story changes. For example, litigation in which liability insurance coverage typically plays a significant role will often involve relatively risk-neutral defendants, since the insurance companies’ entire business model depends upon risk neutrality.213 Self-insured defendants may well be more risk averse, especially as the stakes increase. We might expect to see the same variance among plaintiffs as well. Plaintiffs with few resources and significant injuries may be risk-averse, while plaintiffs who are simply rolling the dice without much personal stake in the game may be risk-neutral or even risk-seeking.214 Similarly, parties whose economic motivations are not purely pecuniary may tend to congregate in certain types of litigation.215 When these differential characteristics align with particular claim types, formally equal rules that fail to account for the differences can have distortive effects.216

Because both components of risk are likely asymmetrically distributed across the range of potential civil disputes, formal equality is sensible only if party responses to the civil rules are effectively risk-independent. This almost certainly is not the case. A risk-averse party may well have different reactions to civil rules than her risk-neutral or risk-seeking analogue.217 If nothing else, risk preference will affect the ex ante expectations of litigating parties—risk-averse litigants will downgrade the expected value of any action and upgrade the expected costs, and risk-seeking litigants will do the opposite. Although risk asymmetry problems occasioned by transsubstantive procedure may be somewhat more difficult to address by way of substance-specific deviation from the transsubstantive default rules, the heterogeneous distribution of risk profiles further undermines any continued commitment to our existing formal equality norm.218

The then-and-now narrative I have offered to explain cost distribution and information distribution may not hold with respect to the risk profile distribution story, or at least not in the same way. Economic incentives influencing litigant behavior with respect to cost and information likely would have been a part of the Rules architects’ background experience, if only at an informal and intuitive level. In other words, it probably makes sense to assume that the Rules architects were thinking about cost and information problems when they put the Federal Rules together, even if they never would have used those terms or invoked microeconomic theory to justify their approach.

Risk profile distribution, however, is different. It seems unlikely that Clark, Sunderland and Pound would have thought much at all about differential degrees of risk involved in civil litigation, except at the most basic level (e.g., considerations regarding the extent to which existing Byzantine procedural regimes might have been used by high-resource litigants to disadvantage less wealthy litigants can be recast in “degree of risk” terms). And it seems almost unthinkable that the Rules architects would have considered differential risk preferences among the various classes of litigants and attorneys involved in civil litigation.219

That said, there is still some reason to believe that there is a then-and-now component to the risk profile story as well. The evolutionary changes to the civil litigation docket I discuss above likely did scramble the distribution of risk profiles also; moreover, that scrambling likely changed that distribution from mostly homogeneous to demonstrably heterogeneous. The rise of statutory causes of action that place plaintiffs at little or no personal risk (e.g., environmental citizen suits) and the creation of the small claims class action almost certainly introduced greater variance into the risk profiles of litigating parties and their attorneys. So too with the rise of different methods of litigation finance, attorneys’ fees statutes, and the like.220 Thus, even though it seems unlikely that Rules architects explicitly considered risk profiles in 1938 to the same extent they likely considered cost and information issues, there is reason to believe that risk profiles are subject to both greater variance and more heterogeneous distribution today than they were at the dawn of the Rules era.

C.     Real-World Examples

Evidence of the civil procedure equality problem is everywhere, if we both know where to look and if we realize that sometimes absence of evidence is in fact evidence of absence. Those experienced with the ins and outs of intra-case economics (i.e., those who actually litigate federal civil cases for a living) can doubtless offer countless examples of scenarios in which modern litigation and formal equality combine with disastrous effect. I will limit myself to just two examples, one from the Supreme Court’s procedural case law, and one from the rulemaking process. Both instances show the need to consider seriously a substantive equality approach to modern, civil litigation.

1.     Twombly, Iqbal, and Conley

In the federal civil litigation system, cases that state a claim under Federal Rule of Civil Procedure 8(a)(2) proceed to discovery, which is often simultaneously the most valuable and the most expensive component of a civil case. A complaint that successfully states a claim satisfies the pleading standard. For the half-century between 1957 and 2007, the Supreme Court took a markedly liberal approach to the pleading standard, following its statement in Conley v. Gibson that a complaint should only be dismissed for failure to state a claim if “the plaintiff can prove no set of facts . . . which would entitle him to relief.”221 Things changed in 2007 and again in 2009, when the Supreme Court apparently adopted a new, more restrictive “plausibility pleading” test under which plaintiffs are required to plead non-conclusory facts that, if true, create a plausible inference of actionable wrongdoing.222 Commentators do not tend to stay on the fence when it comes to the Supreme Court’s jurisprudence on civil pleading standards. An overwhelming majority223 seem to prefer the Court’s old “no set of facts” standard as articulated in Conley v. Gibson,224 while a much smaller minority prefers the new “plausibility pleading” approach225 articulated in Bell Atlantic v. Twombly and Ashcroft v. Iqbal. Both groups are wrong.

The controversy surrounding the proper scope of the pleading standard is a paradigm example of the formal equality problem inherent in transsubstantive civil procedure. And the real problem is not that Conley is right and Twiqbal is wrong, or vice versa. The real problem is that the pleading standard is no longer a good candidate for transsubstantive procedural rulemaking—differences in the economics of different types of cases make transsubstantive application of either a liberal or a strict pleading standard a recipe for disaster in cases that do not fit the paradigm.

In light of this concern, in an earlier work I proposed a single deviation from the transsubstantivity norm to address problems arising from the way the pleading standard interacted with pretrial cost disparities in certain types of cases.226 Specifically, I argued for a heightened default pleading standard in a category of cases characterized by significant cost asymmetries favoring plaintiffs.227 Noting that such dynamics encouraged frivolous litigation and cost-of-defense settlements unconnected to the merits, I suggested “balancing the pleading equation” to encourage merits-driven outcomes.228 At the same time, I recognized that cases characterized by plaintiff-favoring pretrial cost asymmetry often involve significant defendant-favoring information asymmetry.229 I therefore recommended allowing such plaintiffs to “buy” the traditional relaxed Conley pleading standard by purchasing a bond that would be awarded to the defendant only upon a finding that the claim was frivolous.230

My recommendation in Balancing the Pleading Equation is an example of the type of substance-specific rulemaking I propose below in Part V. But the mere existence of cases fitting the cost disparity profile I identified is not in itself sufficient to justify abandoning the transsubstantivity norm. As explained above, while the dynamics I identified might call for a rule change to address the problem, that rule change should still apply transsubstantively if all of the cases on the docket fit essentially the same cost/information profile. Transsubstantivity itself is implicated only if the pretrial cost and information profiles in the civil justice system are heterogeneously distributed. In other words, formal equality only fails when like really is not like.

Consider again the pleading standard example. If rulemakers—or the Supreme Court, as in Twombly231 and Iqbal232—adopt a heightened pleading standard on a fully transsubstantive basis, that standard may do more harm than good, even if it represents an appropriate response to some subset of cases.233 For the large percentage of cases that do not involve significant cost asymmetries,234 the transsubstantive heightened pleading standard would raise error costs in the form of unwarranted dismissals and decreased filing rates for meritorious claims that are likely to be found implausible at the pleading stage. Furthermore, it raises those costs without any real countervailing benefit. At least some plaintiffs who have no incentive or ability to abuse the system through excess cost imposition will nonetheless either decide not to file meritorious claims or will see these claims dismissed at the pleading stage because they cannot satisfy the heightened standard.

Thus, the real problem with pleading law is not Twombly, Iqbal, or Conley. Instead, it is the transsubstantive application of whichever pleading standard the Supreme Court adopts. If Conley is the transsubstantive order of the day, formal equality abridges the substantive rights of defendants in violation of the Rules Enabling Act. Alternatively, if the standard established in Twombly and Iqbal applies to all cases, formal equality abridges the rights of plaintiffs with no incentives to abuse the system and precious little access to the information they need to prosecute their claims. Transsubstantive application of a single pleading standard also wreaks havoc upon procedural justice norms, generating outcomes that have little to do with accuracy and militate against meaningful participation on the part of whichever side happens to be disadvantaged by the economics of the case.

2.     The Proportionality Rules

Much has already been written about the new discovery rules that went into effect on December 1, 2015. Commentators have generally been most concerned by the new approach to “proportionality” in discovery, under which material is no longer deemed discoverable if it is not proportional to the needs of the case.235 This Article will not rehash the full debate over the propriety of such an approach here, primarily because commentators on both sides are committing something close to a category error by assuming transsubstantive application of the rule.236 They are not precisely wrong to do so, of course—the rule does apply transsubstantively, after all. But by focusing their arguments on the types of cases that animate their respective worldviews, the very vehemence of the debate argues eloquently for the position that they are all missing the point. The real problem, again, is transsubstantivity.237

Commentators critical of the proportionality amendments are concerned that the new standard will incentivize district courts to restrict discovery in the cases in which it is most necessary.238 Those who submitted anti-amendment comments routinely raised the specter of unjustified restriction of discovery in employment discrimination cases and other types of claims where the information relevant to a plaintiff’s claim is likely to be buried—and sometimes deliberately hidden—deep within a defendant’s records.239 These commentators lamented the fact that the new proportionality rules give judges too much ability to privilege their own priors in their management of certain categories of cases.240

By contrast, proponents of the amendments emphasized the classes of cases in which impositional discovery tends to distort incentives away from merits-driven outcomes. To them, the proportionality amendments provide judges with a necessary tool that can be used to protect those facing significant cost asymmetries under hyper-liberal discovery from being forced to litigate on the basis of discovery burdens instead of the merits.241

Again, both sides are right—and wrong. There is reason to believe that courts will be reasonably good at making the sorts of discretionary judgments the new rules require, but only in certain categories of cases. The most difficult of the courts’ discretionary determinations under the new Rule 26(b)(1) involves deciding whether discovery is proportional to the needs of the case, considering “whether the burden or expense of the proposed discovery outweighs its likely benefit.”242 Leaving aside for the moment the problems caused by the “discretion dodge” inherent in that sort of provision,243 this sort of determination may be possible for judges in certain types of high-volume, garden-variety federal suits. It is far less likely that judges will be able to make a reliable assessment in less common types of cases.244

As a result, slavish adherence to the transsubstantivity norm will again accomplish precisely the opposite of what it is supposed to accomplish: a tendency to expand or abridge rights in the classes of cases that present economic, information, and risk dynamics outside judges’ comfort zones.      The amendment to Rule 26 allowing “Early Rule 34 Requests” for production of documents offer an additional possible example of the phenomenon.245 For the traditional paradigm case characterized by relative informational parity, the ability to initiate discovery earlier will likely be highly beneficial. For claim types involving significant informational asymmetry, however, the rule change creates a set of perverse incentives to ratchet up discovery pressure against the party in possession of more information.246

D.     Implications of Modern Litigation Dynamics

In sum, the economics of civil litigation in 1938 generally supported implementation of the formal equality approach embodied in the Federal Rules of Civil Procedure. The nominal diversity of the federal civil docket at the level of individual claim types masked a rather striking homogeneity of economic incentives across all doctrinal areas of private litigation. In colloquial terms, a civil case was really just a civil case when the Federal Rules went into effect. As such, formally equal transsubstantive procedure made substantial sense.

