abstract. In How to Save the Supreme Court, we identified the legitimacy challenge facing the Court, traced it to a set of structural flaws, and proposed novel reforms. Little more than a year later, the conversation around Supreme Court reform has only grown louder and more urgent. In this Essay, we continue that conversation by engaging with critics of our approach. The current crisis of the Supreme Court is, we argue, inextricable from the question of the Supreme Court’s proper role in our democracy. For those interested in reform, there are three distinct strategies for ensuring the Supreme Court maintains its proper role relative to democracy: internal restraints, external constraints, and structural reforms. We argue that internal restraints and external constraints both suffer from serious drawbacks as strategies for restraining the Court. Structural reforms remain the most promising option for reforming—and saving—the Supreme Court.

Introduction

Fundamental reforms to the Supreme Court were once the exclusive domain of scholars of constitutional design and theory. But in recent months, they have become a major part of policy debates as many point to a legitimacy crisis engulfing the Court. Underscoring the shift, both Joe Biden and Kamala Harris, as candidates for president and vice president in 2020, would not say whether Democrats would expand the size of the Supreme Court.1 Then-candidate Biden announced instead that, if elected, he would create a bipartisan commission that would provide “recommendations as to how to reform the court system.”2 Such developments would have been unimaginable just a few years ago, as “packing” the Supreme Court was long seen as a third rail in American politics.3 Clearly, the Overton Window4 has shifted—dramatically.

To be sure, major reform remains unlikely in the near term. Although the Democrats now control both the Presidency and Congress, the Senate is closely divided, likely dashing liberals’ hopes for Court expansion and other bold initiatives.5 Nonetheless, the dramatic change in the plausibility of Supreme Court reform is striking. What changed? The most immediate answer turns on politics. Republicans refused in 2016 to give a hearing to President Obama’s nominee, Judge Merrick Garland, on the grounds that it was an election year. Democrats were outraged at this break from norms and were apoplectic over Republicans’ bare-knuckle tactics in confirming three of President Trump’s appointees to the Court—and, in particular, over their hypocrisy in confirming then-Judge Amy Coney Barrett mere days before the presidential election. Facing a solid 6-3 conservative majority on the Court—and thus the possibility of their legislative, regulatory, and policy preferences being blocked for a generation—progressives and Democrats increasingly question the legitimacy of the Court and call for its reform.6 Republicans, for their part, say Democrats are the ones threatening the Court’s legitimacy by calling for Court expansion and other reforms simply because they are unhappy that Republicans have the advantage on the Court.7

Whatever one makes of this immediate crisis, we think it is best understood as the current manifestation of a deeper problem. In 2019, we published How to Save the Supreme Court8 in this journal, identifying the legitimacy challenge facing the Court and tracing it to a set of structural flaws that are putting increasing strain on the Court and its role in our democracy. The problem, we argued, lies in the interaction between growing political polarization in the parties and constitutional practices and traditions that are hard to justify. Lifetime appointments raise the stakes of individual Court nominations and turn essentially random occurrences like the death of a Justice into watershed events that can reshape constitutional law for a generation. Confirmation battles have, as a result, become death matches, with each side seeking to politicize and bury the other’s picks. Moreover, the Court faces a growing democracy deficit: the Electoral College and the role of the Senate in confirming would-be Justices mean that a majority of the Justices on the Court were nominated by presidents who initially lost the popular vote9 and four were confirmed by a narrow Senate majority that represents a minority of the U.S. population.10 In our earlier piece, we outlined two different ways to redesign the Supreme Court—what we called the Balanced Bench and the Lottery Court.11 Our aim was to offer structural proposals that would address what we identified as the key legitimacy problems with the Court while being practically capable of implementation via ordinary statute.

