This is a book about judging. But to talk about judging is not to talk about a single, uniform activity or concept. The specific functions of the various institutions called courts, and the particular tasks performed by the array of officers called judges, vary widely. The United States Supreme Court plays a considerably different role in our society than does, say, the district court in Faribault County, Minnesota, and the day-to-day lives of the jurists who occupy the bench in the two courts are hardly similar at all. So, too, with judges on the United States Bankruptcy Court, the federal Court of Appeals for Veterans Claims, or the Maryland Orphans Court, each of whom dons their robe with a distinct set of concerns and priorities in mind, and each of whom presides over proceedings governed by largely distinct bodies of law.

Yet these roles are not entirely disconnected. They are all part of the same larger system. Imagine a Faribault County deputy were to conduct a search and seizure in a manner not clearly governed by existing Fourth Amendment law, which in turn led to criminal charges. The defendant in the resulting case could raise a claim of the sort that would conceivably make its way from the courthouse on Main Street in Blue Earth, Minnesota, through the Minnesota state court system, and to the courthouse across from the Capitol on First Street in Washington, DC, in which the Supreme Court sits. Each judge or justice along the way would have a different view of their job, in part because their court serves a different function in the larger system, and in part because each would have a unique understanding of such things as how the law works, how it ought to work, and what the appropriate role of courts and judges in our system of governance is. All of this suggests, as Professor Helen Hershkoff put the point, “that judicial power is not a fixed or natural category, but rather a socially constructed (if also socially constrained) institutional form ….”Footnote ¹ One must be careful about generalization.

Still, it seems inevitable that there should be threads connecting these roles and that by moving away from particulars and toward abstraction we ought to be able to identify, in a general way, the functions courts are designed to serve and the role judges are assigned to play. We ought to be able to establish some form of baseline against which to assess courts’ functioning and by reference to which to consider modifications. Much of the rhetoric in Rucho, in both Roberts’s majority and Kagan’s dissent, proceeds from the understanding that there are certain tasks that are outside the scope of judicial competence and therefore there are certain problems the nature of which make them not the kind that courts are suited to solve. Courts are among society’s tools, the logic runs, and as with any tool, they are best suited for certain uses and not at all suited for others.

To a degree this will be a moving target. The nature of society’s needs, the institutional and procedural architecture of courts, and thus the nature of the functions they are designed to serve all undoubtedly change with time and circumstance. Some, however, remain constant. The most basic, surely, is providing a mechanism for the peaceful resolution of disputes. Another involves the making and refinement of the standards by reference to which that happens. These two functions – often referred to as the dispute-resolution and law-declaration functions – are frequently in tension.Footnote ² The best resolution of a given individual dispute could easily conflict with the best rule covering that general type of dispute. “Hard cases make bad law,” as the saying has it. This, too, is part of the disagreement in Rucho. The majority does not quite say that the best result would be to strike the maps down, but it expresses its clear distaste and disapproval of them. Its concern is with the next case, and the unending stream of cases it foresees after that, most of which would be more difficult and all of which collectively would threaten to undermine the judiciary’s institutional legitimacy. The dissent does not share that fear. It believes that this case can be resolved, and that similarly egregious future cases can be resolved, without encouraging challenges to every set of maps. The dynamic arises in more routine situations as well. There are, for example, good reasons to have clear rules against certain conduct – trespassing or driving above a posted speed limit – even if one knows that in many individual cases no harm will come from the violation of these rules, and indeed there may even be compelling reasons to violate them. The values driving law declaration sometimes outweigh those pressing for an ideally just resolution of each dispute.

As the example that opened the chapter suggests, the relative prominence of these functions varies according to a court’s place in the larger system. The judge on the Faribault County District Court will focus on resolving the narrow question of whether this search and seizure in this case violated the Fourth Amendment. The judge’s orientation will be almost exclusively retrospective, focused on making the best sense of existing authorities and determining how they apply to the set of facts at hand. The Supreme Court justice’s concerns, by contrast, will be primarily forward-looking. The resolution of the specific dispute will matter, but it will serve primarily as a vehicle for considering and deciding more general issues. The focus will be less on this case than on this type of case. The fundamental point, again, is this: Courts are functional institutions, and each court and judge plays a different part in advancing the larger purposes of the judicial system. Those purposes are often in tension with one another, and they do not remain fixed from one court system or historical era to the next. Generalization is difficult.

Almost always, however, the invocation of judicial power arises in connection with the call to resolve a dispute of some sort, and it seems inevitable that the origin story of the various institutions we call courts is rooted in that basic social need. Human history has undoubtedly featured a wide variety of mechanisms for resolving conflicts, ranging from duels, feuds, and war to reliance on the wisdom of some trusted elder. But violence has obvious costs, and trusted elders work best within the confines of a small community in which the disputants are both more likely to themselves trust, or at least respect the authority of, such a decision-maker. Thus, at some point in our social evolution, presumably at a stage when interactions – and thus conflicts – between strangers became sufficiently common, we developed the notion of the institutional neutral – the court – as a decision-maker. Party A believes that they have been wronged by Party B. They are not members of the same community, except in some relatively broad sense, so there is no elder or other informal authority figure available whose judgment both will respect. The community has prohibited violence and, in its place, provided a forum for the airing of A’s grievances. It has staffed that forum with decision-makers who are neutral by design. And the community stands ready to use its collective power to secure a remedy from B if A’s grievances turn out to be legitimate.

