Justice and Accountability: Activist Judging in the Light of Democratic Constitutionalism and Democratic Experimentalism

I. Introduction

A familiar theme in American legal discourse pits judicial judgments about justice against democracy. Judges, some argue, should not rely on their understanding of justice when that understanding is in tension with legislative enactments. A still stronger version challenges the idea that such understanding should even guide or supplement interpretation of legislative enactments or common law authority.

The problem with values of justice according to this critique is that people disagree about them. They agree that certain values, like equality and freedom of speech, are important but only at a high level of abstraction. These general concepts do not generate uncontroversial answers to particular issues. Many powerful theories strive to link the general concepts to resolutions of particular cases. However, people disagree both about which of these theories is better, and they often disagree within a given theory about which answer to a concrete dispute the theory supports. For the critics, there is no way to legitimate judicial judgments about justice in the absence of something close to consensus.

According to the critique, it is dogmatism or arrogance for judges or the relatively educated and affluent class from which they emerge to suggest that they have privileged insight into what is good or right. They have no way to validate their premises against the modernist skepticism they themselves deploy against beliefs they do not hold. Imposing such judgments violates the fundamental moral duty of contemporary political thought of respect for people with whom one disagrees. Moreover, it seems incompatible with political accountability. Given judges’ relative immunity from political pressure, errors in their decisions cannot be readily exposed and redressed.

Democracy is one fundamental value to which the critique would have judges commit themselves. They view it as procedural rather than substantive. They seem to believe that democratic values are more widely shared than the substantive values they regard as problematical. But they also make arguments of principle that suggest that democracy is one commitment that does not depend on consensus. They value democracy because it treats people respectfully as equals. One-person/one-vote democracy with universal suffrage and majority decision respects equality strongly in principle and at least minimally adequately in practice.

We can take Antonin Scalia, John Ely, and Jeremy Waldron as familiar examples of this critique. ¹ The reach of their positions varies. Ely and Waldron are largely concerned with judicial review, and Ely concedes a broader role than Waldron for courts in reviewing legislation that impairs the democratic process. Scalia conceives a broader role than Waldron for courts in reviewing legislation that impinges on textually grounded constitutional norms, but unlike Ely’s and Waldron’s, his critique of judicial resort to informal values of justice extends to ordinary interpretation of legislation as well as judicial review. Nevertheless, there is a basic common ground among the three: the idea that legitimacy must come either from consensus or democracy and the consequent belief that judicial judgments in hard cases grounded in ideals of justice are unlikely to be legitimate. The critique is important because it captures widespread anxieties, first, about the public grounding of moral judgments, and second about the exercise of power by judges. Yet, although people often invoke its arguments against judicial decisions they reject, few embrace them consistently.

We can get insight into the limitations of the critique by contrasting its treatment of judicial activism with two other perspectives in recent legal scholarship – Democratic Constitutionalism (DC) and Democratic Experimentalism (DE). The former is exemplified by the work of Bruce Ackerman, Reva Siegel, and various collaborators; ² the latter is exemplified by the work of Charles Sabel and various collaborators. ³ These two perspectives share with the critique doubts that judicial authority can be secured by abstract normative theorizing or doctrinal heuristics. They also share a desire to square judicial independence with democracy. But DC and DE, in common or complementary ways, suggest that the critique oversimplifies its account of activist judging and present an argument for it that responds to at least some of the critics’ concerns.

In particular, DC and DE challenge the critique along three dimensions.

First, they suggest that the critique exaggerates and mischaracterizes the pre-emptive power of courts. Judges do exercise power in important ways, but the exercise is more provisional and contingent and more subject to accountability mechanisms than the critics allow.

Second, the conception of democracy that the critics oppose to activist judging is too narrow. It focuses exclusively on electoral and legislative processes and ignores civil society, in particular, social movement and stakeholder engagement. These latter processes respond to a somewhat different, but nevertheless compelling, conception of democratic equality from the one invoked by the critics. Activist judging can potentially both induce and be disciplined by such processes.

Third, the critics’ idea of consensus is too limited. They assume that legitimation requires a pre-existing consensus. Yet, judicial intervention can induce debate and deliberation that leads to consensus. Such induced consensus can legitimate retrospectively, and the prospect of it can provide a kind of provisional legitimacy.