In the year-plus since our piece went to press, the “looming threat” to the Court’s legitimacy that we described has grown ever more apparent.12 And with it has grown interest in structural change to the Supreme Court.13 At the same time, commentators on both the left and the right have criticized our proposals. Some say we err by politicizing the Court too much;14 others accuse us of trying to depoliticize a fundamentally political institution.15 Some think that our changes are radical and unnecessary; others say that they do not go far enough to disempower or transform the Court.16

In this Essay, we continue the conversation about Supreme Court reform by engaging with critics. Our goal, however, is not to defend our specific proposals, which we are content to let stand on their own merits. Nor will we seek to persuade readers of our diagnosis of the particulars of the legitimacy crisis facing the Court. Although we think that crisis has only worsened since we published our first piece, one’s assessment of the situation may be difficult to disentangle from how one feels about the current composition of the Court.

Instead, we will start by trying to identify common ground with our critics in order to justify our general approach to structural reform of the Supreme Court. The current crisis of the Supreme Court is, we believe, inextricably linked to what is perhaps the central question in American constitutional law and theory: what is the role of the Supreme Court in a democracy? For generations, this question has occupied constitutional scholars, who have debated why and when it is legitimate for unelected judges to overturn the will of majorities. Much of constitutional theory has amounted to an effort to find a principled way to solve this “countermajoritarian difficulty” and reconcile judicial review and democracy.17

Today, there are almost as many constitutional theories as there are constitutional scholars. And there are many different attitudes among theorists toward judicial review. Some seek to expand judicial review beyond its present domain,18 while others would abolish it entirely,19 with many other views lying in between those extremes. What unites most constitutional theories today, however, is at least some lip service to the primacy of democracy. While a few theorists disclaim any particular interest in the will of the people, most try to anchor their theories in some account of how the Court’s role proceeds from American democracy. Put slightly differently, most scholars thinking about the role of the Court recognize that the Court’s role must ultimately be constrained by, and consistent with, democracy—even if they differ wildly in their prescriptions for how to achieve that result.

Two first-rate scholarly responses to our proposals illustrate this point. First, from the right, Stephen Sachs argues that we err by seeing the Supreme Court as powerful enough to do “just anything,” which makes it “too powerful a superweapon to leave lying around in a democracy.”20 From the left, Ryan Doerfler and Samuel Moyn argue that our proposals fail to recognize that American democracy is “beset by deep ills for which Supreme Court power is no part of a cure.”21 Both responses recognize the danger that the Supreme Court can have for a democracy, and both offer significant changes to existing practices that they claim would disempower the Court and thereby remove the threat it poses to democratic self-government.

Yet the nature of their proposals is wildly different—from ours and from each other’s. Sachs argues that instead of restructuring the Court, the way to ensure the Court is legitimate and fulfills its appropriate role in democracy is for members of the Court to adhere to a particular form of jurisprudence: “Limited government, federalism, originalism, and so on.”22 Doerfler and Moyn, by contrast, argue for reform, though a different kind than we proposed. In their view, rather than trying to change the Court’s membership, progressives should pursue “disempowering” reforms, such as jurisdiction-stripping and supermajority voting rules, that would limit or completely restrict the Court’s ability to interfere with democratically authorized legislation.23

We share these critics’ desire to put the Supreme Court in its proper place, ultimately subservient to the democratic process. Where we part ways, however, is in our assessment of the best way to accomplish that goal. In this Essay, we explain why. Our starting point is James Bradley Thayer’s 1893 article The Origin and Scope of the American Doctrine of Constitutional Law.24 We argue that we and our interlocutors can all be seen as descendants of Thayer in seeking a way to ensure that courts, and the Supreme Court in particular, do not improperly usurp democratic authority. We identify different strategies for achieving Thayerian goals and argue that our critics err by focusing too singlemindedly on their favored strategies to the detriment of more promising alternatives.