But in less than three decades, the economic landscape was radically different. Congress began to incorporate private civil litigation as a critical component of its enforcement strategy, adopting a private attorneys general model for many new statutes, and creating new classes of rightsholders for whom the economics of litigation differed substantially from the old paradigm. Radical changes to the information environment affected things as well; claim types that used to fit within the economic envelope of the 1938 paradigm case no longer did, because the changes to the informational context were not necessarily symmetrical on both sides of litigated cases.

Continued application of formally equal transsubstantive procedure to these new (and newly) heterogeneous claim types has consequently resulted in substantial injustice as “unlike” cases are now being treated as “like.” To be clear, there is no clear ideological valence to this story—in some contexts, formally equal procedure has put plaintiffs at a disadvantage relative to similarly situated plaintiffs in a traditional paradigm case, while in others, defendants have suffered. The upshot is not that modern procedure is tilted in a particular direction, but rather that continued application of transsubstantive formal equality is accidentally forcing committee rulemakers to violate the very provisions of the Rules Enabling Act that adherence to the transsubstantivity norm was supposed to promote.

As explained in more detail below, a limited embrace of substantive equality principles will help committee rulemakers comply more fully with the letter and spirit of their charge, and will promote procedural equality in the bargain.

V.     A Substantive Equality Solution

Congress should reimagine the committee rulemaking enterprise to authorize substance-specific deviations from formal equality when the intra-case economics of a category of claims renders them “unlike” relative to the paradigm case that the Rules architects had in mind when they crafted the original transsubstantive rules. As discussed above, both institutional limitations and the importance of procedural justice as an independent value together counsel against employing a substantive equality free-for-all. Rather, the substantive equality approach this Article recommends would authorize rulemakers to deviate from the formal equality default only on limited grounds, and then only when the intervention in question merely balances incentives such that they are consistent with the incentives in paradigm cases involving similar overall stakes.

This particular form of substantive equality is politically feasible and lies within the general institutional capability of the committee rulemaking apparatus.247 It is also broadly consistent with relevant procedural justice norms of improving accuracy and meaningful participation by limiting the extent to which economic imbalances in costs, information, or risk preference drive outcomes away from merits-focused determinations. While not without its own risks, this solution to the civil procedure equality problem is conceptually and normatively superior to solutions proposed by other commentators, and answers most of the objections raised by those who have argued that the problem is effectively insoluble.

A.     The Basic Approach

As Stephen Subrin and David Marcus have noted, the benefits of transsubstantive procedure can be significant.248 I agree—so long as the cases to which transsubstantive rules apply are in fact “like” along the relevant economic characteristics. Thus, I would not authorize substance-specific departures from the formal equality approach embedded in rulemakers’ DNA whenever they feel like it. As Marcus, Burbank, and others have noted, that would be tantamount to writing rulemakers a substantive blank check in direct contravention of the limits imposed by the Rules Enabling Act and general separation of powers concerns.249

This Article therefore proposes only a limited deviation from transsubstantivity. In equality theory terms, it proposes that the Supreme Court and its committee rulemakers take the following approach: Consistent with the Aristotelian conception of the equality–justice relationship, rulemakers must treat like cases as like. Rulemakers should look to intra-case economic incentives alone to assess “likeness.” They should not continue to assume that all civil cases are sufficiently “like” to justify formal equality. Nor should they look to characteristics external to the litigation process itself to identify “likeness.” The injustices worked by continued application of formal equality require rejection of: (1) the “all civil cases” model; (2) the limited role of rulemakers; and (3) fidelity to long-term, inter-case principles of procedural justice that prohibit attempts to pursue other goals. When economic incentives present in a class of cases differ significantly from the paradigm baseline characteristic of the federal civil docket at the inception of the Federal Rules of Civil Procedure in 1938, the Supreme Court and committee rulemakers can act to maximize procedural justice. Accordingly, rulemakers can propose substance-specific rules that will foster accuracy, appropriate types of cost-balancing, and meaningful participation when the existing transsubstantive rules are actually in conflict with the “no substance” provisions of the Rules Enabling Act. However, they should only be authorized to do this if rulemakers can move the economic needle to that economic midpoint without overcorrecting. If an intervention would “solve” a set of incentives that disadvantages defendants only by placing plaintiffs at a similar disadvantage, it should be off the table, and vice versa.250

B.     The Paradigm Case Baseline

There is at least one significant practical and theoretical challenge embedded in a substantive equality approach grounded in economic balancing. Specifically, it will be difficult for rulemakers to engage in the balancing exercise without crossing into substantive territory when determining when the public and private stakes of civil litigation justify engaging the costly machinery of the federal government. Most would reflexively reject the notion that the federal courts should routinely decide $10 federal question claims with little or no extrinsic public value, but rulemakers would exceed their authority if they were to use their substance-specific rulemaking powers to decide the threshold importance level that makes a class of cases worth the federal courts’ time. That is a job for Congress.

Here, the economics of the 1938 “paradigm case” offers some assistance. While rulemakers cannot determine when the stakes of a given class of claims are sufficiently high to justify federal court time, they can perform a comparative exercise. When it comes to baseline incentives to litigation, therefore, their substantive equality interventions should thus have, as their goals, parity with the traditional paradigm case.

More concretely, imagine two cases with precisely the same public and private stakes; that is, two cases in which the value of the claim to the plaintiff and the public importance of the rights in question are identical. One of the cases is a traditional paradigm case of the sort with which Rules architects were familiar (and by extension, for which the formal equality norm serves reasonably well). The other is a “new” case with significantly different economic dynamics. Rulemakers can decide how much of the new claim they think appropriate for the federal courts only by reference to the paradigm case. The goal must be to balance the intra-case economic incentives, not to obtain some hypothetical optimal level of litigation. That “optimality” decision cannot come from committee rulemakers. But those rulemakers can and should be authorized to ensure that plaintiffs pursue the new claim type with the same frequency and to the same extent they would pursue a traditional paradigm case involving identical stakes.

C.     A Brief Note on Congressional Intent

As explained earlier, we cannot infer congressional assent to the Supreme Court’s or rulemakers’ decisions from the fact that Congress has not responded. Additionally, as demonstrated in earlier work, there is ample reason to doubt that silence means much with respect to congressional preferences.251 Admittedly, this current approach presumes that Congress does not legislate from a position of perfect knowledge. Put differently, it assumes that when Congress is enacting substantive legislation with a private civil enforcement component, it does not fully understand and assimilate the distortive effects of transsubstantive procedure into its legislative scheme.

I reserve a full discussion of the justifications for making this assumption for future work. However, there are reasons to believe that Congress does not have the ability to anticipate the interactions between its substantive enactments and transsubstantive procedural law with sufficient foresight to intend the procedurally unjust effects economic mismatches engender. Moreover, there are equally good reasons for the Supreme Court and committee rulemakers to act as if Congress legislates under the assumption that the civil litigation system will generate procedurally just results in private litigation.

D.     Implementing the Solution

The federal judiciary’s civil procedure rulemaking apparatus consists of the Advisory Committee on Rules of Civil Procedure, the Standing Committee on Rules of Practice and Procedure, the Judicial Conference of the United States, and the Supreme Court.252 This assemblage of talent is impressive, but neither the Advisory Committee that carries the laboring oar in drafting prospective rules nor any other component of the rulemaking system possesses particularly deep reservoirs of economic expertise. Accordingly, implementation of this proposal will require rulemakers to get some help.

In the most general sense, rulemakers will need the assistance of economists (more particularly, game theory experts and experimental economists) who can model the incentives inherent in different types of civil litigation. They will also need the help of empirical researchers who will be tasked with ensuring good fit between models and experiments and real-world outcomes.

It would be possible for committee rulemakers to simply rent this expertise on an ad hoc basis by retaining the necessary personnel, but the politics of the Rules Enabling Act suggest that another solution might be preferable. To ensure that committee rulemakers are fulfilling their obligations without exceeding their authority, it may be preferable to create something akin to the Office of Information and Regulatory Affairs (“OIRA”) within the rulemaking apparatus. OIRA is an executive branch institution within the Office of Management and Budget (“OMB”).253 OIRA was created by Executive Order254 and among other things, it is tasked with assessing the costs and benefits of proposed executive branch regulations.255

A rulemaking OIRA would serve much the same function, and could be given both reactive and proactive duties. At the reactive level, a rulemaking OIRA could provide expert analysis of the systemic costs and benefits of existing rules proposals, determining whether application of proposed rules (either fully transsubstantive or substance-specific under this Article’s proposal) would create any perverse economic incentives for litigants that would drive outcomes away from those we consider procedurally just. It could also be tasked with retrospective economic review of the existing rules regime, identifying those places in which the existing formal equality framework is already creating injustices for actual and potential litigants. A rulemaking OIRA could also provide expert input on how to accomplish the legitimate goals of the rulemaking system in situations involving particularly complex economic incentives.

An OIRA-style approach would offer several significant advantages compared to the alternative of having committee rulemakers take the entire economic operation completely in-house. First, a rulemaking OIRA could serve the entire judicial rulemaking apparatus, providing expertise of the same sort in the appellate, criminal, bankruptcy, and evidence contexts, in addition to its civil procedure responsibilities. More important, even located within the judiciary, a rulemaking OIRA would exercise a certain level of independence from the Court and other rulemakers. Its primary clients would not be the Judicial Conference, Supreme Court, or even the Standing Committee, but would instead be the various advisory committees charged with doing the dirty work of rulemaking. By serving all of the different advisory committees subject to the same overarching charge to pursue substantively equal procedural justice using economic incentives to differentiate cases, a rulemaking OIRA would enjoy more credibility with a Congress eager to ensure that rulemakers are not trespassing on congressional prerogatives.

Relatedly, the existence of a dedicated cost-benefit arm within the judicial rulemaking structure would enable superior communication between the legislative and judicial branches. A rulemaking OIRA could interact at a relatively high level with congressional bodies like the Congressional Budget Office or even the U.S. Government Accountability Office (“GAO”), both of which conduct similar types of analyses at the legislative and regulatory levels.256

While, admittedly, the solution to create a federal agency is rarely ideal, the creation of a rulemaking OIRA would help rulemakers to meet their obligation to craft rules that do “not abridge, enlarge or modify any substantive right,”257 an obligation that current formally equal rulemaking fails to satisfy.