Providing a peaceful means of dispute resolution, then, is the first and core function of courts, and it entails a certain institutional logic. Professor Martin Shapiro surveyed the literature on courts and extracted from it “an ideal type, or really a prototype, of courts involving (1) an independent judge applying (2) preexisting legal norms after (3) adversary proceedings in order to achieve (4) a dichotomous decision in which one of the parties was assigned the legal right and the other found wrong.”Footnote ³ As a rough cut, this captures well enough the typical conception of the American judiciary that we might think of it as the “standard model” of adjudication: An independent, reactive judge resolving a dispute framed by the parties and doing so, via what Professor Paul Carrington called “a simple-minded formalism,”Footnote ⁴ by drawing upon identifiable, preexisting legal standards – which is to say, lest the point pass without emphasis, that courts in this model exist primarily to apply rather than to make law.

The standard model has a clear intuitive appeal. Party A will be more inclined to resort to the courts, rather than violence or some other means of self-help, if A feels confident that the judge is subject to no influences that might distort consideration of A’s claim and that the dispute will be decided according to law that was known (or at least knowable) to both parties at the time of the conduct in issue. It further helps that this will happen via a proceeding in which A has an opportunity to make their case in an effort to persuade the judge and, in some cases, jury that the law and the facts are on their side.

Discussions of the judicial role in the American system often assume this vision of the role as a baseline and more than that tend to give it normative significance. Chief Justice Roberts’s opinion in Rucho serves as an example. The problem with judicial attempts to resolve claims of partisan gerrymandering, he repeatedly emphasized, is that there are not (and, in his view, cannot be) sufficiently precise standards to generate resolutions. Because it, therefore, would be impossible to have preexisting norms, any decision in any case would appear to be the product of the judges rather than the law. What’s more, given the existence of real and perceived connections between the justices and the political actors who appointed and confirmed them, coupled with the inherently partisan nature of disputes involving election maps, it is nearly impossible for the Court or a justice to appear to be independent. The problems presented by partisan gerrymandering are real, as the Chief Justice acknowledged. But, he reasoned, it is beyond the proper role of the judiciary to try to fix them. The intuitive appeal of this conception of the judicial power is such that most of the state supreme courts that have decided cases involving claims of extreme gerrymandering in the wake of Rucho have simply assumed that it also describes the power they possess. They have done so despite there being compelling reasons – among them the fundamentally different natures of the governments brought into being by state constitutions and the fact that in many states the justices of the supreme court are elected – not to blithely make that assumption.Footnote ⁵

Tempting though it may be to accept the standard model as an appropriate archetype, careful consideration reveals that it captures, at best, only a portion of the entire array of courts and the entire range of judicial functions. Courts in the United States do much more than deciding disputes between individuals. Judges preside over bankruptcy proceedings and class actions, oversee the administration of school systems, counsel drug offenders, and generally find themselves involved in just about every aspect of society. Many of these functions fit uneasily, at best, within the standard model. Indeed, Shapiro concludes, “If we examine what we generally call courts across the full range of contemporary and historical societies, the prototype fits almost none of them.” Things are more varied, and nuanced, than the standard model suggests.

The standard model’s components, which the remainder of this chapter will consider in turn, consistently reveal themselves to be “essentially contested concepts” – ideas as to which there are no generally accepted definitions and no available benchmarks by which to conclude that one definition or another is correct. The core of the judicial role is more elusive than first meets the eye, and judges inevitably – and necessarily – have more decisional space in which to operate than the model suggests. Operating within that space inescapably involves the exercise of judgment.

Judicial Independence

That we want independent judges is axiomatic, what we mean by that considerably less so. As Professor Frank Cross put the point, “Few cows are more sacred than judicial independence. Yet the concept of an independent judiciary is more commonly apotheosized than analyzed.”Footnote ⁶ Judges certainly do not, and should not, enjoy complete freedom to do whatever they please. A crucial question thus becomes independence from what or from who? Certainly from the parties themselves, but who else? From elected officials? From voters or from the public more generally? Must independence look and function the same in a small community in which everyone knows everyone else as it does in a large metropolitan area? Can judicial independence be the same thing in Delaware or Wyoming as it is in California or New York? Is judicial independence an end in itself? Or does it serve a purpose?

The answer to the last question seems clear. Judicial independence exists to serve a purpose rather than for its own sake. If, as I have suggested, courts are functional institutions, then it follows that the role of judicial independence is to provide courts and judges with the freedom to best serve those functions. Independence is a component in a larger, complex arrangement of governmental institutions, relationships, and processes that collectively serve to mediate and advance the various overlapping and conflicting ends of society. Its primary function is undoubtedly to give judges the space to follow the law (whatever it might mean to do so) rather than the wishes of some other group or audience except, ideally, insofar as those groups or audiences are able to exert pressures that work to push courts toward greater or more consistent adherence to the law. But it can also serve to enable judges to depart from the letter of the law when doing so is necessary to achieve a just outcome. The key point is that, as with the judiciary itself, judicial independence as a concept does not exist for its own sake but rather as a means to an end or, more accurately, a collection of ends. And the identity and nature of those ends might vary from one context to another, and they might at times be in tension if not outright conflict with one another.