Thayer argued that the appropriate role for federal courts was deference to the political branches, striking down federal statutes as unconstitutional only when the court determines that the legislature has clearly erred in its own constitutional judgment.25 That is, his approach was to ask judges to understand their proper role and to exercise restraint. We call this classically Thayerian approach internal restraint because it focuses on how judges behave. Few modern constitutional theorists endorse Thayer’s clear-error rule. Nonetheless, we think that most leading interpretive methods fall into this broad category because they aim to ensure judicial restraint or because they claim to offer a methodology that provides clear rules for when courts should—and should not—strike down legislation. As we explain in Part I, we see Sachs as one contemporary practitioner within this broader paradigm. We criticize his approach, and the internal-constraint approach more generally, as unlikely to succeed for a variety of practical reasons, such as indeterminacy and the lack of meaningful enforcement mechanisms.

Instead, we argue that neo-Thayerians—those who believe that federal courts should be restrained in a democracy—should instead focus on two other categories of deference-forcing strategies: external constraints and structural reforms. External constraints do not alter the Court’s structure but instead seek to cabin, prohibit, discipline, or narrow the Court’s exercise of the power of judicial review. Structural reforms, in contrast, seek to restrain the actions of the Court by changing its very design. Within these broad categories are many possibilities, ranging from slight limits on Court power to more severe restrictions. Reams of paper have been spent on debates about internal restraint on the Court, but there has not been the same level of interest in external constraints and structural reforms. This is surprising because the Madisonian approach to the Constitution sought to restrain power through external and design mechanisms, rather than assuming that officials would follow the correct theory of their own proper role.26 The Madisonian approach has its downsides, given that it rests on overly simplistic assumptions about the motivations of government actors, who are more likely to advance the interests of the political parties to which they belong rather than the interests of the branch of government that they inhabit.27 But for those with neo-Thayerian instincts, it may prove more helpful going forward than interminable and intractable debates on internal restraints.

As we explain in Part II, Doerfler and Moyn’s proposals are examples of external constraints on the Court. Such reforms can, we think, play an important role in keeping the Court in its proper lane in a democracy. We argue, however, that Doerfler and Moyn create a false dichotomy between “disempowering” and “personnel” reforms—one that obscures the critical fact that there are both direct and indirect ways to disempower the Court. Personnel reforms can themselves be disempowering. More substantively, we note that their assessment of the benefits of the reforms they favor—in particular, jurisdiction-stripping—is far too rosy. The practical details of how such a measure would work are essential to achieving their own goal of implementing progressive policies. But Doerfler and Moyn do not work out how progressive policies would be stable if their aggressive proposals to disempower the Court were adopted.

Instead, we argue in Part III that those who seek to restrain the Court should look hard at structural reforms. Such reforms are, in our view, in the best tradition of American constitutional design and are most likely to provide solutions to the problems that the Supreme Court currently faces with the least practical difficulties. Our proposals are examples of structural reforms, but they are far from the only ones. Some form of structural redesign must be central to the conversation about how to reform the Supreme Court—and, indeed, how to save it.

I. internal restraints

We begin by considering the original Thayerian strategy for keeping courts in their proper role in a democracy: what we call internal restraints. And that means starting—of course—with Thayer himself. It is debatable how much Thayer deserves credit for the idea of judicial restraint.28 What’s unquestionable, though, is that he was the most influential academic proponent of the idea.29 Thayer’s famous article offered three justifications for why courts should exercise extreme restraint in declaring federal laws unconstitutional. First, he believed that the coordinate branches of government were themselves charged with making their own judgments as to the constitutionality of legislation.30 Second, he thought “the [C]onstitution often admits of different interpretations.”31 There is a “range of choice and judgment,” and “much which will seem unconstitutional to one man, or body of men, may reasonably not seem so to another.”32 This inherent uncertainty cut in favor of deference to Congress. Finally, he worried that if the judiciary became the sole expositor of constitutionality, Congress’s capacity to deliberate on the constitutionality and wisdom of legislation would
wither.33 For these reasons, he urged courts to overturn legislation only when “those who have the right to make laws have not merely made a mistake, but have made a very clear one,—so clear that it is not open to rational question.”34

Thayer’s ideas continue to influence legal scholarship more than a century after his article was published. But while his ideas may have once informed the practices of actual judges, his approach seems decidedly out of favor today in the judiciary. Supreme Court Justices appointed by both Republicans and Democrats seem more than willing to declare federal statutes unconstitutional, even if they do not agree about which statutes should be struck down.35 No Supreme Court Justice in recent decades could be plausibly described as one of Thayer’s disciples.