II. Democratic Constitutionalism

Brown v. Board of Education is a problem for the critics. The correctness of Brown is the bedrock of American rights discourse. There is disagreement about its implications in many situations, but no one can expect to be taken seriously in rejecting the case’s holding that racially segregated public facilities violate constitutional equal protection. If there is anything about public law that Americans agree on, it includes this. Justice Rehnquist and other conservatives of his generation had to distance themselves strenuously from their early criticism of the ruling. ⁴ Herbert Wechsler’s negative assessment, which was once considered the height of sophistication, is virtually unintelligible today. ⁵

Moreover, the antidiscrimination understanding that has been accepted as foundational in American public law extends beyond the Brown holding. It condemns racially-based exclusion from, in addition to education, other public services, the electoral process, public and private employment, “public” accommodations even when privately owned, and state-sanctioned marriage and family relations. This non-discrimination principle extends to disadvantages based on ethnicity and religion as well. To be sure, controversy remains about some applications. There is intense debate about the legality of affirmative action and of activity with “disparate impact” (decisions based on neutral criteria that have foreseeably disproportionate effects on protected groups). But there is a large core that rests on consensus.

The Brown core is the most important, but it is not the only example of judicially-declared doctrine of substantive justice that has gained virtual consensus status. Consider, for example, the “de facto Equal Rights Amendment”. The Equal Rights Amendment (ERA) was never enacted as an Article V amendment, but something very close to the norms its proponents thought the amendment would mandate became law through other means, including judicial decision. There has been little public dissent from a core set of gender-equity norms since Robert Bork was denied confirmation of his Supreme Court nomination after criticizing the Supreme Court cases interpreting the 14th amendment equal protection guarantees to apply to gender discrimination. The core of the de facto ERA is presumptively equal access for women to education and employment, as expressed in strict scrutiny of gender classifications. Not included in the core are claims to supportive services such as childcare or abortion rights.

It is possible that we are witnessing the formation of another consensus of principle around state neutrality toward sexual identity. Fighting continues around gay marriage, but the age break-down in reported views suggests that it will not be long before consensus arrives.

The troubling feature of these examples, and especially the Brown core, for the critique is that they are matters of substantive justice. If the most agreed-upon principle in American public law – the Brown core – is substantive, that presents at the very least a big counter-example to the critics’ picture of disagreement about justice as a starting point for constitutional theory.

The critics have recognized this problem and responded by trying to characterize Brown in other ways. Most notably, Ely sought to explain Brown as a response to a procedural defect in majoritarian democracy – the role of prejudice in subverting legislators from fair consideration of the interests of minorities. But the argument was widely considered unsuccessful, and few advance it today. The basic problem is that it is hard to distinguish in procedural terms the kind of prejudice to which Brown responded from myriad other kinds that do not raise constitutional problems (for example, prejudice against burglars) without resort to substantive values. ⁶ In another effort to ground Brown without resort to substantive values, Michael McConnell tried to show that Brown could be understood in originalist terms of the sort defended by Justice Scalia, but again, the effort has not proved broadly persuasive. ⁷ Of course, Ely and McConnell were not doing this work to support Brown. Brown needed no support. They were working to shore up their non-substantive approaches by showing that they were compatible with Brown.

Brown, as well as the gender-equity core, are best understood as substantive, and they rest on a powerful base of social agreement. But of course, this agreement did not legitimate judicial action in the manner contemplated by the critics. It did not pre-exist the judicial decisions. Indeed, there was massive and sometimes violent contestation around them. The consensus emerged from this conflict. Consensus did not produce the judicial decisions; it was in substantial part produced by them.

Democratic Constitutionalism has analyzed the process by which constitutional principle becomes entrenched in this manner. The key processes are different from those portrayed by the critics. The critics contrast a pre-emptive mandate imposed by a court with one enacted by a democratically constituted legislature. In the DC account, basic constitutional norms emerge from interaction among all three branches and between government and the People, acting both as electorate and civil society.

Judicial decisions state the law in a partial fashion. They typically permit or require substantive elaboration and enforcement processes that the executive and legislative branches are best equipped to provide. These other branches have ample opportunity to advance or resist the courts’ projects. If the process is salient and sufficiently prolonged, it will become a subject of electoral debate and contention. Legislative and executive activity will feed back on the agenda of the courts. All three branches will respond to experiences of enforcement. Election returns will influence legislatures directly and courts indirectly. At the same time, social movement activism will influence the electoral process and informal public deliberative processes.

The court’s role in this picture is not as strongly pre-emptive as the critics tend to portray it. In the long run, a court cannot prevail over persistent widespread opposition. The court will be compelled to back up, even to reverse itself, or its pronouncements will become dead letter. Conversely, entrenchment requires support from all three branches and validation in elections.