But providing more space to judges involves tradeoffs. More independence not only enables judges to follow the law wherever it leads but also allows the ability for other factors to influence decision-making – things such as ideology, bias, and other improper considerations of the sort that Chapters 2 and 3 will explore. The underlying dynamic is a version of an agency problem.Footnote ⁷ The principal – in this case, society or a subset of it such as the people of a state – wants the agent – its judges – to faithfully carry out the law and to administer justice, and, in general, to do so despite the fact that it will often be in the short-term interest of powerful groups for judges not to do so. That entails giving judges the freedom to discern what the law and justice require but not so much that they are free simply to do as they please. As Professor Stephen Burbank has highlighted, viewing the matter from this perspective makes it apparent that judicial independence and judicial accountability are not “discrete concepts at war with each other” but rather “complementary concepts that can and should be regarded as allies. … An accountable judiciary without any independence is weak and feeble. An independent judiciary without any accountability is dangerous.”Footnote ⁸ Further, Burbank argues, this means that we cannot regard the contours of judicial independence, or the judicial power more generally, as fixed. The judiciary is “one part of a complex governmental apparatus” that changes over time in response to social, intellectual, and other trends and developments. To conclude that the balance of independence and accountability, and the specific mechanisms for maintaining it, were fixed at the moment the judiciary was created is to unwittingly err in one direction or the other, simply because the balance of forces influencing the judiciary will inevitably have shifted over time. It follows that judicial independence cannot be a monolithic concept. State and federal judiciaries, and governmental structures more generally, differ in ways that have implications for the functions of the judiciary as a whole and thus for the role, nature, and characteristics of judicial independence in serving them.

Not surprisingly, then, we see judicial independence operationalized in a range of ways, which vary in both selection procedures and conditions of service. Article III federal judges enjoy the most robust independence, which is assured by life tenure conferred after appointment by the President and confirmation by the Senate, with the only available mechanism for removal being impeachment under the Constitution. Judges on Article I courts, by contrast, generally serve fixed terms and often may be removed pursuant to statutory provisions that reach a greater range of conduct than what’s covered by the Constitution. On some Article I courts, there are limits on the number of judges who may be affiliated with a given political party. Selection methods and terms of service in state courts vary widely. Many judges are elected; some are appointed. Most serve fixed terms, typically with the possibility of reelection or reappointment, but a few enjoy something approaching the life tenure of federal judges. As we will see in Chapter 5, the selection processes for both federal and state judges have changed over time, sometimes in ways that have implications for judicial independence.

These differences in the conditions under which judges serve – which are to a large extent differences in the degree and nature of their independence – undoubtedly affect their behavior as judges. Federal court judges, with their near-complete insulation from political pressure, are often regarded as better protectors of rights than their state court counterparts.Footnote ⁹ For example, studies have suggested that state court judges behave differently when reelection draws near, such as by being less inclined to cast votes that favor criminal defendants.Footnote ¹⁰ At the same time, elected state court judges may have a legitimate claim to intervene more aggressively in the workings of government since they draw on a base of democratic legitimacy that federal judges do not have. None of these arrangements is better or worse in some absolute manner. Each must be assessed in a functional sense according to standards set by reference to the nature of the power the courts hold, the nature of the power that courts are responsible for checking, and the ultimate aims of government. To paint with a broad brush, a governmental structure with the preservation of liberty as its primary goal would require a different sort of judiciary than one designed primarily to facilitate majority rule or one intended to advance some vision of the common good.Footnote ¹¹ The questions cannot be answered in the abstract.

Preexisting Legal Norms

There is likewise more to the prototype’s second component – preexisting legal norms – than is first apparent. Standing behind it is the rule of law ideal. The rule of law is, of course, most often contrasted with the “rule of men,” and it is meant to encapsulate the idea that, in Professor Richard Fallon’s formulation, “the law – and its meaning – must be fixed and publicly known in advance of application, so that those applying the law, as much as those to whom it is applied, can be bound by it.”Footnote ¹² The law should apply alike to everyone, with no prohibitions created in an ad hoc or after-the-fact fashion. So stated, however, the rule of law is an ideal, and, as Fallon observes, “It is strongly arguable that no plausible legal system could avoid departing from it in some respects.”Footnote ¹³ Any moderately difficult case illustrates the point. The contours of the law are unclear, and the dispute before the court concerns whether this situation falls inside or outside of the boundary. Did the police violate the Fourth Amendment when they conducted this search or not? Is this a statute that regulates commerce, such that Congress had the power to pass it, or not? Did this defendant’s conduct in this unique set of circumstances amount to disorderly conduct? There will often be uncertainty at the edges of the law, and someone will have to decide whether some set of facts fell just within or just outside of the boundary. To the losing side the law will not seem to have preexisted the ruling. Some degree of departure from the fully realized ideal is thus inevitable. The question thus arises: What sorts of departures from the ideal we are willing to tolerate and to what extent?