Why has the Thayerian project failed? Richard Posner has argued that the traditional Thayerian approach collapsed with the rise of constitutional theory.36 Constitutional scholars in the mid-to-late twentieth century increasingly offered interpretive methodologies—like originalism, textualism, and political-process theory—that claimed to provide an objectively correct answer to constitutional questions. This put Thayerian restraint under two attacks. It undermined Thayer altogether by denying the uncertainty inherent in many constitutional questions. But even if one accepted a Thayerian approach, constitutional theory expanded the category of “clear error,” thereby narrowing the window for deference. With the right method, there would be little that “admits of different interpretations.” Either way, the Court would not need to defer as much to legislatures.

Yet from a different vantage point, most modern constitutional theories can also be seen as descendants of Thayer. That is because most theories purport to offer some account of why and when judicial review is ultimately consistent with democracy. John Hart Ely’s process theory, for example, contends that under some circumstances courts can reinforce democracy by striking down legislation.37 Leading conservative jurists and legal scholars contend that originalism ensures judicial restraint vis-à-vis the will of democracy.38 Justice Scalia, for example, famously argued that “[o]riginalism is the only approach to text that is compatible with democracy.”39

With these observations to set the stage, we now are prepared to consider Sachs’s response to How to Save the Supreme Court. Sachs offers a number of thoughtful responses to the details of our proposals, which we will not quibble with here. Instead, we will start with a point of common ground: Sachs fears an all-powerful Court that can be used as a partisan weapon, just as we do. As he puts it, “[a] Court that can do just anything is too powerful a superweapon to leave lying around in a democracy; sooner or later, someone is bound to pick it up.”40 In our view, it is precisely the Court’s ability to do so much that makes it ripe for partisan capture.

Where we part ways with Sachs, however, is on how to solve the problem. He argues that if our goal is “to restore legitimacy, legal conservatives might have a few ideas to offer”41:

Given the depth of our country’s polarization, maybe we should require less by way of social agreement, relying somewhat more on private ordering and reducing the number of questions that the political process needs to answer. Maybe we should reduce the scope of that process, encouraging working agreements by different parts of the country when consensus is lacking in the whole. And maybe, to reduce the threat of the Supreme-Court-as-superweapon—capable of vaporizing any target that shows up in the Justices’ gunsights—we should precommit to limiting the Court’s freedom of action, binding it to some discrete set of preexisting rules until there is a very broad consensus for changing them. (We could even write those rules down on a piece of paper, to be kept in the National Archives—and change them only by agreement of, say, two-thirds of each House of Congress, and some three-fourths or so of the states.)42

Here, Sachs does not invoke Thayer, and he is certainly not endorsing anything like Thayer’s clear-error rule as an interpretive guide for courts. Nonetheless, we think it fair to cast him among Thayer’s heirs. He, like Thayer, seeks to keep the Court in its proper role in our democracy. Specifically, he proposes to do so by relying on “[l]imited government, federalism, originalism, and so on.”43 Originalism is an excellent example of what we have described as an internal restraint. That is, it requires judges to adopt a certain posture when interpreting the laws; faithful adherence to this interpretive methodology will (the theory goes) ensure that the Court does not usurp power that should belong to the people themselves.