Southern racists talked as if the Supreme Court had dictated abandonment of apartheid. But properly understood, at most, it merely shifted the burden of democratic contestation. They could have prevailed by capturing the three branches of government in a series of elections, as the New Dealers had done to reverse the laissez-faire Constitution of the late 19th and early 20th centuries. In order to do that, they would have to engage their fellow citizens over the issues. Moreover, the range of issues as to which the burden was shifted was ambiguous. What Brown meant beyond condemnation of de jure segregation and even the scope of the idea of de jure segregation was left open. Supporters who wanted to establish an expansive understanding of constitutional equality had ample reason to mobilize on their own initiative.

Bruce Ackerman’s account emphasizes inter-branch engagement and the role of the electorate. The Brown core developed in judicial decisions over two decades. They were complemented by high-profile executive initiatives, including executive orders and agency rule-making, as well as threats and sanctions against recalcitrant state officials and physical protection of demonstrators against lawless violence. Congress intervened with a series of statutes, including two ‘super statutes’: the Civil Rights Act of 1964 and the Voting Rights Act of 1965. The statutes created new opportunities for private enforcement and empowered the executive branch to enforce them in new ways. The core was a focus of debate in the Presidential election of 1964, and Johnson’s landslide victory was a major step in the entrenchment of the core. Consolidation was advanced following the election of 1968, when Nixon made clear that, notwithstanding his hostility to much of the liberal agenda of his predecessor, he accepted the Brown core. The two super-statutes were revised in the 1970s in ways designed, for the most part, to strengthen them.

As Ackerman’s account emphasizes the interaction of official initiative and the electoral process, Reva Siegel’s account of the de facto ERA emphasizes the role of engagement in civil society. The ERA inspired a movement for gender-equity and a counter-movement that sought to limit its ambitions. The amendment failed, in part because of its ambiguity about key issues on which there was strong division, such as childcare entitlement and abortion rights. But a core of norms prescribing equal access to employment and education took root. The Supreme Court played a role by developing a jurisprudence that treated gender as a suspect classification (a holding unwarranted by either originalist or proceduralist premises). Again, there were complementary statutes and executive initiatives. Siegel shows the contours of the core emerging in the discourse of social movement and counter-movement. ERA proponents moderated their positions when confronted by protests that their views devalued traditional family norms and implied a cumbersome expansion of the welfare state. Their concessions facilitated the emergence of the core.

The disciplinary pressures of social movements reach the courts. Some of this pressure is communicated through legislatures. When courts act pre-emptively to hold statutes invalid, they are frequently acting against outlier jurisdictions and can appeal to strong state legislative trends. Social movement pressure can also act more directly. Siegel has shown that the shape of the courts doctrine on both the de facto ERA and the still controversial Second Amendment owe much more to social movement discourse than to formal legal authority. ⁸

Siegel sees deliberation as an important mechanism of consensus formation. In her account the key feature of deliberation is the pressure to understand and respond to views opposed to one’s own. “The quest to win public confidence and to capture sites of norm articulation disciplines change agents, leading them to internalize elements of counterarguments and to other implicit forms of convergence and compromise.” ⁹

Thus, DC suggests that courts often do well to configure doctrine over hotly contested issues in ways that promote deliberative engagement. While Siegel is ambivalent about recent authority on abortion and affirmative action, she argues that the use of conditional standards in cases like Planned Parenthood v. Casey, ¹⁰ and Ricci v. DeStefano ¹¹ has desirable effects in inducing continuing discussion of the issues. ¹² In Casey, the Court backed off of some of the categorical strictures of Roe v. Wade in favor a general standard: the prohibition of “undue burden” on the choice to have an abortion. In its affirmative action cases, the court has refused to either categorically permit or categorically prohibit racially conscious decision-making designed to promote inclusion. Rather than condemning abortion regulation and affirmative action outright, they specify the legitimate purposes of such measures and then require that the measures be narrowly tailored to such purposes. The effect is to encourage both debate and experiment over the elaboration of the standards. Such contestation seems desirable in two ways. First, it holds out the possibility that some form of consensus may emerge. Second, even where they do not prevail, it gives participants a sense of connection to the processes of decision.

The Democratic Constitutionalist picture blurs the distinction between strong and “weak form” judicial review that has become salient recently. In weak form review, as exemplified in Canada or the United Kingdom, the Court can declare a statute unconstitutional, but the legislature can nullify the holding by enacting a new statute explicitly overriding it. In practice, however, legislatures in these systems seem reluctant to do this. ¹³ In the nominally pre-emptive American system, a judicial decision can have some effect over persistent, widespread popular opposition for some period of time. Roe is the prominent modern example. But this effect is limited in scope and (most likely) in time.