Once one moves away from broad definitions, it becomes clear that the rule of law is another essentially contested concept. The literature here is vast and Fallon’s is hardly the final word on the matter, but his analysis effectively illustrates the concepts. As he suggests, some generalization is possible. A nonexhaustive list of the features of a system that facilitates rule-of-law values includes: (1) generality – the rules governing conduct are uniformly applicable and impartially applied to government officials and ordinary citizens alike; (2) publicity – rules can guide conduct only if they are made available to the public; (3) clarity – rules should be understandable in terms of both their substance and their scope; (4) consistency – rules should not contradict one another; (5) congruence – the rules as applied should match the rules as announced; (6) stability – the law should remain consistent so as to enable people to plan and coordinate action over time; and (7) the existence of impartial mechanisms for the enforcement and application of law. Distilled even further, in the ideal rule-of-law world, any of us ought to be able to know in advance, with certainty, whether our conduct will run afoul of the law and further that our conduct will be assessed against rules and through procedures applied consistently and equally to everyone else.

How best to achieve these in an imperfect world is the subject of further disagreement. Fallon suggests that there are four primary versions of the rule-of-law ideal, each based on a distinct understanding of the nature of law and the preferred nature of its implementation. The “historicist ideal type” emphasizes pedigree, based on the premise that lawmaking is an inherently political process that ought to be undertaken by politically accountable actors, such that law consists primarily of rules created in advance by legitimate lawmaking authorities. The primary goal is thus to figure out what rules those authorities created. Originalism as an approach to constitutional interpretation is an example. It regards the meaning of the Constitution to have been fixed at the moment of ratification, such that the primary task of satisfying the rule of law in that context is to determine and comply with that original meaning. The “formalist ideal type” emphasizes the need for legal standards that have clear edges allowing for easy application and for those who are subject to them to be able to plan their conduct with relative certainty. On this view, legislatures should avoid imprecise language and courts should prioritize deciding cases in a way that remedies any legislative failures to produce the requisite clarity. The more the endeavor can be made to resemble an algorithmic process, the better. The “Legal Process ideal type,” as its name suggests, emphasizes process. The concern is not with the pedigree or clarity of rules, but with whether the means by which those rules are created and applied are fair, governed by reason, and appropriately transparent. A citizen may not be able to know in advance whether their conduct falls within the scope of the rules, but they should at least be confident that their position will receive due consideration. Finally, the “substantive ideal type” looks to whether the law satisfies external criteria of morality. On this view, a rule could meet the demands of the other three types and still be fundamentally unjust. The American legal system when it countenanced slavery stands as an example of a system that could be described as satisfying the first three versions of the ideal while failing the fourth. The four versions Fallon identifies are not mutually exclusive. One could, for example, be an originalist who also favors rule-like legal standards, thereby emphasizing the concerns at the heart of the first and second versions, as was the case with Justice Antonin Scalia.Footnote ¹⁴ Or one could favor a process-based approach that is also bound by substantive limits. To a point, at least, it is a matter of emphasis.

Among the menu of options Fallon provides, the formalist ideal type is the most intuitive and enduring. The popular, superficial conception of the idea is most famously encapsulated in Chief Justice Roberts’s opening statement at his Senate confirmation hearing. There he likened the role of the judge to the role of a baseball umpire: “Judges and justices are servants of the law, not the other way around. Judges are like umpires. Umpires don’t make the rules; they apply them. The role of an umpire and judge is critical. They make sure everybody plays by the rules. But it is a limited role. Nobody ever went to a ball game to see the umpire.”Footnote ¹⁵

Underlying Roberts’s characterization is the embrace of what is often referred to as a formalistic approach to judging. Formalism is yet another term that is widely used but often not defined. As Professor Frederick Schauer put it, “Even a cursory look at the literature reveals scant agreement on what it is for decisions in law, or perspectives on law, to be formalistic, except that whatever formalism is, it is not good.”Footnote ¹⁶ The most common usage is one that imagines judges as ideally engaged in an essentially mechanical activity, an elaborate paint-by-numbers exercise in which answering legal questions is simply a matter of ascertaining the correct answers without exercising discretion or allowing values or ideology to influence the process.Footnote ¹⁷ On this view, legal interpretation is often conceived of as a linguistic or semantic exercise involving the identification and definition of the legal concepts that apply to a given situation, whether they are found in a statute or constitution or in the language of past judicial decisions. The goal is to figure out what, for example, a statutory provision says rather than to focus in some more abstract or general sense on the problem it is designed to solve or how it was meant to work. In Professor Adrian Vermeule’s formulation, the recipe calls for judges to “stick close to the surface-level or literal meaning of clear and specific texts, resolutely refusing to adjust those texts by reference to the judges’ conceptions of statutory purposes, legislators’ or framers’ intentions or understandings, public values and norms, or general equity.”Footnote ¹⁸ Relatedly, formalists tend as well to prefer legal standards that take the form of rules, which work to limit the number of factors to be accounted for in decision-making and thus to further the appearance of objectivity.Footnote ¹⁹