While we admire Sachs’s scholarship on originalism, we think that his response to Supreme Court reform ultimately reveals why the internal-restraints approach is unlikely to succeed—indeed, why we think it is doomed to failure. Consider the specifics of Sachs’s proposed solution. He says the key to disarming the Court is to ensure that it follows a “discrete set of preexisting rules” written down on “a piece of paper . . . kept in the National Archives.”44 That is, the key is for the Court to simply follow and strictly enforce the constitutional text (and, presumably, its original meaning). Some version of this command has been the marketing slogan for textualists and originalists for quite some time. Consider, for example, then-Judge Barrett’s opening statement at her Supreme Court confirmation hearing, in which she explained that she learned from (self-described originalist) Justice Scalia that “[a] judge must apply the law as written, not as the judge wishes it were.”45

As a strategy for ensuring democratic restraint, these theories suffer from many objections that have long been discussed among constitutional scholars. First of all, even an approach that could consistently and correctly identify the objective meaning of every constitutional provision would be subject to the criticism that it is undemocratic to allow choices by long-dead framers and ratifiers to overturn the will of present-day majorities.46 But even if one can find some theory to resolve this dead-hand problem, the other difficulty is that neither the text nor the original meaning of the Constitution—the document stored at the National Archives—uncontroversially resolve the disputes that most divide our polity and that the Supreme Court has taken upon itself to decide.

The legal conservative movement uses the slogan “apply[ing] the law as written” to justify overturning cases like Roe v. Wade47 and its progeny. But Justices part of that same movement have brought us no shortage of decisions like Shelby County v. Holder, in which the Court’s conservative majority overturned the preclearance formula in Section 5 of the Voting Rights Act based on an atextual “principle of equal sovereignty.”48 And Shelby County is only one example of a case in which conservative Justices aggressively interfered with democracy based on less-than-clear constitutional text and history. Such cases have struck down democratically enacted legislation concerning voting rights,49 campaign-finance reform,50 affirmative action,51 and gun control,52 among other examples. In these cases, neither constitutional text nor history unambiguously compelled the result that the Court reached.

One possible response, of course, is that the Supreme Court and even some scholarly commentators have failed to be sufficiently faithful to text and original meaning. That is, perhaps the correct approach to constitutional interpretation would lead to more restraint. But that is part of the problem. Even limited to using the tools of textualism and originalism, American lawyers and judges can develop arguments for many outcomes that have at least a veneer of legal plausibility.53 And, as we see over and over, Justices’ views about the plausibility of those arguments seems closely correlated with their ideological attitudes toward the political and policy consequences of any particular decision. Moreover, as theorists spend time working out the details of interpretive theories like originalism, the theories themselves end up getting foggier—offering a less-sharp edge with which to slice legal problems. As Jeremy Kessler and David Pozen have argued, prescriptive legal theories have a tendency to “work themselves impure” as they mature, acquire more adherents, and confront and respond to criticisms and flaws.54

These problems are not unique to constitutional law. Take an example from statutory interpretation, where textualism is increasingly the dominant approach. On its face, textualism offers very limited resources to determine the meaning of statutes, so it might seem to offer an even greater likelihood of determining unambiguously “correct” answers to legal questions. And yet, in prominent Supreme Court cases, such as Bostock v. Clayton County,55 dueling opinions come to opposite conclusions—while both simultaneously claiming the banner of textualism. Tara Grove thus observes that the question in Bostock is really “which textualism” will apply.56

Even if legal elites could agree on a single interpretive modality (which they obviously cannot, even after many decades of debate and reams of writing on interpretation), within any particular methodology, there are innumerable variants or flavors. Not all textualists are originalists, and vice versa. And Sachs’s originalism is not Stephen Calabresi’s or Larry Solum’s or Jack Balkin’s or Akhil Amar’s—or for that matter, Hugo Black’s.57

It gets worse. Because even if the legal profession agreed to adopt a single strain of a single interpretive methodology, interpreters are consistently inconsistent. Judges who claim to adhere to a certain methodology frequently fail to do so. Internal restraints, it turns out, are not very constraining. Justice Scalia himself famously admitted that he was a “faint-hearted” originalist precisely because he was not always willing to decide cases in the manner that originalism required.58

This, of course, raises the bigger question of whether politicians or judges much care about the theories themselves. Think about it this way, would Republicans enthusiastically support Jack Balkin for a position on the Supreme Court because he is an originalist? We doubt it. The reason, of course, is that Balkin is a liberal originalist, not a conservative one. Theories like originalism or textualism are much bandied about, but political actors seem, quite simply, to prefer those who support their policy preferences.