VI.     Conclusion

Civil procedure has been too long without a theory. The erosion of the theoretical underpinnings of the traditional substance–procedure dichotomy seems to have driven many experts into a sort of nihilistic funk in which the prevailing mentality is “just do the best we can.” Others have identified the same phenomenon as an opportunity to smuggle higher-order social preferences into the procedural realm. But there is a way out of the wilderness. By fully embracing modest but robust procedural justice norms, theorists can begin to see what civil procedure should become—an enterprise dedicated to intra-case accuracy, meaningful participation, and cost balancing. And by thinking of the procedural enterprise explicitly in terms of equality along the economic lines that drive litigation behavior, those same theorists can begin to see what needs to be done to fix the problem.

For the last eight decades, federal civil procedure rulemaking has held steadfast to a formal equality norm that treats all claim types identically. But the decision to employ formally equal transsubstantive procedure seems to have been based upon the erroneous assumption that all federal civil cases were and always would be sufficiently “like” along the relevant dimensions to justify a formally equal approach. If the original Rules architects were aware that internal economic incentives are also part of the engine propelling litigation results, they showed no sign of that in their contemporaneous writings. And yet that oversight has been haunting the federal civil justice system for several decades now, as the litigation docket has moved from fortuitous economic homogeneity at the inception of the rules to today’s far more heterogeneous distribution of cost, information, and risk preference profiles.

Continued application of formal equality in civil procedure necessarily means continued substantive distortions from supposedly neutral rules. But while substantive equality interventions in other contexts are sometimes difficult to control, substantive equality rulemaking is a different story. Cabined by appropriate limitations derived from the Rules Enabling Act, separation of powers concerns, and the limits of rulemakers’ institutional capacity, a substantive equality approach will work. By focusing on balancing the idiosyncratic internal economic incentives associated with certain types of modern litigation, substantive equality rulemaking can better serve the goals of the Rules Enabling Act while simultaneously fostering procedural justice vital to the long-term health and legitimacy of the civil dispute resolution system.


  1. [1]. Charles R. Lawrence III et al., Introduction, in Words that Wound: Critical Race Theory, Assaultive Speech, and the First Amendment 1, ]p 6 (1993).

  2. [2]. Id. at 6–7.

  3. [3]. Richard Delgado, Words that Wound: A Tort Action for Racial Insults, Epithets, and Name-Calling, 17 Harv. C.R.-C.L. L. Rev. 133, 133–49, 181 (1982).

  4. [4]. R.A.V. v. City of St. Paul, 505 U.S. 377 (1992) (striking down law banning cross-burning on First Amendment grounds).

  5. [5]. Mari J. Matsuda & Charles R. Lawrence III, Epilogue: Burning Crosses and the R.A.V. Case, in Words that Wound: Critical Race Theory, Assaultive Speech, and the First Amendment, supra note 1, at 133, 136.

  6. [6]. Robin West, Progressive Constitutionalism 55–56 (1994).

  7. [7]. Lawrence B. Solum, Procedural Justice, 78 S. Cal. L. Rev. 181, 183 (2004).

  8. [8]. See id.

  9. [9]. See id. at 244–52 (describing the accuracy model of procedural justice and its pitfalls).

  10. [10]. See id. at 277–81 (describing the importance of participation in procedural justice).

  11. [11]. See infra Part III.A.

  12. [12]. Stefan Gosepath, Equality, Stan. Encyclopedia Phil. (June 27, 2007), http://plato.stanford.
    edu/entries/equality; see also Peter Westen, The Empty Idea of Equality, 95 Harv. L. Rev. 537, 542–48 (1982) (describing the Aristotelian basis of formal equality).

  13. [13]. See infra Part II.B ; see also Solum, supra note 7, at 242–72 (analyzing several models of current procedural justice and their pitfalls).

  14. [14]. 28 U.S.C. § 2072(a)–(b) (2012).

  15. [15]. Fed. R. Civ. P. 1.

  16. [16]. See West, supra note 6, at 55–56 (describing the feminist critique of the rationality model); Michel Rosenfeld, Substantive Equality and Equal Opportunity: A Jurisprudential Appraisal, 74 Calif. L. Rev. 1687, 1698–1708 (1986) (describing the benefits of a shift to substantive equality). See generally Gary Goodpaster, Equality and Free Speech: The Case Against Substantive Equality, 82 Iowa L. Rev. 645 (1997) (describing problems with substantive equality and the importance of balancing substantive equality with formal equality).

  17. [17]. See Gosepath, supra note 12, and Rosenfeld, supra note 16, at 1696–1708, for general discussions of substantive equality approaches. See also West, supra note 6, at 56, and Lawrence et al., supra note 1, at 6–7, for examples of substantive equality-influenced reform proposals. Consider traditional debates regarding legal responses to racial or gender inequality. Affirmative action programs establishing preferences for disadvantaged groups would be an example of a substantive equality intervention. By contrast, a formal equality approach to the same problem might simply prohibit discrimination on the basis of race and/or gender. As Rosenfeld and others have observed, the devil is in the details for substantive equality proponents, not least because, on the margins, the concept of equality is highly contestable and is damnably difficult to specify operationally in any event.

  18. [18]. See generally Solum, supra note 7.

  19. [19]. One might conceivably argue that such a system is “just” under some higher-order conception of justice, but any such approach inevitably devolves into nihilistic incoherence. See id. at 184–89.

  20. [20]. See Westen, supra note 12, § 5.1(arguing that “equality” is inherently derivative and that the only coherent approach is to define it by reference to the possession and protection of the rights from which it derives). But see Rosenfeld, supra note 16, at 1691–98 (describing limits to “narrow conception of opportunity” when defined as a “negative freedom to be left alone”).

  21. [21]. See Rosenfeld, supra note 16, at 1695–96; see also Solum, supra note 7, at 184–89.

  22. [22]. See Rosenfeld, supra note 16, at 1688–89.

  23. [23]. Prospect-regarding equality norms present their own set of complex subsidiary questions, but we need not engage these here, because the limitations imposed by the civil litigation environment effectively take prospect-regarding equality of opportunity off the table. See infra notes 67–81 and accompanying text.

  24. [24]. 28 U.S.C. § 2072(a) (2012) (emphasis added).

  25. [25]. Id. § 2072(b).

  26. [26]. Ashcroft v. Iqbal, 556 U.S. 662, 663–64 (2009); Bell Atl. Corp. v. Twombly, 550 U.S. 544, 554–63 (2007). Together, these cases establish a pleading standard under which a federal complaint can survive a motion to dismiss only if the non-conclusory factual allegations in the complaint set forth a “plausible” claim.

  27. [27]. For a detailed discussion of the asymmetry problem as applied to pleading standards, see generally Paul Stancil, Balancing the Pleading Equation, 61 Baylor L. Rev. 90 (2009).

  28. [28]. Office of the Curator, Supreme Court of the U.S., The West Pediment: Information Sheet 2 (2003),

  29. [29]. See generally id.

  30. [30]. See Rosenfeld, supra note 16, at 1700–01 (“[S]ince the eighteenth century Western society has valued the proposition that individuals are morally equal.”).

  31. [31]. Id. at 1701.

  32. [32]. See generally Gosepath, supra note 12.

  33. [33]. See generally J.R. Lucas, Against Equality, 40 Phil. 296 (1965); Westen, supra note 12.

  34. [34]. See Lawrence et al., supra note 1, at 13­–15; Stephen N. Subrin, Fudge Points and Thin Ice in Discovery Reform and the Case for Selective Substance-Specific Procedure, 46 Fla. L. Rev. 27, 45–56 (1994).

  35. [35]. See Paul J. Stancil, Close Enough for Government Work: The Committee Rulemaking Game, 96 Va. L. Rev. 69, 71 (2010).

  36. [36]. See Gosepath, supra note 12.

  37. [37]. For an excellent summary of the history of “equality” in anthology form, see generally George L. Abernethy, The Idea of Equality (1959), tracing writings about equality from the Bible through Brown v. Board of Education.

  38. [38]. See Gosepath, supra note 12, § 2 (“Equality in its prescriptive usage has, of course, a close connection with morality and justice in general and distributive justice in particular.”).

  39. [39]. See id. §2.1 (discussing Aristotle’s conceptual framework of equality as laid out in his Nicomachean Ethics and Politics).

  40. [40]. See id. §2.2 (explaining the equality principle developed by Aristotle that people with equal status must be treated equally); see also Abernethy, supra note 37, at 17 (discussing Aristotle’s comments on proportionate equality).

  41. [41]. See Michel Rosenfeld, Affirmative Action, Justice, and Equalities: A Philosophical and Constitutional Appraisal, 46 Ohio St. L.J. 845, 852 (1985).

  42. [42]. Abernethy, supra note 37, at 17; Aristotle, Nicomachean Ethics, reprinted in The Basic Works of Aristotle 927, 1006–07 (Richard McKeon ed. 1941). As a technical matter Aristotelian formal equality is a far broader concept that in fact encompasses most of what would today be categorized as a “substantive equality” approach. But within popular legal theory, the label “formal equality” has come to stand for what Aristotle himself described as “numerical equality,” and I adopt modern terminology for convenience and consistency with current scholarship.

  43. [43]. Abernethy, supra note 37, at 17; Aristotle, supra note 42, at 1007–10; Aristotle, Politics, reprinted in The Basic Works of Aristotle, supra note 42, at 1113, 1187–93 [hereinafter Aristotle, Politics]; see Gosepath, supra note 12, § 2.2 (“[A] form of treatment of others or distribution is proportional or relatively equal when it treats all relevant persons in relation to their due.” (emphasis in original)).

  44. [44]. See Aristotle, supra note 42, at 1006–07.

  45. [45]. See Abernethy, supra note 37, at 17 (“It should be noted that in his numerous comments on equality Aristotle is usually discussing not arithmetic equality, but proportionate equality—to each according to his deserts or virtue. This kind of equality is, for Aristotle, just and necessary for the good life.”).

  46. [46]. See id. (“But Aristotle did not think it undemocratic to exclude from the body of the citizens, not only the slaves, resident aliens, and women, but the mechanics, tradesmen, and husbandmen.”).