The Supreme Court’s reasoning about the scope of the Commerce Clause during what is known as the Lochner era provides one example. The clause gives Congress the power “to regulate Commerce … among the several States,” and the Lochner Court resisted Congress’s reliance on the clause as the source of its authority for broad regulation of the economy. For the majority of justices on that court, much turned on the meaning of the word “commerce,” which Congress did have the ability to regulate, as contrasted with things like manufacturing, agriculture, and so on, which it did not. From there, it was just a matter of determining which box the regulated activity in question fell into. Thus, in United States v. E.C. Knight Co., the Court held that Congress lacked the power to prevent an acquisition that led to the formation of a monopoly in the sugar refining industry on the grounds that what Congress sought to do was an attempted regulation of manufacturing rather than commerce – even though a monopoly in refining would unquestionably have effects on the market for sugar.Footnote ²⁰ Consequences didn’t matter, labels did.

A second aspect of formalism relates to the conceptual nature of law. In the formalist view of the world, legal concepts have a necessary and fixed content. They are, in Justice Oliver Wendell Holmes’s famous critical characterization of the view, part of “a brooding omnipresence in the sky,”Footnote ²¹ a perfectly integrated, logically consistent, comprehensive, and timeless structure of concepts for the regulation of all aspects of human affairs.Footnote ²² This was a crucial idea during the heyday of the common law, when most of the law was found in judicial decisions. The judge’s task under this approach is to make the best use of the available evidence – in the form of past decisions and the rules and concepts that can be derived from them – to reason toward the proper result in the present case. Done correctly, according to this logic, the processes of legal reasoning are objective and determinate, and thus neither their workings nor their conclusions depend on the identity of the judge who undertakes them. The job of the judge faced with a novel question is not to make law but instead to discover it.

In the standard story about the development of American law, formalism was the dominant approach until the emergence of the Legal Realist movement in the early decades of the twentieth century. The Realists were, in Professor Lawrence Friedman’s characterization, “a variegated bunch” about whom it is easier to identify what they were not than what they were.Footnote ²³ And what they were not were formalists. Among the Realists’ revelations was that legal rules are indeterminate – they do not and cannot provide clear answers to all questions – and that as a result legal reason cannot be the mechanistic, deductive process that formalists desire. Holmes again: “General propositions do not decide concrete cases. The decision will depend on a judgment or intuition more subtle than any articulate major premise.”Footnote ²⁴ Rules can never be precise enough to provide the guidance that formalism demands across a range of cases. There will be gray areas in which the application of a rule is unclear. A rule pointing toward a decision in one direction will sometimes be accompanied by a different, equally valid, and equally applicable rule pointing in the other direction. Often the law will allow multiple and perhaps even contradictory outcomes.

Take, for example, a case with which I begin the semester with my first-year Criminal Law students. Under Wisconsin law, the penalties for committing a theft are more severe if the property is taken “from the person of another.” In State v. Hughes, the defendant took a purse that was hanging on the handle of a wheelchair. One could argue, as the defendant did, that “from the person of another” requires that the victim actually be touching the property. At first glance, this seems to provide a clear line. But further consideration reveals that clarity to be an illusion. It can’t be correct to require that the victim’s skin must be touching the stolen item, because that approach would exclude takings of items from pockets, which are separated from their owner by at least one layer of fabric, but which surely fall within the scope of “from the person of another.” If we conclude that clothing is an extension of the person, then we could say the same about a wheelchair. But if we accept that logic, then we have to ask whether it extends to things such as bicycles, golf carts, or cars. Other efforts to define the phrase likewise fail to produce a distinct, easy-to-administer line that mechanically and satisfactorily governs all cases. Another approach would be to look behind the language of the provision to seek an understanding of the reasons underlying it. Perhaps takings from the person of another are penalized more severely because they tend to make a situation more dangerous by increasing the likelihood of a confrontation. Or maybe it’s that the victim feels more violated by such a taking, as contrasted with a theft of property when the victim is not present. There are, of course, multiple permutations and extensions of all these arguments, and it quickly becomes apparent that no mechanical solution is available. Appropriate “legal” reasoning is available to justify either result in the case, though none of it is as simple as calling a ball or strike. Judgment is necessary, together with resort to something beyond a narrow, formalistic view of law. The words of the statute do not alone provide an answer to the questions. The same is true of concepts extracted from judicial decisions.

That was not the Realists’ only insight. Not only are rules indeterminate, but they are often also misleading. A newcomer to an area of law might read the applicable legal materials and conclude that a case involving a given set of facts will be resolved in a particular way, just as one unfamiliar with the customs of the United States might imagine that drivers will be fined for even minor violations of speed limits. But a study of the pattern of results over a range of cases may reveal that something other than the factors expressly identified in the law itself best explains the results. An oft-repeated phrase relating to speeding has it that, from the perspective of a police officer, the true trigger is double-digit speeding: “Nine you’re fine, ten you’re mine.”Footnote ²⁵ Professor Karl Llewellyn, among the most prominent of the Realists, captured the idea in the distinction between “paper rules” – the formal legal rules that purport to govern the resolution of dispute – and the “real rules” that actually do.Footnote ²⁶ Judges may not even realize the extent to which their decisions depart from what the governing law appears, to an outsider, to require.