Returning to the Supreme Court, there is another problem, one that we think Sachs does not take seriously enough. Does a commitment to the variant of “[l]imited government, federalism, originalism, and so on” that Sachs endorses require the left to simply live with decisions striking down democratically enacted legislation abhorred by conservatives? Such decisions deprive political actors of the ability to resolve difficult and sensitive matters in the name of conservative legal values, and many do so in a way that advances the electoral and policy interests of conservatives and the Republican Party. If Sachs’s proposed solution would require accepting decisions like this, his approach would entail not mutual disarmament but rather unilateral surrender by progressives as the Court advances conservative policy preferences under an originalist banner.

Indeed, Sachs himself mentions “[l]imited government, federalism, [and] originalism,” in a manner common to the conservative political and ideological project, but without grappling with the fierce conflict between these ideas. What of the tension between originalism and the transformative impact on American federalism that occurred with the Reconstruction Amendments?59 What of the conflict between originalism and the history of the expansive powers of the federal government?60 Indeed, we might even think of the political project of “[l]imited government, federalism, [and] originalism” as itself a type of Supreme Court reform—and one that has been particularly successful, given the multigenerational effort by Republican politicians as well as organizations like the Federalist Society to reshape the judiciary. Few should be surprised if progressives thus decline to go along.

And while Sachs dismisses our reform proposals as “radical,” we think that label is quite apt for the solution he proposes.61 However radical our interpretation of, say, the phrase “one Supreme Court” in Article III, that seems to us trivial when compared with the radical changes to American law and American life that would be required if our country (as it may, without Supreme Court reform) goes all-in on the approach Sachs urges.

Such a commitment might well lead the Court to invalidate the Affordable Care Act.62 It could lead the Court to declare all federal administrative agencies unconstitutional.63 It might even require the Court to hold that paper money is unconstitutional.64 Sachs asks us to consider the compromises that those who drafted and amended the Constitution made, as they “lived through civil war, economic crisis, and profound moral disagreement (over human slavery, among other topics).”65 Yet originalism seems to demand that we pay no heed to other judgments and compromises by our political predecessors that were no less important.66

Ultimately, though, our problem is not with any particular variety of interpretive theory. It is with the idea that the complete solution to the problem of a Supreme Court in a democracy—an institution that has grown too powerful, and that is resolving too many questions that should be left to political actors—lies in any kind of internal restraints imposed by judges themselves. Whether that restraint is classical Thayerianism, originalism, textualism, political-process theory, or some other interpretive methodology, history and common sense give us no reason to believe that relying on judges to police themselves will provide the proper checks. Moreover, in an age of extreme polarization, motivated reasoning, and biased perceptions and interpretation, it is hard for us to see how leaving restraint to judges alone will result in more than the rhetoric of restraint masking the reality of rivalry.

Perhaps we are wrong. Perhaps American judges can show the kind of heroic self-restraint that is worthy of epic poetry. But perhaps not. After all, even Ulysses needed to be tied to the mast.67 The difficulties for the mortal judge are, as we have noted, that there are fewer clear answers to legal questions than one might hope, that scholars and judges cannot agree on a single interpretive theory, that even working within one interpretive theory, there are often plausible arguments for both sides of important legal questions, and that adherents to interpretive methodologies are unreliable in their faithfulness to those theories. These are difficult challenges for mere mortals to overcome.