  47. [47]. See Catharine A. MacKinnon, Substantive Equality: A Perspective, 96 Minn. L. Rev. 1, 4–5 (2011) (“[Wechsler’s] ‘neutral’ approach, now a doctrinal term for formal equality, has continued, in the guise of principle, to provide the undertow, if not the text, of much if not most constitutional adjudication in the equality area in the United States to the present.”); see also West, supra note 6, at 53–56; Duncan Kennedy, The Critique of Rights in Critical Legal Studies, inLeft Legalism/Left Critique 178, 206–07(Wendy Brown & Janet Halley eds., 2002) (describing problems encountered in feminist legal theory when “[a] series of efforts to use the notion of a right to equal treatment as the basis for a program of law reform ran up against the classic problem of deciding between formal and substantive equality as the content of the right”). West quotes feminist legal theorist Mary Becker for the proposition that “[f]ormal equality ... can effect only limited change” and can actually be counterproductive for women’s rights because it fails to account for a variety of differences between men and women that make identical treatment unjust. See West, supra note 6, at 56 (quoting Mary E. Becker, Prince Charming: Abstract Equality, 1987 Sup. Ct. Rev. 201, 247).

  48. [48]. See, e.g., Anthony E. Cook, Beyond Critical Legal Studies: The Reconstructive Theology of Dr. Martin Luther King, Jr., 103 Harv. L. Rev. 985, 994–95 (1990); Kimberlé Williams Crenshaw, Race, Reform, and Retrenchment: Transformation and Legitimation in Antidiscrimination Law, 101 Harv. L. Rev. 1331, 1334 (1988).

  49. [49]. See MacKinnon, supra note 47, at 4; see also Parents Involved in Cmty. Sch. v. Seattle Sch. Dist. No. 1, 551 U.S. 701, 748 (2007) (plurality opinion) (“The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.”).

  50. [50]. See Rosenfeld, supra note 16, at 1700–01.

  51. [51]. See MacKinnon, supra note 47, at 9–11.

  52. [52]. See Aristotle, supra note 42, at 1006­–07.

  53. [53]. See, e.g., Rosenfeld, supra note 16, at 1698–1708 (analyzing equality of opportunity through the lens of substantive equality); Subrin, supra note 34, at 45–56 (asserting a case for substance-specific procedure). See generally MacKinnon, supra note 47 (offering an in-depth analysis of substantive equality from a gender perspective).

  54. [54]. See MacKinnon, supra note 47, at 3–5 (acknowledging Aristotle’s postulate but ultimately disagreeing with it).

  55. [55]. See id. at 13 (arguing for substantive treatment between sexes, particularly for women).

  56. [56]. See Mari J. Matsuda, Public Response to Racist Speech: Considering the Victim’s Story, in Words That Wound: Critical Race Theory, Assaultive Speech, and the First Amendment, supra note 1, at 17, 17–51.

  57. [57]. See Aristotle, Politics, supra note 43, at 1193–94 (claiming that statesmen “must be freemen and tax-payers”). See generally Maryanne Cline Horowitz, Aristotle and Woman, 9 J. Hist. Biology 183 (1976) (criticizing Aristotle’s view of women).

  58. [58]. See, e.g., Matsuda, supra note 56, at 17–51 (examining substantive equality through the victim’s perspective); MacKinnon, supra note 47, at 15–16 (discussing substantive equality in the context of Reedy v. Evanson).

  59. [59]. See, e.g., Lawrence et al., supra note 1, at 6–7 (defining critical race theory); MacKinnon, supra note 47, at 4–5 (applying Aristotelian formal equality to sex discrimination).

  60. [60]. See Lawrence et al., supra note 1, at 10–15; MacKinnon, supra note 47, at 3–5.

  61. [61]. See Rosenfeld, supra note 16, at 1698–1708. For an example of a formal equality approach to equal opportunity, see generally Westen, supra note 12. Westen’s contention that equality is inherently derivative and can only be expressed coherently in terms of negative freedoms is ultimately a formal equality position, since he contemplates that only law ensures the absence of legal impediments to opportunity. See id. at 543–49.

  62. [62]. Moreover, even those who agree on the goal may well disagree as to the appropriate form of the intervention. Even those who support equality of result sometimes part ways when “leveling down” is the only practical means by which to achieve their ends. See Gosepath, supra note 12, § 5.1 (describing the arguments directed at the theoretical possibility of “depressing the level of everyone’s life” in order to achieve equalization).

  63. [63]. See Westen, supra note 12, at 543–49.

  64. [64]. See Rosenfeld, supra note 16, at 1698–1708.

  65. [65]. See Lawrence et al., supra note 1, at 6–7. Examples of this type of evaluation can be found in Matsuda’s work, see Matsuda, supra note 56, at 38–46 (classifying specific categories of hate speech), and in MacKinnon’s work. See MacKinnon, supra note 47, at 14–15 (examining the law of prostitution).

  66. [66]. I leave aside for now the added complications engendered by the law of unintended consequences.

  67. [67]. See generally Stancil, supra note 35.

  68. [68]. Id. at 76.

  69. [69]. 28 U.S.C. § 2072(a) (2012).

  70. [70]. Id. § 2072(b).

  71. [71]. Stancil, supra note 35, at 76–78, 132.

  72. [72]. See id. (summarizing the process of committee rulemaking).

  73. [73]. See, e.g., Bell Atl. Corp. v. Twombly, 550 U.S. 544, 555–69 (2007) (interpreting Rule 8(a)(2)); Celotex Corp. v. Catrett, 477 U.S. 317, 319–28 (1986) (interpreting Rule 56); Schlagenhauf v. Holder, 379 U.S. 104, 106–22 (1964) (interpreting Rule 35(a)).

  74. [74]. See Stancil, supra note 35, at 77–78 (discussing the timeline and procedure of rule implementation); see also 28 U.S.C. § 2074(a).

  75. [75]. See Stancil, supra note 35, at 97–102 (noting the factors that preclude an inference that congressional silence is tacit approval).

  76. [76]. See id. at 85–87 (pointing out the difficulties of getting a fully informed Congress to act and the costs of locating information on the implications of a decision).

  77. [77]. See id. (referring to the costs of becoming fully informed).

  78. [78]. See 28 U.S.C. § 2072(b) (establishing boundaries in connection with Supreme Court rulemaking authority).

  79. [79]. But see David Marcus, Trans-Substantivity and the Processes of American Law, 2013 B.Y.U. L. Rev. 1191, 1250 (arguing in favor of deviations from transsubstantivity norm precisely when deviation would serve broader substantive policy goals).

  80. [80]. See Stancil, supra note 35, at 99–100 (“Members of both [the Standing Committee and the Advisory Committees] are appointed by the Chief Justice, and can be expected to share the Chief Justice’s preferences to some degree.”).

  81. [81]. See infra Part V.

  82. [82]. See Solum, supra note 7, at 183 (“Meaningful participation requires notice and opportunity to be heard, and it requires a reasonable balance between cost and accuracy.”).

  83. [83]. Solum demonstrates that his version of the accuracy criterion is defensible even under the deflationary view that “procedural justice as an independent criterion of fairness is empty,” a position that he ultimately rejects. Id. at 184–85.

  84. [84]. A coin flip would never be truly accurate in the sense Solum intends, because it reaches a result unrelated to the merits of claims or defenses. A coin flip might reach the “right” result in the sense that it identifies as the winner the same party that would have won if law were applied correctly to the facts, but it will never be “accurate.”

  85. [85]. See Solum, supra note 7, at 247–52 (contrasting “systemic accuracy” with “case accuracy”).

  86. [86]. See id. at 206 (exceptions from intra-case accuracy requirement do not include service to broader social goals of underlying substantive law).

  87. [87]. Id. at 191.

  88. [88]. See id. at 185.

  89. [89]. See id. Many first-time civil litigants would likely identify with that hypothetical under the current system. But things could always be worse.

  90. [90]. See id. at 253 (discussing the Mathews v. Eldridge balancing test, 424 U.S. 319, 334–35 (1976)).

  91. [91]. Solum, supra note 7, at 257; see also id. at 253–59.

  92. [92]. See id. at 257–58.

  93. [93]. I reserve for future work a full exploration of this position, but the basic intuition is straightforward. If committee rulemakers were to attempt resource-balancing, they would: (1) necessarily stray far beyond their range of expertise; (2) invite strategic responses from litigants; and (3) trespass on congressional prerogatives with respect to the threshold determinations about which cases are worth the federal courts’ time.

  94. [94]. See Solum, supra note 7, at 275–89.

  95. [95]. See id.

  96. [96]. See id. at 275–84.

  97. [97]. See id. at 292.

  98. [98]. See infra Part V.

  99. [99]. See supra Part II.A.

  100. [100]. See Subrin, supra note 34, at 41. See generally David Marcus, The Past, Present, and Future of Trans-Substantivity in Federal Civil Procedure, 59 DePaul L. Rev. 371 (2010).

  101. [101]. See Marcus, supra note 100, at 376. In other words, transsubstantive procedure represents a commitment to formal equality such that claim type has no effect upon the procedural treatment each case receives.

  102. [102]. Id. at 376–77.

  103. [103]. Id. at 376.

  104. [104]. Marcus cites Rule 8(a)(2), which establishes a federal civil pleading standard for all cases not expressly exempted from coverage, and Rule 26(a)(1)(B)(vii), which exempts federal student loan collection suits from initial disclosure requirements, as examples of geographically uniform rules—applicable throughout the federal system—that are transsubstantive and substance specific, respectively. Id. at 376 nn.27–28 (citing Fed. R. Civ. P. 8(a)(2); Fed. R. Civ. P. 26(a)(1)(B)(vii)).

  105. [105]. See Marcus, supra note 100, at 377.

  106. [106]. Id. Though Marcus is technically correct, a disuniform system of the sort he suggests may be only nominally transsubstantive to the extent that different levels of case size or complexity are predictably associated with different types of claims. A disuniform system applying separate deposition rules for class actions effectively repudiates the transsubstantivity norm insofar as medical malpractice tort claims are concerned, since malpractice claims rarely if ever qualify for class treatment.

  107. [107]. See id. at 378.

  108. [108]. Id.

  109. [109]. See id. at 379–80. I adopt a structurally similar argument in my own analysis. See infra Part V.

  110. [110]. See Marcus, supra note 100, at 380.

  111. [111]. See Ashcroft v. Iqbal, 556 U.S. 662, 662 (2009) (imposing a “plausibility” requirement upon federal civil pleadings); Bell Atl. Corp. v. Twombly, 550 U.S. 544, 569–70 (2007) (same).

  112. [112]. See Private Securities Litigation Reform Act of 1995, Pub. L. No. 104-67, 109 Stat. 737 (1995) (establishing heightened pleading standard for securities litigation).