From the point of view of democratic legitimacy, activist judging can be considered in three categories.

First, some of the courts’ decisions purport to reinforce the democratic character of the political process. Ely’s proceduralist theory allowed for “representation-reinforcing” review. His vision of democracy was focused on elections. So the interventions he defended most vigorously were First Amendment protections of political speech and application of equal protection norms to the voting process.

Democratic Constitutionalism has a broader vision of the political process that implies a broader conception of procedural preconditions. It values free speech, not just as essential to electoral accountability, but also as safeguarding the process of popular consensus formation. Consensus formation requires opportunities for organized deliberative engagement outside as well as inside government. It also requires opportunity to protest against prevailing views. Consensus is a basis for legitimacy only when it is voluntary and reflective. Legitimacy is enhanced when positions are subjected to the “full blast” of competing views.

At the same time, DC values organizational efforts that induce and facilitate extra-electoral deliberative engagement. Thus, it should accord prominent place in the canon to the decisions that protected social movement initiative. The Supreme Court struck down bar regulations that prevented movement organizations from supporting or coordinating lawsuits brought by individuals, invalidated maintenance and champerty rules that banned solicitation of potential litigants; protected the privacy of organizational member and contributor lists; and carved out a space for public demonstrations and concerted economic pressure against trespass, disorderly conduct, and economic conspiracy rules. ¹⁴

Ely’s argument was that the critique of judicial activism should not apply to “representation-reinforcing” decisions because the critique presupposes a democratic legislature. A judicial decision that imposes a condition of democracy on the political process cannot be criticized as undemocratic. Or at least, such a decision is democratic in effect if not in origin. The argument is appealing, but one can see why critics like Scalia and Waldron are wary of it. Most claims about substantive justice could be packed into conceptions of democracy, and the concession could thus end up swallowing the critique. This is not a problem for DC, however, because it defends activist judging inspired by substantive values.

The second of the DC categories of activist intervention involves the enforcement of a dominant trend against outliers. Cases like Gideon v. Wainwright, ¹⁵ requiring appointed counsel for criminal defendants, and Griswold v. Connecticut, ¹⁶ striking down contraceptive prohibition, followed growing and widespread support for their principles in federal and state statutes and judicial decisions. The Court’s constitutional holdings thus rested on incomplete but emerging consensuses. Such decisions follow broad national deliberation in circumstances where there is ample evidence of broad national support for them.

The third and most challenging category of activism involves cases like Brown where the court takes an initiative on the basis of a substantive value without the support of a dominant trend or emerging consensus. DC defends such cases in part as deliberation-inducing. These cases do not categorically pre-empt democracy in the manner the critique implies. They do involve the exercise of power, but this power takes the form of three kinds of more modest and contingent effects. First, the court may have prestige due to features of the judicial role – political independence, disinterestedness, reason-giving – and this prestige may give persuasive force to its views. Second, the court exercises agenda control. It creates a focus of public deliberation and forces officials to address issues they would have preferred to ignore or defer. Brown forced political leaders like Presidents Eisenhower and Kennedy, who were sympathetic to racial equality but who might otherwise have been inclined to ignore or downplay the issue, to address it saliently and concretely. Third, the court can shift the burden of initiative. Opponents can no longer remain passive and enjoy the status quo; they must affirmatively mobilize to secure what they want. It is generally harder to enact a new rule than to block change of an old one in any system, and more than usually so in the American. Thus, the shifting of the burden can be highly significant.

It is not clear how much this revised picture of judicial activism would placate the critics even if they accepted its descriptive accuracy. The critique objects most strongly to categorically pre-emptive judicial decisions. It sometimes suggests that, as long as judicial decisions can be reversed by new legislation, the value of democracy is satisfied. On this view, there is room for quite a bit of activism. But the critics would be right to point out that the powers that the court exercises in even “weak form” review are substantial, and they are not democratic in the distinctive sense emphasized by the critics of affording equal opportunities for participation for all citizens. On the other hand, it would be difficult to deprive the judiciary of some measure of these powers, and doing so would not necessarily contribute to a more democratic system. Public officials have the potential to acquire influence through prestige of office in any system. It is doubtful whether there is any viable system of agenda-setting that respects the critics’ equality principle. And the power to shift the status quo is in tension with democracy only if the status quo has some democratic warrant. Often it will not. The status quo may be the product of some long-ago legislation that persists more through inertia than popular support, or it may simply be a matter of informal social custom and power relations reinforced by background laws that regulate property and general social action.