The Realists also attacked the conceptual component of formalism. Law, they asserted, is not some timeless set of fixed and self-contained principles but rather a constantly changing mass of rules connected to and arising out of a shifting collection of social norms and mores. Here, too, Holmes stands as the acknowledged progenitor of the idea:

The job of the judge (especially the common-law judge) working within this understanding cannot be simply to reason from one abstraction to the next. It instead requires attentiveness to the law as it actually functions and consideration of the social context in which it arose and the nature of the problems it was designed to address. A judge resolving a novel case or issue, then, necessarily makes law insofar as the decision in the present case has binding effect in future cases. To some greater or lesser degree, depending on the case, the standard model’s preexisting legal norms do not – and cannot – exist.

There is more. The realism thus far described, which focused on the limits of legal materials and of judges in dealing with them, is that of the “rule skeptics.” There was a second strand as well, which consisted of the “fact skeptics.” From their perspective, not even perfectly determinate legal rules would necessarily result in completely predictable decision-making. The reason is that the facts of each individual case must be discovered and characterized. They must be assigned significance and weight, and credibility determinations must be made. Legal rules might spell out which facts matter, but they cannot remove the uncertainty and unpredictability from the process through which those facts are determined to be present or not. One judge might conclude a plaintiff has provided enough evidence to survive summary judgment, the next might not. So, too, with key evidentiary rulings. The jury’s verdict might turn on the happenstance of its membership, or on the fact that a key witness had a bad day or gave an improbably good performance. None of this can be reliably predicted in advance.Footnote ²⁸

Finally, there is the obvious fact that courts do make law. The law-declaration function is an acknowledged part of the judicial role, especially but not only for judges on apex courts. Indeed, most of the law that existed at the nation’s founding, and through much of its history, was common law generated via case-by-case decision-making. Once one rejects the idea of a brooding omnipresence, it can only be the case that it was the judges who were making, rather than finding, the law. Courts’ role as primary authors of the law governing our day-to-day interactions has largely shifted, with legislatures and administrative agencies assuming primary responsibility for that role. But statutes and regulations inevitably leave gaps and include ambiguities that courts must address.

Having laid out the Realist critique, a pair of caveats are in order. The first is to acknowledge, as most Realists did, that there are many easy cases in which the rules can be applied in a determinate fashion to generate clear results. To take an easy example, speed limits present no problems of indeterminacy in their paper format, and the applicable real rule likewise provides no room for manipulation in the case of someone driving 120 miles per hour on a highway. Beyond that, a great many potential disputes do not even appear in the legal system at all given the clarity of the law’s application to the situation involved, and compliance with clear legal rules enables people to structure their affairs in such a way as to prevent disputes from arising in the first place. When disputes do arise, a substantial portion are likewise, when viewed from a detached perspective, equally clear in terms of what the law requires. Easy cases abound.Footnote ²⁹

The second caveat is that the traditional story of the transition from formalism to realism is, at best, overdrawn. Professor Brian Tamanaha studied judicial and scholarly commentary concerning the processes of judging during the era in which formalism reputedly dominated and concluded that almost no one in that era actually held the beliefs the Realists ascribed to them.Footnote ³⁰ Rather than thinking that judges mechanically applied law to facts, judges “had a realistic awareness of the complexities involved in judging and of the foibles of human judges.”Footnote ³¹ They understood “that the law has inconsistencies, runs out, and routinely comes up against unanticipated situations and that judges possess a substantial degree of flexibility when working with legal materials. It was obvious to observers that the law can interpreted differently by judges with different views.”Footnote ³² The difference between the Realists and the judges of the formalist era, Tamanaha concludes, is not that they held different understandings of the way in which law and judging work, but that they held different understandings of the significance of those realities. The Realists were skeptical of judging; their predecessors were not. “While the historical jurists recognized that there are social influences on judicial reasoning, they did not describe this as a flaw in judging, but as an essential – necessary, beneficial – link that keeps law and society on the same plane.”Footnote ³³ The play in law’s joints was, in the current lingo, a feature rather than a bug.

Adversarialism

The standard model’s third component – the adversarial process – stands at the heart of popular conceptions of how the American legal system works. Lawyers act as zealous advocates for their clients, and the resulting clash provides the decision-maker, whether judge or jury, with the raw materials it needs to get to the truth. This has consequences for the judicial role. It implies judicial neutrality and passivity – a Sphinx-like judge who acts only after the parties have provided all the input they have to provide.Footnote ³⁴ In Professor Lon Fuller’s classic development of the idea, adjudication’s distinguishing characteristic “lies in the fact that it confers on the affected party a peculiar form of participation in the decision, that of presenting proofs and reasoned arguments for a decision in his favor.”Footnote ³⁵ It follows that anything that enhances party participation also enhances adjudication, while anything that hinders participation or otherwise renders it less effective thereby undermines adjudication. A judge who assumes anything but a reactive posture thus threatens to undermine the legitimacy of the process. Without realizing it, such a judge may come to favor one side as a result of being proactive. The simple act of asking questions, for example, could lead the judge to unwittingly develop a preference by enticing the judge to view the dispute from one party’s perspective or could reveal something about the judge’s thought processes that a party could use to their advantage. And such an intervention risks creating a perception of prejudgment, regardless of whether the judge has actually departed from neutrality.Footnote ³⁶