  113. [113]. See generally Robert G. Bone, Making Effective Rules: The Need for Procedure Theory, 61 Okla. L. Rev. 319 (2008); Robert G. Bone, The Process of Making Process: Court Rulemaking, Democratic Legitimacy, and Procedural Efficacy, 87 Geo. L.J. 887 (1999) [hereinafter Bone, Making Process]; Judith Resnik, Failing Faith: Adjudicatory Procedure in Decline, 53 U. Chi. L. Rev. 494 (1986); Stephen N. Subrin, How Equity Conquered Common Law: The Federal Rules of Civil Procedure in Historical Perspective, 135 U. Pa. L. Rev. 909 (1987).

  114. [114]. See Marcus, supra note 100, at 381–83; see also Subrin, supra note 113, at 914–18.

  115. [115]. See Marcus, supra note 100, at 381–83; see also Subrin, supra note 113, at 926–29.

  116. [116]. Subrin, supra note 113, at 915.

  117. [117]. See id. at 929–30.

  118. [118]. See Marcus, supra note 100, at 384–86.

  119. [119]. Id. at 384 (quoting Jeremy Bentham, Principles of Judicial Procedure, in 2 The Works of Jeremy Bentham 5, 5–6 (John Bowring ed., 1843)).

  120. [120]. See Solum, supra note 7, at 244–52 (discussing the accuracy theory of procedural justice).

  121. [121]. Marcus, supra note 100, at 385.

  122. [122]. See id.

  123. [123]. Id.

  124. [124]. Id. at 386.

  125. [125]. See id. at 386–92.

  126. [126]. David Dudley Field, What Shall Be Done with the Practice of the Courts?, in 1 Speeches, Arguments, and Miscellaneous Papers of David Dudley Field 226, 229 (A.P. Sprague ed., 1884).

  127. [127]. Marcus, supra note 100, at 389.

  128. [128]. See id. This theme was a near-universal constant among reformers until the rise of legal realism. See, e.g., John Norton Pomeroy, Code Remedies: Remedies and Remedial Rights by the Civil Action According to the Reformed American Procedure § 37, at 57 (4th ed. 1904) (criticizing state practice of continuing to distinguish between types of action despite fact that distinctions had been abolished); Fred A. Maynard, Reform in Legal Procedure., 2 Mich. L.J. 75, 75 (1893) (embracing development of “a legal procedure” that allows controversies to be “settled quickly, inexpensively, and as nearly right as possible”); Edmund M. Morgan, Judicial Regulation of Court Procedure, 2 Minn. L. Rev. 81, 83 (1918) (“The sole object of any procedure system should be the attainment of a just and speedy decision upon the merits, according to the principles of substantive law, at the lowest practicable cost, of all disputes between litigants. The attainment of this end is possible only under a plan which recognizes -the impossibility of foreseeing the effects of the application of any procedural rule in all contingencies, and the impossibility of devising a code which will cover every procedural contingency. Consequently, a satisfactory system must be flexible and must provide an easy method for wise amendment.”); see also Roscoe Pound, The Causes of Popular Dissatisfaction with the Administration of Justice, in Report of the Twenty-Ninth Annual Meeting of the American Bar Association 395, 404–06 (1906) (bemoaning the fact that common law pleading made litigation an adversarial “sport” and not a mechanism for resolution of cases on the merits).

  129. [129]. See Marcus, supra note 100, at 392.

  130. [130]. Charles E. Clark, The Code Cause of Action, 33 Yale L.J. 817, 819 (1924). The “rules as end” phenomenon was just an earlier version of the cost-imposition games with which this Article is concerned; more able and knowledgeable lawyers were able to obtain favorable results for their clients by taking advantage of the pretrial cost disparities engendered by that knowledge gap.

  131. [131]. Marcus, supra note 100, at 394–95. Marcus notes that the Rules’ transsubstantive approach is in part grounded in legitimacy concerns, and in part upon a commitment to “simple rules that were designed to vest ample discretion in trial judges.”

  132. [132]. See id. at 396–97.

  133. [133]. Charles E. Clark, History, Systems and Functions of Pleading, 11 Va. L. Rev. 517, 519 (1925); see also Charles E. Clark, Handbook of the Law of Code Pleading, 26–29 (2d ed. 1947).

  134. [134]. Morgan, supra note 128, at 83.

  135. [135]. See Pierre Schlag, Law and Phrenology, 110 Harv. L. Rev. 877, 896–900 (1997) (explaining the growth, importance, and failings of the legal school that viewed the “[l]aw as [s]cience”).

  136. [136]. See Marcus, supra note 100, at 399–401 (detailing the concerns of the legal theorists that championed the promulgation of the 1938 Federal Rules that the distinction between substance and procedure had blurred to the point of making the distinction unworkable); see also Grant Gilmore, The Ages of American Law 77–86 (1977); Laura Kalman, Legal Realism at Yale 1927–1960, at 115–44 (1986); Resnik, supra note 113, at 502–07 (summarizing the various interests that motivated those who worked on procedural reform in the 1930s).

  137. [137]. See Gilmore, supra note 136, at 91–92 (describing an “Age of Anxiety” in the 1920s and 1930s during which traditional understandings of law eroded); see also Walter Wheeler Cook, “Substance” and “Procedure” in the Conflict of Laws, 42 Yale L.J. 333, 334–35, 343–44 (1933) (suggesting that the “tacit assumption that the supposed ‘line’ between [substance and procedure]” is at most something more akin to the “no-man’s land” between the “face” and the “background” of a soft-focus photograph of a person, and arguing that the terms “substance” and “procedure” mean different things in different contexts); Letter from Ernest G. Lorenzen to Edgar B. Tolman (July 11, 1935), in Records of the U.S. Judicial Conference: Comms. on Rules of Practice and Procedure, 1935–1988, microformed on CIS No. CI-931 (Cong. Info. Serv.) (describing distinction as “necessarily an arbitrary one”); see also Karl. N. Llewellyn, The Bramble Bush 82–83 (1930) (“The differentiation between substantive law and adjective law is an illusion, although the prevalence of this illusion [as of any other] has results in human behavior, and must be taken account of.”). See generally Edgar H. Ailes, Substance and Procedure in the Conflict of Laws, 39 Mich. L. Rev. 392 (1941).

  138. [138]. See Marcus, supra note 100, at 401 (citing Thurman Arnold, The Rôle of Substantive Law and Procedure in the Legal Process, 45 Harv. L. Rev. 617, 643–45 (1932)).

  139. [139]. Arnold, supra note 138, at 643.

  140. [140]. Id.

  141. [141]. Id. at 645.

  142. [142]. See Clark, supra note 133, at 519.

  143. [143]. Bone, Making Process, supra note 113, at 894.

  144. [144]. Edson R. Sunderland, An Inquiry Concerning the Functions of Procedure in Legal Education, 21 Mich. L. Rev. 372, 381–82 (1923). Sunderland’s argument is simultaneously incorrect and prophetic. Though he wrongly concludes that there is a relatively clear divide between “substance” and “procedure,” he was prescient with respect to his assertion that “procedural” revolution could take place without causing social unrest. Activists from all points on the political spectrum have smuggled effectively “substantive” reforms into law by way of ostensibly “procedural” changes; it seems clear that the hue and cry from such maneuvers has been far less than one would have expected from frontal assaults on the “substantive” law.

  145. [145]. See Bone, Making Process, supra note 113, at 897.

  146. [146]. See Resnik, supra note 113, at 498–507 (discussing the drafters’ differing opinions to “make some sense of why the drafters did what they did”).

  147. [147]. This is true despite the fact that the substance/procedure dichotomy already rested on shaky theoretical ground at the inception of the Federal Rules and had lost most of its conceptual foundations by the 1970s. See Bone, Making Process, supra note 113, at 900–07 (exploring the history surrounding the rise of procedural skepticism). I reserve discussion of other critiques of the transsubstantivity norm—both descriptive and theoretical—for infra Part IV.B. I do note, however, that the continued popularity of the substance/procedure dichotomy in practice may be at least in part because it has significant instrumental value, Marcus, supra note 100, at 379–80, and is perhaps to some degree a product of the same sorts of intuitions that animate Rosenfeld’s “postulate of equality.” See supra note 61 and accompanying text.

  148. [148]. See supra Part II.

  149. [149]. A more coherent defense of formal equality norms exists, but it has little rhetorical traction. One might defend formal equality not on the “first-best” ground that all participants are in fact “like” across all relevant dimensions, but rather on the ground that it is the best that can be done given the challenges associated with implementation of genuinely fair substantive equality approaches. While this defense is powerful, and indeed seems to have persuaded some commentators, I demonstrate below that the limited substantive equality procedural rulemaking I propose is far closer to a “first-best” solution.

  150. [150]. See Resnik, supra note 113, at 508–11 (discussing the various types of cases with which the drafters of the Rules were familiar, and their similarities).

  151. [151]. See infra note 166 and accompanying text.

  152. [152]. See infra notes 169–70 and accompanying text.

  153. [153]. To be clear, cases in 1938 obviously involved cost distribution, information distribution, and risk-preference profiles. They were simply fairly consistent from case to case, regardless of claim type. See infra note 166 and accompanying text.

  154. [154]. I recognize that the economic incentives facing litigants are sometimes more complex. See generally Stancil, supra note 27 (explaining that full economic costs and benefits of litigation are not limited to gross expenditures and judgment/settlement amounts).

  155. [155]. See Jeffrey J. Rachlinski, Gains, Losses, and the Psychology of Litigation, 70 S. Cal. L. Rev. 113, 127 (1996) (discussing the policy implications of the framing theory of litigation). “Systematic irrationality” refers to persistent and ubiquitous behavior inconsistent with traditional notions of economic rationality. While anecdotal, idiosyncratic irrationality on the part of a small percentage of individuals does not undermine substantial proposals based upon an assumption of economic rationality, systematic irrationality does.

  156. [156]. See Chris Guthrie, Prospect Theory, Risk Preference, and the Law, 97 Nw. U. L. Rev. 1115, 1116 (2003) (explaining “prospect theory” tenet that, while people do try to maximize outcomes, they “fail to do so in systematic and predictable ways”).

  157. [157]. Daniel Kahneman & Amos Tversky, Prospect Theory: An Analysis of Decision Under Risk, 47 Econometrica 263, 263 (1979).

  158. [158]. See Guthrie, supra note 156, at 1118 (explaining the same phenomenon in people choosing between options that will result in a loss (risk-seeking) or a gain (risk-averse)); Rachlinski, supra note 155, at 157–60 (analyzing data showing that defendants exhibited risk-seeking behavior, while plaintiffs exhibited risk-averse behavior).

  159. [159]. See Rachlinski, supra note 155, at 129–30.

  160. [160]. See Citizens United v. Fed. Election Comm’n, 558 U.S. 310, 392 (2010) (Scalia, J., concurring).