III. Democratic Experimentalism

Constitutional theory tends to pause when key principles and paradigm cases have been accepted and political mobilization has waned. But much work remains to be done. At this point, both the nature of the task and the processes of addressing it have changed. We are no longer concerned with entrenchment so much as elaboration. Disputes are less likely to involve general principles than their application to varied and incompletely understood circumstances. The institutional configuration of citizen engagement may also have changed. As the intensity of informal social movement recedes, it leaves behind newly strengthened organizations that can continue to engage in a more focused and structured manner.

Democratic Experimentalism has been especially interested in activities at this stage. Like DC, DE appeals to deliberative engagement and potential consensus as a path to legitimation. But it focuses on institutional processes that are more formal than social movement activism and less comprehensive than general elections and legislation. The key institutions might be called contextualizing regimes.

Consider what some call “second generation” race discrimination, or more generally, civil rights problems. First generation law focused on consciously and explicitly invidious discrimination or, in related areas such as the Fourth Amendment, egregiously reckless conduct. By contrast, second generation issues arise from conduct that is not invidiously motivated or egregiously reckless but is foreseeably disproportionately harmful to groups or values protected by civil rights law. Disparate impact discrimination claims are a key example.

A disparate impact claimant challenges a practice, such as a college-degree requirement for a job that foreseeably disadvantages a protected group. The plaintiff may assert that the requirement was motivated by a desire to achieve this effect and is thus merely a disguised form of invidious discrimination. The defendant responds by adducing non-discriminatory grounds that might justify the practice. In the employment context, she might say that the requirement correlates with higher productivity. From the plaintiff’s point of view, traditional doctrine is unsatisfactory because the existence of a possible legitimate ground does not mean that the decision was not influenced by discriminatory sentiment. Experimental psychology has taught that racial preconceptions are pervasive but, in the post-Brown era, often unconscious and nearly always unacknowledged. However, if doctrine permits an inference of discrimination from foreseeably disparate impact, it may put the defendant in an unfair situation. Since it is often impossible to conclusively demonstrate the productivity effects of an employment requirement, cases will often be decided by the allocation of the burden of proof. The losing party often feels that he lost because of unrealistic evidentiary burdens.

An important response to such difficulties emphasizes notions such as “less restrictive alternative” and “reasonable accommodation.” ¹⁷ Under the strong versions of these requirements, where there are alternative practices that serve the defendant’s legitimate purposes while doing less harm to protected groups or values, the defendant should adopt them. Where doctrine remains focused on intent, the reason to adopt mitigating measures is to escape an inference that the more burdensome practice was invidiously motivated. But in the more demanding versions, intent is no longer the touchstone. The duty to assess impact of practices on protected groups and to search for less burdensome alternatives becomes a core element of the duty of equal protection.

Second-generation civil rights issues are part of a broader class of issues that confront the modern state characterized by variation and fluidity. Democratic experimentalism views these conditions as calling for responses customized to local contexts or revised more or less continuously as new understanding accrues. In these situations, neither ends nor means may be fully clear prior to intervention. Thus, intervention must be provisional, and it must take the form in part of investigation. In regimes of this kind, a central institution authorizes and oversees local experimentation within uniform parameters. Local initiatives are transparent and their effects are routinely assessed. The center holds the local units accountable to the parameters and aggregates information about local efforts. Local initiatives are developed through stakeholder deliberations. So initially, general principles are elaborated locally in varied ways. But there remains the possibility that as some local initiatives gain recognition, their premises will come to be widely shared and eventually may be incorporated into the uniform framework.

Local investigation in experimentalist regimes typically involves stakeholder deliberation. Such deliberation has an instrumental rationale. Stakeholders have information that officials cannot easily gather without engaging them, and implementation may require cooperation by stakeholders that engagement may induce. But there is also a democratic rationale. Stakeholder deliberation is a form of self-government, one that focuses more concretely than general elections or legislatures on problems that affect the participating citizens.