Here, too, matters are not as straightforward as the standard model would lead us to believe. There are both theoretical and practical objections to this conception of the judicial role. Complete passivity seems undesirable, and so does neutrality insofar as it is viewed as requiring that a judge work solely with what the parties have presented. Not all litigants are equally situated. There are often differences in wealth, sophistication, and access to counsel.Footnote ³⁷ Sometimes this will mean that one of the parties will have misunderstood or failed to recognize the applicability or significance of a legal theory.Footnote ³⁸ One side will have been better able to develop the facts. Often such imbalances will be consistently present in cases, such as those that pit sophisticated, repeat players against unsophisticated individuals.Footnote ³⁹ Sometimes the litigation will have consequences that fall on people whose interests the parties do not share. To remain passive in the face of an obvious imbalance in power or misalignment of incentives would mean to adhere to the requirements of an idealized conception of judging at the expense of justice as it is generally conceived. A baseline obligation of responsiveness to the parties’ claims and arguments remains desirable, lest judges use a perceived freedom to depart from the contours of the dispute as the parties have presented it to skirt claims and arguments they find inconvenient or inexpedient,Footnote ⁴⁰ but meeting that responsibility need not entail comprehensive passivity.

Moving from the abstract to the concrete, there are various ways in which the judicial process has evolved over the course of American history that involve departures from, or at least consequential variations on, the standard model’s prescriptions. A striking early innovation in the American system compared to the English system from which it descended was the product of geographical necessity. England long followed the “oral tradition,” in which all proceedings take place orally and in open court, with very little reliance on written submissions.Footnote ⁴¹ But the requirement that parties and judges be in the same location for as long as necessary to work through a case at each stage was impracticable in the United States, with its comparatively sparse and geographically dispersed population and small number of lawyers.Footnote ⁴² The US courts accordingly shifted to the use of written briefs and opinions. And not only do American courts issue written justifications for their decisions (as opposed to delivering them extemporaneously in open court), but they also do so via a single opinion for the court, rather than separate, seriatim opinions setting forth the views of each judge or justice. Writing differs from orality in its influence on thought, and, in the case of multi-member courts, the process of generating a single opinion joined by a majority of judges leverages the collective wisdom of the group in a way distinct from a process in which each judge offers their own conclusion. The evolution did not stop there. Law clerks, rather than judges, now serve as the initial authors of nearly all opinions issued by appellate courts, and the once unheard of nonprecedential opinion is now a common thing. Their precise nature is difficult to identify, but such changes surely have subtle but nonetheless real effects on the way in which the adversarial process plays out. We will return to them in Chapter 4.

There are other common things courts do that fall outside the bounds of the standard model’s call for adversary proceedings. Probate and family courts, for example, often deal with matters that do not involve parties who are opposed to one another in any real sense. Drug courts and the various problem-solving courts that have developed in their wake feature at their core judges who are actively involved in steering a path forward for defendants. All of these, and more, are precisely what Shapiro observed: entities that we call courts in which judges routinely act in ways that do not fit comfortably within the prototype.

A Dichotomous Decision

The preceding two components of the standard model have implications for the types of disputes that are appropriate for resolution by courts and thus bring us to the fourth, the idea of the dichotomous decision declaring a winner and a loser. The resolution of A’s claim against B will be that one or the other of them will prevail. Either B wronged A in some legally cognizable sense or did not, and if the wrong occurred, B will be responsible for somehow making it right. Compromise solutions are not on the table. In their influential-though-unpublished materials entitled The Legal Process, Professors Henry Hart and Albert Sacks argued that this is a necessary part of the standard model, that it follows from an understanding that adjudication works best in the context of disputes that can be resolved by reference to preexisting general rules, that “an either-or question of the existence vel non of any right to a remedy seems ordinarily to be essential to an adjudicable dispute.”Footnote ⁴³ The exercise of ad hoc discretion may be appropriate in the context of fashioning a remedy, they allow, but “only when it is strictly ancillary to a clear-cut determination of right, and pertains only to what is necessary to the due effectuation of the right.” On this view, decisions that require a forward-looking management of affairs or that otherwise involve what Fuller called polycentric disputes, which involve consideration of an array of interacting and incommensurable factors, are beyond the reach of judicial competence.Footnote ⁴⁴ These are, of course, the concerns that run through the Chief Justice’s analysis in Rucho. Though his analysis is unmindful of the distinction between right and remedy – the ability to identify situations in which partisan gerrymandering has gone too far does not turn on the need for a precise formula by which to make things right – it repeatedly invokes the difficulty for judges of weighing and balancing the various considerations that go into the drawing of district boundaries.