  161. [161]. Both within a case and, for repeat litigants, across cases.

  162. [162]. See Rachlinski, supra note 155, at 118.

  163. [163]. And possibly public value. See generally Stephen B. Burbank, The Rules Enabling Act of 1934, 130 U. Pa. L. Rev. 1015 (1982).

  164. [164]. Of course, Congress retains the ability to encourage claims of even that size if public policy so requires. See 17 U.S.C. § 504(c)(1) (2012) (authorizing copyright plaintiffs to seek “statutory damages ... of not less than $750” in lieu of actual infringement damages). The point is not that no nominally small claims are worth the time of the federal courts but rather that there are some legally valid claims for which the expenses of a federal action exceeds the social benefit.

  165. [165]. There are both structural and constitutional reasons for rejecting a means-regarding form of substantive equality that attempts to balance the parties’ resources. The system already does this to some extent in particularly egregious cases. For instance, pro se litigants are generally afforded considerable procedural deference relative to represented parties. But it is also worth noting that procedural justice does not necessarily require a single-shot framing in which cost-balancing must be performed on a case-by-case basis. Even if we could figure out how to balance resources, we might not wish to. First, the market will sort good lawyers from bad over time. More importantly, that sorting will in turn serve the system by helping to sort good claims from bad. Good lawyers will presumably be able to identify good claims, and at least as a systemic matter, this might help to define the threshold level of “value” worthy of the courts’ time and resources in functional terms. This assumes the market for legal services is relatively well-functioning, of course. If the distribution of legal skill across various doctrinal areas of law is normally distributed, resource limitation will be a problem only in the odd or idiosyncratic case. While this concern is very real, there is unfortunately little that can be done to fix this problem. If instead the distribution is discontinuous or bimodal, it might be a bigger deal.

  166. [166]. See supra Part III.B (discussing the historical progression following the Rules). The explosion was multifaceted, corresponding roughly with the rise of the regulatory state during the New Deal, the enactment of broad-scope remedial legislation (e.g., civil rights, environmental), the elimination of privity requirements in products liability cases, the increased willingness of courts to infer implied private rights of action with respect to existing statutes, the amendment of the Federal Rules of Civil Procedure to permit small claims class actions under Rule 23(b)(3), and Congress’s increased general willingness to embrace the private-attorneys-general model of enforcement in addition to or even in lieu of robust government enforcement. See infra Part IV.B.1 (discussing the evolution of the Rules).

  167. [167]. See Robert G. Bone, Modeling Frivolous Suits, 145 U. Pa. L. Rev. 519, 542–46 (1997) (discussing information asymmetry); Guthrie, supra note 156, at 1118–19 (discussing effects of risk preferences on litigation behavior). See generally Stancil, supra note 27 (analyzing the economics of the civil litigation and specifically pleading).

  168. [168]. We cannot prove the Rules framers’ subjective states of mind either empirically or by reference to the historical record. They simply did not talk about these things. But as demonstrated below in Part IV.B.1.i–ii, we can gain significant insight from an examination of the civil dockets of the time. This Article makes a very tentative, largely anecdotal start on that project; the literature would benefit substantially from a more thoroughgoing examination of the civil litigation backdrop against which the Federal Rules architects operated.

  169. [169]. Note that this is not the same thing as uniform distribution of those characteristics among plaintiffs and defendants. Transsubstantive procedure can deal with asymmetries; it cannot deal with asymmetries distributed in a nonuniform way.

  170. [170]. As I tentatively define the term. See Stancil, supra note 27, at 120–21, 160–63 (emphasizing in the context of an argument for selectively stricter pleading standards that cost disparity must be calculated using both internal and external cost functions and arguing that the triggering disparity should be sufficiently large to assuage an implementing court’s error and transaction cost concerns).

  171. [171]. See 1936 Att’y Gen. Ann. Rep. 162–79; Stephen B. Burbank, Vanishing Trials and Summary Judgment in Federal Civil Cases: Drifting Toward Bethlehem or Gomorrah?, 1 J. Empirical Legal Stud. 591, 619–20 (2004); Resnik, supra note 113, at 508–09. See generally Am. Law Inst., A Study of the Business of the Federal Courts: Part II Civil Cases (1934).

  172. [172]. See 1936 Att’y Gen. Ann. Rep., supra note 171, at 162–79; Am. Law Inst., supra note 171, at 59.

  173. [173]. See 1936 Att’y Gen. Ann. Rep., supra note 171, at 162–79. See generally Am. Law Inst., supra note 171. Although the cost-mismatch dynamic arguably comes into play occasionally in cases involving the United States as a party, I generally bracket those cases on the assumption that the United States’ incentive structure when it litigates is not motivated by simple economic desires to profit maximize.

  174. [174]. See 1936 Att’y Gen. Ann. Rep., supra note 171, at 162–79.

  175. [175]. In other words, federal civil cases between exclusively private parties.

  176. [176]. See Stancil, supra note 27, at 132–33, 160–63 (arguing in favor of a heightened pleading standard only when a sufficiently significant cost disparity exists).

  177. [177]. See Resnik, supra note 113, at 514 (suggesting that many claims were left unlitigated because many plaintiffs lacked the resources to pursue them, and modern correctives were not yet in place).

  178. [178]. See Stancil, supra note 27, at 127 (summarizing data on pretrial costs in tort litigation).

  179. [179]. See Resnik, supra note 113, at 498–526. Resnik states:

    To summarize, in the past fifty (and particularly in the past twenty) years, the docket of the federal courts has changed in several significant ways. First and foremost in the minds of many jurists, there are a lot more cases to decide. Second, some of those cases involve disputes between the very powerful (often but not exclusively the government) and the very poor. Third, new federal legislation has served as the basis for complex cases in which the underlying disputes may involve multiple incidents and require the production of numerous witnesses and thousands of documents. Fourth, an important set of cases involves group litigation, in which representatives seek recognition to pursue the rights of individuals not physically before the court and sometimes not even aware of the pendency of the lawsuit. Fifth, in those cases in which court-based activity actually occurs, the litigation is sometimes protracted and complex, and the opportunities for adversarialism in the pretrial process or at trial are multiplied. Sixth, because that adversarialism occurs under the umbrella of the court-based discovery rights established by the Federal Rules of Civil Procedure and in the context of an individual calendar system and fee-shifting statutes, judges see a good deal of aggressive or inept attorneys. Finally, many other institutions within the society have been obligated to perform court-like functions in a court-like manner, and the growing use of such alternatives raises questions about the appropriate domain of the federal courts.

    Id. at 525–26 (footnotes omitted).

  180. [180]. See id. at 514.

  181. [181]. See Act of July 2, 1890, ch. 647, § 7, 26 Stat. 210, superseded by Clayton Act, § 4 (current version at 15 U.S.C. § 15 (2012)); History of the Antitrust Division, U.S. Dep’t Justice,
    atr/about/division-history.html (last updated June 25, 2015).

  182. [182]. History of the Antitrust Division, supra note 181.

  183. [183]. 15 U.S.C. § 15(a). As an aside, it is plausible to characterize the passage of the private-right-of-action portions of the Clayton Act as an enforcement tweak intended to move the law toward the negotiated social outcome the Sherman Act represented.

  184. [184]. Although the analysis is admittedly a bit rough-and-ready, quick examination of reported antitrust cases decided before 1938 bears this out. I performed a Westlaw search in the “ALLFEDS” database looking for all pre-1938 cases containing the words “Sherman Act,” “antitrust,” or “anti-trust,” but not containing “United,” “federal,” “people,” or “State” in the case title. Of the first 25 results returned, six cases involved actual antitrust claims; all involved a business entity suing another business entity for allegedly anticompetitive behavior. The captions of those six cases give a sense of the paradigmatic antitrust claim of the times. See generally Terminal Warehouse Co. v. Pa. R.R. Co., 297 U.S. 500 (1936); Radio Corp. of Am. v. Raytheon Mfg. Co., 296 U.S. 459 (1935); Ind. Farmer’s Guide Publ’g Co. v. Prairie Farmer Publ’g Co., 293 U.S. 268 (1934); Levering & Garrigues Co. v. Morrin, 289 U.S. 103 (1933); Cent. Transfer Co. v. Terminal R.R. Ass’n, 288 U.S. 469 (1933); Story Parchment Co. v. Paterson Parchment Paper Co., 282 U.S. 555 (1931).

  185. [185]. Given the nature of antitrust claims (for example, the paradigm business-versus-business antitrust claim that the defendant illegally monopolized a market), it is plausible but by no means certain that private antitrust plaintiffs of the time tended systematically to be smaller than the antitrust defendants they sued.

  186. [186]. See Ind. Farmer’s Guide Publ’g Co., 293 U.S. at 272–74 (plaintiff farmer’s guide publisher alleging that a cooperative of other publishers conspired to drive plaintiff out of business by attempting to monopolize the sale of commercial advertising to farm-related businesses).

  187. [187]. In the case of antitrust specifically, another dynamic is in play: changes in the substantive law over time. Through the 1960s, antitrust law was relatively friendly to claims made by one competitor against another. See generally Klor’s, Inc. v. Broadway-Hale Stores, Inc., 359 U.S. 207 (1959). But by the late 1970s, the tables had turned, and the law clearly favored claims by injured consumers over competitor suits. See Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 489 (1977) (establishing an “antitrust injury” requirement such that antitrust plaintiffs must prove “injury of the type the antitrust laws were intended to prevent and that flows from that which makes defendants’ acts unlawful”); see also Brooke Grp. Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 224–25 (1993) (imposing a “recoupment” requirement upon predatory pricing claims in a competitor suit alleging predatory pricing by rival cigarette manufacturer). This increasing hostility to competitor suits, while arguably positive for the overall direction of antitrust law, put additional pressure on rules of procedure by favoring consumer suits in which problematic cost disparity is substantially more likely. But see generally Ill. Brick Co. v. Illinois, 431 U.S. 720 (1977) (limiting federal antitrust suits to direct purchasers of the good or service allegedly subject to illegal restraint of trade).

  188. [188]. I say “typically.” There were, of course, exceptions—including antitrust. But those exceptions likely constituted too small a fraction of the federal docket to negate the value of transsubstantive procedure generally. The vast majority of the relevant private party v. private party federal civil docket of the day consisted of diversity-based common law case.

  189. [189]. Even the punch cards used for data storage (only by governments and the very largest corporations) at the time were technically hard copy.