Civil rights issues in juvenile detention provide an example. Pretrial detention is constitutionally permissible only on the basis of demonstrable risk that the defendant will fail to appear for trial or will re-offend if left at large. In addition, detention decisions must be race-neutral. Yet, many studies conclude that decisions do not correlate with indicators of risk and that they are racially biased. The studies are usually controversial, however, and it is typically open to defendants to argue that the data is inadequate or improperly measured or that relevant variables have been omitted. Moreover, even if inappropriate detention could be established through aggregate data, the consequent remedy might be controversial. It would not be enough to order the defendants to detain only where the risks justified it or to stop discriminating. It would be necessary to tell them how to do so, and there would likely be controversy at that stage.

In recent years, federal legislation and foundation initiatives have produced a distinctive approach to the problem. Courts have not participated in this development in the pre-emptive manner that the critics worry about. But the development exemplifies a conception of right that seems to moot the critics’ objections to judicial intervention. As we will see, the courts have intervened in activist fashion in other areas on the basis of this conception.

With respect to racial disparities, local juvenile justice agencies are obliged under the emergent regime to measure the racial incidence of decisions that potentially lead to detention, and when they find disparities, to develop plans for mitigating them. The Department of Justice and the Annie E. Casey Foundation provide technical assistance and prescribe metrics for measuring disparities. These efforts, known as the Juvenile Detention Alternatives Initiative (JDAI), have generated a network of state and local agencies that involves about half of the relevant agencies in the country. The agencies pool information and engage in various informal modes of peer review.

JDAI has transformed the process of juvenile detention in the past quarter century. Detention has fallen dramatically during this period. While aggregate racial disparities have persisted, they have fallen in some localities, and since minorities are over-represented in detention, the aggregate reduction has benefited them disproportionately. ¹⁸

Two elements of this transformation are especially interesting. One is the replacement of informal probation officer judgments about pretrial detention with empirically validated risk assessment instruments. The instruments dictate a decision on the basis of scores determined by objective factors such as prior offenses or the availability of an adult to take responsibility for the child. (The numerical scores can be over-ridden where the official believes the instrument does not adequately account for some aspect of the situation but only with supervisor approval and subsequent review.) Agencies are supposed to validate their instruments on the basis of data from their own and others’ experience. Each criterion is tested for its power to predict the consequences that justify detention – re-offense and failure to appear in court.

The other key JDAI practice has to do with the attempt to diagnose the sources of racial disparity and find ways to mitigate them. A mundane but potentially important example concerns notification and transportation initiatives. Some jurisdictions have achieved major reductions in failures to appear by routinely telephoning parents prior to a scheduled appearance, reminding them of the date, and telling them exactly where the child should appear and what to expect. Sometimes, notification is coupled with an offer of transportation. In Santa Cruz County, California, Latinos are concentrated around Watsonville in the southern part of the county, while the courthouse is located in a sparsely developed area of the north poorly served by public transportation. When the probation department inaugurated a bus service from Watsonville to the courthouse, failures to appear dropped significantly.

More complex mitigation strategies involve the development of less restrictive forms of supervision than incarceration. Electronic monitoring is a technology-based example. Others, such as after-school reporting centers, mentoring or coaching programs, and substance abuse treatment, involve social services. The reasonableness of a decision to detain will depend substantially on the availability of such alternatives. A community that fails to develop them will have more occasions to incarcerate. Individual decisions may seem reasonable viewed in isolation, but the community’s failure to develop its institutions may seem unreasonable when viewed in terms of its aggregate effects on incarceration. (Thus, do negative rights to liberty shade into positive rights to government services.)

JDAI prescribes an evidence-based approach to reform. Agencies must specify metrics for assessing the success of their initiatives and periodically re-assess them in the light of experience. They must establish inter-agency and interdisciplinary collaborations. Typically, criminal justice agencies form collaborations with social service agencies and private service providers. And continuous consultation with community leaders and organizations over both detention criteria and alternatives is standard.

Viewed as a process of rights elaboration, JDAI has two stages. In the first stage, responsible officials have a duty to assess the effects of their practices on relevant civil rights values and to scan for less harmful alternative practices. The second stage occurs once experimentation produces consensus around specific interventions. At this point, officials have a duty to adopt proven interventions unless they can articulate plausible reasons why they would not be effective in their circumstances.

JDAI illustrates a process of consensus formation over constitutional elaboration. The mechanism of the process is slightly different from the one emphasized in DC. DC most often portrays movement toward agreement arising from the pressure to provisionally identify with an antagonist’s perspective in order to respond to her concerns in the context of a commitment to a common constitution. In DE, progress is more often portrayed as coming from pressure to dissolve abstract assertions of value or position into more concrete propositions, the discovery of common ground among these more concrete propositions, and the subjection of some propositions to empirical testing. The course by which vague general judgments about dangerousness were translated into concrete propositions and then tested systematically is an example.