But here, too, consideration reveals that we frequently, and often noncontroversially, ask courts to do things that fall outside the contours of the standard model. In his 1976 article The Role of the Judge in Public Law Litigation, Professor Abram Chayes observed that much federal litigation at the time concerned “the vindication of constitutional or statutory policies”Footnote ⁴⁵ and was “recognizable as a lawsuit only because it takes place in a courtroom before an official called a judge.”Footnote ⁴⁶ Prominent examples included school desegregation and prison reform litigation. He cataloged eight ways in which such litigation differs from the standard model, including with respect to its “sprawling and amorphous” party structure, the overtly active nature of the judicial role, and the forward-looking nature of its factual inquiry and remedial focus.Footnote ⁴⁷ Three years later, Professor Owen Fiss articulated a justification for the active judicial role in such cases – primarily to account for the representational difficulties and disparities that inhere in such cases due to the fact that not everyone who will be affected by the resolution of the suit will be represented effectively or even at all – and suggested that the classic model’s emphasis on courts’ dispute resolution function overlooked the crucial fact that courts “exist to give meaning to our public values.”Footnote ⁴⁸

All of this leaves us roughly where it left Shapiro. The standard model has considerable intuitive appeal. At first blush, it seems to describe our judicial system well and to provide a solid baseline against which to assess whether a given innovation is sensible. But examination reveals that the American judicial system does not consistently satisfy any of the standard model’s components, and moreover that there is ample room for disagreement about what those components mean in their particulars and how to prioritize the various often-conflicting values the system seeks to advance. Whether and to what extent judges are independent depends on the answer to questions such as “independent from what?” and “toward what end?” Whether they can truly be described as applying preexisting legal norms depends on the clarity of the rule and the facts to which the rule applies that is only sometimes present. Adversary proceedings do not always fit the bill, and even when they do, there are often reasons why it makes sense to expect judges to depart from a purely reactive orientation. In similar fashion, we can conclude that dichotomous decisions are sometimes inappropriate to the nature of the problem at hand.

The key point, again, is this: courts do not exist for their own sake, or to meet the demands of a particular concept of what they are to be. They exist to serve social functions, and Louis Sullivan’s phrase “form follows function” serves just as well to capture the reality of courts as it does the design of buildings. The nature of judicial power, and of the ways in which it is used and constrained, will vary across time and contexts.Footnote ⁴⁹ The boundaries between courts and other parts of the political system are not fixed, and the powers and responsibilities of the institutional actors within government will shift and overlap.

For these reasons, Shapiro rejects the standard model as a depiction of the necessary ingredients of courts and offers instead “a root concept of ‘courtness’” that better captures the core logic of courts and court-like institutions. At its heart is a triad: disputants who find that they cannot resolve their conflict turn to a third party, a neutral standing in an equivalent relation to both. This triad features “a basic instability, paradox, or dialectic that accounts for a large proportion of the scholarly quarrels over the nature of courts and the political difficulties that courts encounter in the real world.”Footnote ⁵⁰ That instability is this: once the third party has reached a decision, what was previously a triadic relationship breaks down and threatens to appear, to the losing party especially, as two-against-one. The neutral has taken a side. “To the loser,” Shapiro concludes, “there is no social logic in two against one. There is only the brute fact of being outnumbered.” This, in turn, has significant implications for institutional procedures and design: “A substantial portion of the total behavior of courts in all societies can be analyzed in terms of attempts to prevent the triad from breaking down into two against one.”Footnote ⁵¹ The upshot is that much (maybe most) of what matters in establishing and maintaining the real and perceived legitimacy of the judiciary is the perception of the court as a true neutral, not only in the stages that precede the moment of decision, but in those that follow as well.Footnote ⁵²

To accept the logic of the triad is not to reject the standard model completely. Indeed, considered in the abstract, the model seems to provide a good mechanism for implementing the triad. An independent judiciary applying preexisting legal norms via a process in which the disputants participate as adversaries can be recharacterized as one in which the judicial role entails the freedom to follow the law wherever it leads, and to do so after allowing the parties an opportunity to make their strongest possible cases. The judge can avoid the breakdown of the triad by pointing to the law, rather than the judge, as the author of the decision. (The dichotomous decision, by contrast, does not seem to be a necessary implication of the desire to maintain triadic integrity.) Indeed, both the majority and the dissent in Rucho implicitly accept something akin to this depiction of the standard model as a baseline, and in a manner that also recognizes the value of the triad’s logic. Both accounts provide valuable fodder for any analysis of the judicial role in the United States, and we will return to both throughout the remainder of this book.

The analysis thus far has revealed that the bedrock concepts that make up the standard model, ideas that are routinely invoked as if they have a settled definition, are in fact both contestable and contested as to their meanings and are ideals that are in practice impossible to meet. The usual materials of law cannot provide clearly predetermined answers in all cases, and the litigants will not always be well-positioned or incentivized to direct the court’s attention to all the pertinent materials. The path from law to decision – or, as it is often called, judgment – requires judgment. We want judges who have the freedom to exercise judgment in appropriate ways – whatever we might mean by that – but not to do whatever they want. Understanding the nature and role of courts and judges, and their various manifestations in our society, requires more than abstract reasoning from idealized conceptions. Such thinking can help with both understanding and setting limits. But we must also consider the incentives, capacities, and pathologies that affect individual judges on the ground as they go about their work, the various processes and other features of institutional design that work to channel judicial behavior in the desired directions, as well as larger questions about jurisprudential and intellectual trends that shape the atmosphere in which this work, and analysis of it, takes place. The remainder of the book takes up that work.