  190. [190]. Chester Carlson invented electrostatic photocopying in 1938, but the first automated photocopier did not appear until 1960. Document duplication techniques that were available (e.g., photostatic copying, carbon-paper copying, typographic transcription) were labor intensive, expensive, and relatively low quality. David Owen, Making Copies, Smithsonian Mag. (Aug. 2004),

  191. [191]. Although not as important to the ultimate analysis as cost parity/disparity distribution, it is also worth noting that the civil litigation environment before the adoption of the Federal Rules was characterized by search costs substantially higher than those present today. See supra Part IV.B.1.a. Before the arrival of the Information Age, it was simply more difficult to obtain and analyze relevant information; reformers thus recognized that byzantine complexity was one of the primary weaknesses of the common law regime. See Field, supra note 126, at 230 (proposing “general conformity in the different cases, so that, while the particular circumstances of each may receive such remedy as they require, the outline of the proceedings in all may be the same, and a knowledge of the course pursued in one may serve as a guide in the others”). And reformers were critical of common law procedure’s tendency “to make litigation an adversarial sport and not a mechanism for resolution of cases on the merits.” Marcus, supra note 100, at 394; see also Pound, supra note 128, at 405. It would have been almost impossibly difficult for a layperson to navigate the intricacies of common law or even code pleading; in a very real way, litigating lawyers in the pre-transsubstantivity era commanded high fees as much for their procedural expertise as for their substantive legal acumen. Moreover, it would have been almost impossibly expensive for a layperson to acquire the knowledge necessary to navigate a common law system; the expense associated with this knowledge acquisition is another argument in favor of a transsubstantive approach.

  192. [192]. Another argument in favor of abandoning strong form transsubstantivity for a default rule approach is dynamism. It is possible that evolving technology will again shift information and cost dynamics heterogeneously across claim types. For example, the use of algorithmic search technology and technology-assisted review (“TAR”) in discovery may bring some claim types closer to cost parity while having little effect on others. For a general introduction to the relevant technologies, see generally The Sedona Conference, The Sedona Conference Best Practices Commentary on the Use of Search & Information Retrieval Methods in E-Discovery, 15 Sedona Conf. J. 217 (2014). The use of TAR technologies can in certain cases drastically reduce the amount of time and effort required to parse large quantities of digitally stored information. Id. Rulemaker flexibility is key to keeping pace with these sorts of changes.

  193. [193]. See, e.g., Civil Rights Act of 1964, Pub. L. No. 88-352, 78 Stat. 241 (codified as amended in scattered sections of 42 U.S.C.); William H. Timbers & David A. Wirth, Private Rights of Action and Judicial Review in Federal Environmental Law, 70 Cornell L. Rev. 403, 405 & n.8 (1985) (listing federal environmental statutes with express “citizen suit” provisions).

  194. [194]. See, e.g., Greenman v. Yuba Power Prods., Inc., 377 P.2d 897, 901 (Cal. 1963); Henningsen v. Bloomfield Motors, Inc., 161 A.2d 69, 77 (N.J. 1960).

  195. [195]. If not necessarily inappropriately.

  196. [196]. 42 U.S.C. § 2000e-2 (2012).

  197. [197]. See McDonnell Douglas Corp. v. Green, 411 U.S. 792, 802–03 (1973) (describing burden-shifting analysis for proof of discrimination by indirect evidence).

  198. [198]. By demonstrating membership in a protected class, application and qualification for job opening, rejection for the position, and that the position remained open thereafter. Id. at 802.

  199. [199]. See id. at 804–05.

  200. [200]. A typical citizen-suit provision authorizes “any person [to] commence a civil action ... against any person” alleged to be in violation of the relevant environmental standard. See 15 U.S.C. § 2619(a)(1) (2012) (expressly authorizing citizen suits in the context of the Toxic Substances Control Act). Citizen-suit plaintiffs’ own activities are irrelevant because they were not involved in any of the transactions that allegedly constitute wrongdoing, and because they need not have suffered any damages or other particularized harm in order to bring suit. This Article is primarily concerned with private defendants because the paradigm private case seems most likely to produce troubling cost dynamics. Thus, the citizen-suit environmental litigation in question would typically involve only private defendants. This is not to say that government litigants are irrelevant to my analysis, they just deserve independent consideration. In addition, this should not be read to suggest that citizen-suit environmental litigation needs its own enforcement regime. There are too many variables at play to allow such a conclusion after only a cursory analysis.

  201. [201]. Fed. R. Civ. P. 23(b)(3).

  202. [202]. The astute reader will recognize that the 1986 amendments that created the modern small-claims class action themselves present concerns under my analytical framework. This is true, especially to the extent that the 1966 Amendments to the Federal Rules of Civil Procedure represent the output of a committee rulemaking process effectively unchecked by congressional oversight. See generally Stancil, supra note 35. I generally bracket this concern for now, taking the existence of the small claims class action as a given. Moreover, my concerns with Rule 23(b)(3) are solely institutional—Congress clearly would have had the authority to address the “death by 1000 cuts” problem by adopting Rule 23(b)(3) expressly.

  203. [203]. See Stephen B. Burbank, The Class Action Fairness Act of 2005 in Historical Context: A Preliminary View, 156 U. Pa. L. Rev. 1439, 1484–89 (2008) (describing the 1966 amendments).

  204. [204]. Fee-shifting statutes do not wholly solve the problem, if for no other reason than they routinely award a “reasonable” attorney’s fee. Reasonableness is typically measured against the recovery obtained, and in the absence of particularly strong social policies and statutory enforcement norms, a risk-averse attorney would be unlikely to take a five dollar case in the hope of earning a multi-million dollar fee.

  205. [205]. In stark contrast to the reported cases from before 1938, see supra note 184, where an identical Westlaw search without the date limitation suggests that consumer class actions are the order of the day in modern antitrust litigation. Of the first 25 hits from the search (all opinions from 2011), eight cases involved actual antitrust claims. Of those eight, six were class actions, and five were traditional consumer class actions.

  206. [206]. I bracket for now the impact of the 2015 proportionality amendments to the Federal Rules. Suffice it to say I am deeply skeptical that these amendments will solve the problem.

  207. [207]. It is at least plausible that for a few types of modern claims, increases in cost asymmetry variance are attributable to new cost asymmetries favoring defendants. For example, the proliferation of professional malpractice claims may well mean that relative to 1938, there are also more cost asymmetric cases in which the defendant enjoys the advantage.

  208. [208]. See supra notes 186–94 and accompanying text.

  209. [209]. See Resnik, supra note 113, at 508 (noting that “paradigm cases” included “private monetary dispute[s]” as well as “[e]quity cases” and “labor injunctions and receiverships”).

  210. [210]. Robert Bone, Professor, Univ. of Tex. Law Sch., Settlement Panel Remarks at the 2010 Conference on Civil Litigation (May 11, 2010),

  211. [211]. For example, a wealthy plaintiff who suffered relatively minor injuries and/or no significant disruption to her income stream may be more willing to take risks than a poor plaintiff whose injuries rendered her unable to work for a significant period.

  212. [212]. Substance-specific rulemaking is only appropriate to correct risk profile heterogeneity to the extent that heterogeneity is predictable and systemic with reference to claim types. I am not arguing for a bespoke procedural regime that attempts to delve into the minds of individual litigants on a case-by-case basis to ascertain risk tolerance and preference. Some variance from litigant to litigant within claim types is inevitable. It is also simply not amenable to procedural correction, no matter what its distortive effects on litigation outcomes may be.

  213. [213]. See, e.g., Rachlinski, supra note 155, at 129–30.

  214. [214]. See generally Note, Risk Preference Asymmetries in Class Action Litigation, 119 Harv. L. Rev. 587 (2005).

  215. [215]. For example, environmental or structural reform litigation plaintiffs, or defendants whose motivations are at least in part “for the principle of the thing.”

  216. [216]. Among other things, if risk preference profiles differ from case type to case type, using the same set of rules to govern every type of case will create optimal incentives for litigants rarely, and then only by accident. For example, Rule 11 governs the filing of frivolous claims or defenses (among other things). See Fed. R. Civ. P. 11. If a certain class of plaintiffs (or their attorneys) is risk-seeking, the traditional approach to sanctions (functionally predicated on either risk neutrality or risk aversion) simply will not work.

  217. [217]. For example, a risk-averse litigant is likely to exercise particular care in the fulfillment of her discovery responsibilities, while a risk-seeking litigant may be more willing to take aggressive positions that limit her upfront costs, at the risk of later incurring the court’s wrath.

  218. [218]. The story of cost and information asymmetry distribution since the effective date of the Rules is one of increased asymmetry and increased variance. As the federal docket has expanded to include new statutory causes of action based upon a defendant’s state of mind (e.g., civil rights claims based upon intent), citizen suits, and small claims class actions, the cost and information economics change accordingly. See infra Part IV.D. It is less clear that the distribution of risk profiles has changed as dramatically since 1938. While a narrative supporting increased variance exists, it is equally possible that we have had substantial variance the whole time, and that the Rules architects simply did not consider it in their deliberations. See generally Resnik, supra note 113 (examining the intentions Rules’ architects).

  219. [219]. The idea of “risk neutrality” and the concept of risk preference really only entered the economic academic consciousness with the publication of Theory of Games and Economic Behavior in 1944. See John von Neumann & Oskar Morgenstern, Theory of Games and Economic Behavior
    162–65 (1944) (describing a quantitative theory of risk aversion). The discipline of “law and economics” got its start no earlier than 1946, the year Aaron Director joined the faculty of the University of Chicago Law School. See Douglas Martin, Aaron Director, Economist, Dies at 102, N.Y. Times (Sept. 16. 2004), While the framers of the Federal Rules were both prominent and skilled lawyers and legal academics, nothing in their histories suggests any familiarity with economic theory.

  220. [220]. Attorney’s fees statutes raise interesting questions in the context of this Article. Congress can and does use attorneys’ fees provisions to tip the scale when it concludes that the default enforcement regime (the so-called “American Rule”) does not generate sufficient enforcement. But this in turn raises at least three additional important questions mostly beyond the scope of this Article. First, as an empirical matter, do attorneys’ fees statutes work and under what conditions? In two separate Supreme Court cases, the majority and dissenting opinions expressed equally uninformed but stridently divergent views on the effectiveness of attorney’s fees statutes under different conditions. See Buckhannon Bd. & Care Home, Inc. v. W. Va. Dep’t of Health & Human Res., 532 U.S. 598 (2001); Evans v. Jeff D., 475 U.S. 717 (1986); see also generally John J. Donohue III, Opting for the British Rule, or If Posner and Shavell Can’t Remember the Coase Theorem, Who Will?, 104 Harv. L. Rev. 1093 (1991) (discussing the costs and benefits of the American and English rules regarding attorneys’ fees); John Hause, Indemnity, Settlement, and Litigation, or I’ll be Suing You, 18 J. Legal. Stud. 157 (1998). Second, should the courts be allowed to imply attorney’s fees provisions or depart from the plain language of existing attorney’s fees statutes to further “correct” perceived imbalances? See Hensley v. Eckerhart, 461 U.S. 424, 429 & n.2 (1983) (holding that statutory language authorizing attorneys fee awards to “prevailing parties” in civil rights litigation should be interpreted such that prevailing plaintiffs “should ordinarily recover an attorney’s fee” while prevailing defendants should recover fees “only where the suit was vexatious, frivolous, or brought to harass or embarrass the defendant”). Finally, can or should committee rulemakers diverge from the American Rule on a substance-specific basis—i.e., should attorney’s fees be in the toolkit, given the recommendations I make? Regardless of the theoretical answer to that question, the practical answer is almost certainly “no.” It is hard to envision even the most articulate and thoughtful committee rulemakers successfully defending a decision to shift from the American Rule to the English Rule, much less vice versa.