Litigation has not played a strong direct role in the JDAI process. Courts have participated as partners in the administrative coalitions that have designed the reforms, and they have implemented the reforms in individual case decisions. They have not, however, mandated them as a matter of equal protection doctrine. The Department of Justice has primary enforcement responsibility and it has favored informal pressure over litigation. ¹⁹

We do, however, find courts imposing analogous remedies in institutional reform cases in such areas as schools, policing, prisons, and housing. The process in these cases is another example of the complex relation of substance and process, judicial judgment and extra-judicial deliberation. ²⁰

Typically, the plaintiffs will prove a series of individual instances of official conduct they allege violate civil rights. The child welfare agency maintains children in foster placements where they are under-nourished. Police officers use deadly force to stop fleeing unarmed people suspected of nonviolent crimes. A prison places prisoners in long-term solitary confinement as punishment for disrespect to guards. The court makes a substantive judgment in each of these instances as to whether the conduct is permissible under statutes or the Constitution. The constitutional and statutory norms overlap and both tend to be stated in terms of very general substantive values like equality, due process, or cruel and unusual punishment. These are clearly the kinds of judgments about which the critics are uneasy.

But these judgments are rarely the controversial part of the cases. Most often, when the court finds liability in these individual cases, its judgments are supported by broadly shared popular sentiment or by professional standards that the defendants often concede are applicable. Controversy arises at one or all of the next stages. In the immediate next stage, the court determines whether these incidents, in combination with evidence of general administrative practice and structure, indicate that violations are systemic and warrant structural relief. In the next stage, the court enters a remedial order and then finally, the court determines when there has been sufficient compliance with the order to terminate the intervention.

Once a systemic violation is found, the case starts to play out in a manner that strongly resembles the JDAI process. The court orders the defendants to negotiate a remedial regime with the plaintiffs. With surprising frequency, defendants prove willing to do so without much coercion, but where necessary, the court will push with a threat of some harsh default remedy such as contempt sanctions or the closing of a program or facility. The parties then typically produce a decree that the court confirms.

When the decrees contain substantive provisions, such as limits on cell size or requirements that children in foster care get medical examinations, the provisions are often based on standards from professional organizations. Then the decrees will provide for extensive monitoring and reporting. And further, they provide for re-assessment of practices in the light of the experience shown in the monitoring.

Two sorts of controversy are common. The defendants and their supporters often object to structural relief on the ground that it would interfere with democratic processes. And they often object to particular remedial provisions on the ground that they are not entailed by the substantive violations the court has found. For example, they might object to a requirement that police carry body cameras on the ground that there are many ways to monitor misconduct and the court’s findings of substantive violations do not require this particular remedy. These objections resemble the points that the critique makes against pre-emptive judicial review. While structural injunctions limit executive rather than legislative power, they too involve the constraint of elected officials by democratically unaccountable judges.

The way rights get elaborated in JDAI and institutional reform litigation has resemblances to DC, and the responses in DC to concerns about democracy overlap those in DE.

To begin with, as in DC, the Experimentalist interventions involve a vision of democracy reinforcement that goes beyond Ely’s electoral view. As DC emphasizes the importance of social movements, DE emphasizes the importance of stakeholder participation. Statutes like Juvenile Justice and Delinquency Prevention Act, which underpins the JDAI, encourage local agencies to engage stakeholders in devising solutions to specified problems. Structural injunctions pressure defendants to engage with the plaintiffs and other stakeholders and frequently mandate participatory processes explicitly.

In litigation, the rationale for the court’s mandate is that the substantive violations found in the liability determination demonstrate a failure to take account of interests of stakeholder constituencies and consequently a defect in the democratic process. Substantive violations are treated as symptoms of democratic failings. ²¹

In addition, as in DC, intervention is often supported by an emerging consensus and has the effect of codifying or consolidating it. Many institutional reform decrees force outlier institutions to adopt practices widely viewed elsewhere as standard. For example, some of the early prison cases dealt with prisons that still employed the system by which guards delegated disciplinary authority to inmate “trusties”. The system had been abandoned throughout most of the country and was widely condemned within the corrections profession. A series of decrees prohibiting it in recalcitrant institutions basically codified this view. ²² In juvenile justice, the practice of validated risk assessment instruments and detention alternatives such as electronic monitoring seem to be emerging as consensus norms. Were a plaintiff to show repeated instances of substantive violations by an agency that lacked such practices, a court would have a substantial basis on which to order them.