  221. [221]. Conley v. Gibson, 355 U.S. 41, 45–46 (1957), abrogated by Bell Atl. Corp. v. Twombly, 550 U.S. 544 (2007).

  222. [222]. Ashcroft v. Iqbal, 556 U.S. 662, 680–84 (2009); Bell Atl. Corp., 550 U.S. at 556–58 (2007).

  223. [223]. See, e.g., Joshua Civin & Debo P. Adegbile, Restoring Access to Justice: The Impact of Iqbal and Twombly on Federal Civil Rights Litigation 2 (2010),
    sites/default/files/Civin_Adegbile_Iqbal_Twombly.pdf; Kevin M. Clermont & Stephen C. Yeazell, Inventing Tests, Destabilizing Systems, 95 Iowa L. Rev. 821, 840 & n.70 (2010); Arthur R. Miller, From Conley to Twombly to Iqbal: A Double Play on the Federal Rules of Civil Procedure, 60 Duke L.J. 1, 10 (2010). See generally Joseph A. Seiner, The Trouble with Twombly: A Proposed Pleading Standard for Employment Discrimination Cases, 2009 U. Ill. L. Rev. 1011 (2009).

  224. [224]. See, e.g., Civin & Adegbile, supra note 224, at 7, 9–10, 16–19; Clermont & Yeazell, supra note 224, at 824­–32; Miller, supra note 224, at 14–15, 71–72; Seiner, supra note 224, at 1018–24.

  225. [225]. See, e.g., Mark Herrmann et al., Plausible Denial: Should Congress Overrule Twombly and Iqbal?, 158 U. Pa. L. Rev. PENNumbra 141, 145–47 (2009); Keith N. Hylton, When Should a Case Be Dismissed? The Economics of Pleading and Summary Judgment Standards, 16 Sup. Ct. Econ. Rev. 39, 41–42 (2008).

  226. [226]. See generally Stancil, supra note 27.

  227. [227]. Id. at 146–49.

  228. [228]. Id.

  229. [229]. Id. at 146.

  230. [230]. Id. at 150–51.

  231. [231]. Bell Atl. Corp. v. Twombly, 550 U.S. 544, 570 (2007).

  232. [232]. Ashcroft v. Iqbal, 556 U.S. 662, 681 (2009).

  233. [233]. See Stancil, supra note 27, at 146–49 (assessing the effect of strict pleadings on various types of claims). It is likely that the TwomblyIqbal standard is not such a response, even for the set of cases motivating the decision.

  234. 235. See id. at 126–27, 146.

  235. [235]. See Fed. R. Civ. P. 26(b)(1) (providing the new discovery rule).

  236. [236]. See generally, e.g., Stephen B. Burbank, Proportionality and the Social Benefits of Discovery: Out of Sight and Out of Mind?, 34 Rev. Litig. 647 (2015) (discussing the revision to the rules); Patricia W. Hatamyar Moore, The Anti-Plaintiff Pending Amendments to the Federal Rules of Civil Procedure and the Pro-Defendant Composition of the Federal Rulemaking Committees, 83 U. Cin. L. Rev. 1083 (2015) (arguing against the new restrictions on discovery); Craig B. Shaffer, The “Burdens” of Applying Proportionality, 16 Sedona Conf. J. 55 (2015) (discussing discovery abuse); see also generally Suja A. Thomas & Dawson Price, How Atypical Cases Make Bad Rules: A Commentary on the Rulemaking Process, 15 Nev. L.J. 1141 (2015) (describing commentary on both sides of proposed amendments to the Rules). In addition, consider that the Advisory Committee on Civil Rules heard from 120 testifying witnesses (including the author) and received over 2,300 written comments regarding the 2015 discovery amendments. See Thomas Y. Allman, The 2015 Civil Rules Package as Transmitted to Congress, 16 Sedona Conf. J. 1, 4–5 (2015) (describing the amendment process). For a digital repository containing the written comments, see Proposed Amendments to the Rules of Civil Procedure, Regulations, (last visited Mar. 30, 2017).

  237. [237]. See Stancil, supra note 27, at 146–49 (assessing the effect of strict pleadings on various types of claims).

  238. [238]. See Thomas & Price, supra note 237, at 1151­–57.

  239. [239]. See, e.g., Charles Guerrier, Comment Letter on Preliminary Draft of Proposed Amendments to the Federal Rules of Civil Procedure (Feb. 16, 2014),
    D=USC-RULES-CV-2013-0002-1490; Gordon Leech, Wisconsin Association for Justice, Comment Letter on Preliminary Draft of Proposed Amendments to the Federal Rules of Civil Procedure (Feb. 19, 2014),; David Mehan, Comment Letter on Preliminary Draft of Proposed Amendments to the Federal Rules of Civil Procedure (Feb. 18, 2014),

  240. [240]. See Thomas & Price, supra note 237, at 1155 (“For many on the [Advisory] Committee, the ‘typical’ litigation experience appears to be the atypical case. Additionally, research shows the limited ability of humans to consider all the facts, circumstances, and implications of a problem. Instead, they inordinately focus on what is before them. So, the rulemakers may be able to see only the problem presented to them—that is, high costs and delays in discovery, thus motivating the rule change.” (footnote omitted)).

  241. [241]. See Michael E. Lackey, Jr. & Andrew Pincus, Comment Letter on Preliminary Draft of Proposed Amendments to the Federal Rules of Civil Procedure (Feb. 18, 2014), https://www.

  242. [242]. Fed. R. Civ. P. 26(b)(1).

  243. [243]. The same dodge is present in Iqbal: “Determining whether a complaint states a plausible claim for relief will, as the Court of Appeals observed, be a context-specific task that requires the reviewing court to draw on its judicial experience and common sense.” Ashcroft v. Iqbal, 556 U.S. 662, 679 (2009).

  244. [244]. Again, leaving aside the problems engendered by delegating too much discretionary authority to judges.

  245. [245]. See Fed. R. Civ. P. 26(d)(2).

  246. [246]. It might also be worth considering whether the Supreme Court’s 1966 adoption of the small claims class action rules embodied in Fed. R. Civ. P. 23(b)(3) creates the same problem. Rulemakers’ inability or unwillingness to see the dark side of small claims class actions and opt-out classes is troubling, but the inverse rights-without-remedies problem was significant as well.

  247. [247]. With some help. See infra Part V.D.

  248. [248]. See Stephen N. Subrin, The Limitations of Transsubstantive Procedure: An Essay on Adjusting the “One Size Fits All” Assumption, 87 Denv. U. L. Rev. 377, 387–88 (2010). See generally Marcus, supra note 100 (discussing the value of transsubstantivity).

  249. [249]. See Stephen B. Burbank, Pleading and the Dilemmas of “General Rules, 2009 Wis. L. Rev. 535, 543; Marcus, supra note 100, at 417–26.

  250. [250]. An argument exists for the proposition that rulemakers should pursue the 1938 baseline instead of procedural justice in the abstract. Congress, it might be said, approved a particular social policy when it approved the initial Federal Rules of Civil Procedure, and there is no reason to believe that Congress necessarily intended plaintiffs and defendants to be evenly matched. There are at least two reasons to reject this interpretation. First, as I note below and will explain in greater detail in future work, there are good reasons for rulemakers to act as if that Congress did intend that sort of balance. Second, the economic analysis will be difficult enough when performed against a neutral economic baseline; it is likely impossible for rulemakers to calibrate against a different baseline with anything approaching accuracy. Notwithstanding all of this, the 1938 paradigm case remains important for at least one critical reason. See infra Part V.B.

  251. [251]. See generally Paul Stancil, Congressional Silence and the Statutory Interpretation Game, 54 Wm. & Mary L. Rev. 1251 (2013).

  252. [252]. See Stancil, supra note 35, at 71–72, 76–77.

  253. [253]. See Exec. Order No. 12,866, 58 Fed. Reg. 51,735 (Sept. 30, 1993).

  254. [254]. Id.

  255. [255]. See, e.g., Cass R. Sunstein, The Office of Information and Regulatory Affairs: Myths and Realities, 126 Harv. L. Rev. 1838, 1839 (2013); Regulations and the Rulemaking Process,, https://www. (last visited Mar. 30, 2017); see also Office of Mgmt. & Budget, 2015 Report to Congress on the Benefits and Costs of Federal Regulations and Agency Compliance with the Unfunded Mandates Reform Act 1 (2015),
    web/20170112091133/ (last visited Mar. 30, 2017).

  256. [256]. See, e.g., 2 U.S.C. § 602 (2012) (setting forth duties of Congressional Budget Office); Budget and Accounting Act of 1921, Pub. L. No. 67-13, §§ 301–318, 42 Stat. 20 (establishing the General Accounting Office and setting forth its duties); GAO Human Capital Reform Act of 2004, Pub. L. No. 108-271, § 8(a), 118 Stat. 811, 814 (codified as amended at 31 U.S.C. § 702 (2012)) (redesignating the General Accounting Office as the Government Accountability Office).

  257. [257]. See 28 U.S.C. § 2072(b) (2012).


Professor of Law, J. Reuben Clark Law School, Brigham Young University.

Thanks to Judith Resnik, Stephen Burbank, and Robert Bone for their comments on earlier versions of this Article. Thanks also to Emery Lee, Suja Thomas, Aaron Nielson, Carolina Nuñez, and Ronnell Andersen Jones, as well as participants at the First Annual Civil Procedure Workshop, the Junior Federal Courts Conference, and the faculties at Indiana University—Bloomington and Brigham Young University for their comments on earlier versions of this piece. Special thanks to Jarrett Szcezny, Alexandra Sandvik, Matthew Jaynes, and Jesse Houchens for able research assistance. All errors are my own.