Next, structural intervention has deliberation-inducing tendencies analogous to those portrayed in DC. At their most pre-emptive, the courts impose substantive rulings directly. They may rule that a certain level of prison crowding or the punitive imposition of solitary confinement violates the Eighth Amendment, for example. Here the court puts the weight of its authority most strongly behind propositions of substantive justice. But again, such propositions are subject to political contestation. The court’s rulings can become a subject of debate in civil society and the electoral process, and the other branches can respond either supportively or in opposition. Prison decrees, in particular, have engendered extensively controversy and legislative response. In consequence, the courts have increased the burdens on plaintiffs to establish liability and have tailored decrees more narrowly. But a set of core principles and interventions seems to have been established. ²³

However, a striking characteristic of structural interventions is their tendency to refrain from specifying substantive duties categorically and instead to mandate directly that officials engage with stakeholders. Defendants are induced to negotiate with plaintiffs both over the terms of the decree and throughout its life over issues of compliance. Judicial decrees typically mandate various types of deliberative engagement directly. For example, the recent federal decree against the New York City police department requires the department to hold a series of community forums to explain the decree and seek suggestions for its implementations. Police decrees typically contain provisions prescribing the operation of civilian complaint bodies and, sometimes, of more proactive citizen oversight processes. The Department of Justice, which has negotiated consent decrees with police departments throughout the country, has played a role in facilitating the exchange of information and developing standards across jurisdictions.

Judicial practice in this regard takes a form similar to what Congress mandated by statute in the juvenile justice area. The Juvenile Justice and Delinquency Prevention Act encourages deliberation in various ways. Agencies have incentives to exchange information to demonstrate the seriousness of their efforts to the Department of Justice. State and local agencies have formed the Coalition for Juvenile Justice to facilitate collaborative support, and the Casey Foundation brings agencies together routinely. The Department of Justice pressures laggards to get peer assistance from other agencies. Another dimension of deliberation involves agencies and local civic leaders and non-governmental organizations (NGOs). The design of detention alternatives typically benefits from consultation with stakeholders, and some programs are most appropriately administered by local NGOs.

The type of coercion involved here is predominantly agenda-setting and decision-forcing. Officials are forced to address issues to an extent greater than they would prefer. They are required to act more explicitly and transparently than they would prefer. And they are induced to address stakeholders they would prefer to ignore. When these interventions come from courts, they are not democratic in the sense that the critics mean, and when they are mandated by Congress, they override local democracy in a way that many object to on grounds analogous to those of the critique. ²⁴ But if they are not democratic in origin, they are democracy-reinforcing in effect. In making practice more explicit and transparent, they facilitate oversight in the legislative and electoral process. At the same time, they introduce a less episodic and more concentrated type of accountability to stakeholders.

Such interventions could be defended on the ground that they typically are designed to protect people, such as minorities, prisoners, mental health patients, or children in dysfunctional families, who seem especially vulnerable in majoritarian political processes. Note, however, that judicial practice in institutional reform cases is responsive to concerns that democracy-reinforcement arguments tend to underestimate the capacities of electoral institutions or to too readily translate substantive claims into procedural defects; for judicial intervention only occurs when its proponents have demonstrated a systemic failure on the part of the defendant to fulfill its responsibilities.

Finally, structural intervention has a dimension analogous to the shifting of the burden of initiative in judicial review. With conventional judicial review, the court’s invalidation of a statute forces proponents to resort to the political process to seek to re-establish their position. This effect may also occur with structural decrees. However, there is also a shifting of burdens that occurs within the lawsuit. We have noted that the court tries to avoid imposing relief directly and instead induces the parties to negotiate. This is because the parties have superior information and expertise. The court’s goal is to induce them to deal with each other productively and in particular, to get the defendant to engage with the plaintiffs. Intervention is thus substantially indirect. It can involve a penalty for failure to undertake open-ended deliberative effort or a reward for doing so. In JDAI, the federal government threatens to withdraw funding for other criminal justice efforts when efforts to mitigate racial disparities are inadequate and both the government and the Casey Foundation provide grants for such efforts. In structural litigation, the court threatens a penalty that neither party desires, such as closing a facility or holding officers in contempt, in order to induce the parties to negotiate a better one.

IV. Activism and Accountability

The argument of the critics is that judicial interventions based on controversial judgments about substantive justice are illegitimate because they are incompatible with democracy. DC and DE challenge each of the three key premises of the argument – intervention, democracy, and legitimacy.