More generally, we find fanciful the idea that the best way to curb a too-powerful institution is by developing a better theory of how actors within that institution should behave. Ultimately, every theory of constitutional interpretation that claims to successfully rein in judges fails to account for the many and various imperfections of humans. To paraphrase then-Judge Barrett, we should take people as they are, not as we wish them to be.68

Indeed, if we want to consider the wisdom of the Founders, as Sachs asks us to, one thing we might do is observe how they chose to deal with the problem of power. After overthrowing the rule of a monarchy through a violent revolution, they designed a new government that they hoped would be less oppressive than the one they had renounced. Their solution was not simply to adopt a new monarchy, supplemented with a new, improved theory of how a just monarch should behave. Instead, they created structures and institutions that would distribute and cabin power.69 Though the Founders’ specific choices are open to question,70 this approach to constitutional design generally is, we think, the right way to think about Supreme Court reform in particular. And in the next two Parts, we discuss what we see as better solutions to the problem of restraining the power of the Supreme Court in a democracy.

II. external constraints

If internal restraints on judges are likely to prove unavailing, what other options are available for those seeking to rein in the Supreme Court? The alternatives, as we see it, are strategies that either impose constraints on the Court or that reconfigure the institution of the Supreme Court itself. We call these approaches external constraints and structural reform, respectively. In this Part, we will discuss the first option, external constraints. This method takes for granted the basic structure of the Supreme Court as an institution, but imposes some restrictions limiting its power or constraining its decisionmaking.

In framing the category of external constraints on the Court, we disagree with Doerfler and Moyn’s taxonomy of possible Court reforms. They identify two opposing categories of reform. The first is “personnel reforms”; that is, proposals “to alter the Supreme Court’s partisan or ideological composition,” a category that includes Court expansion, term limits, partisan-balance requirements, and both of our own
proposals.71 Their second category, “disempowering reforms,” are reforms which “take aim at what the Supreme Court is permitted to do.”72 This second category, which Doerfler and Moyn favor, includes jurisdiction-stripping, supermajority-voting requirements, and legislative overrides.

We value Doerfler and Moyn’s contribution to this debate, and in particular we think they bring into focus important questions about how progressives should think about courts. Their taxonomy, however, does not persuade us, because the two categories they identify are not mutually exclusive. The opposite of disempowering reforms is empowering reforms, not personnel reforms; personnel reforms can themselves be disempowering. Their taxonomy is like saying “there are two kinds of cars: electric ones and red ones.”

Nor is it obvious how some reforms even map onto their taxonomy. Would a reform that replaced life tenure for the Justices with at-will service at the pleasure of the President be a “personnel” reform, or a “disempowering” one? Doerfler and Moyn do not say; such a reform would have major consequences for both the Court’s composition and for its power (as it would be the effective end of judicial independence). The fundamental problem is that Doerfler and Moyn take a formalistic approach to disempowering reforms—looking only to actions that directly constrain the Court as currently constituted. But there are many indirect ways to constrain the Court as well. If the goal is to understand the ways to ensure a restrained Court in a democracy, we think our taxonomy better organizes the options than their approach.

In this Part, we examine external constraints more closely. We have two goals. First, we attempt to catalogue the various types of Supreme Court reforms that can be characterized as external constraints—that is, reforms that do not change the basic structure of the Court itself, but rather limit its ability to interfere with democracy by imposing some external limit on the Court’s decision making or power. Second, we offer some thoughts on when and why external constraints might be appropriate, and what their drawbacks are, with particular attention to the reforms favored by Doerfler and Moyn.

Let us start with what Doerfler and Moyn categorize as disempowering reforms: jurisdiction-stripping, legislatively mandated supermajority-voting rules, and legislative overrides. We think all are best understood as external constraints—that is, limitations or checks placed on judicial behavior. For each reform, the current structure of the Court itself remains unaffected, but its ability to render certain kinds of decisions is limited. With jurisdiction-stripping, the Court (as well as possibly other federal and state courts) is denied the ability to hear certain classes of cases, such as those involving “affirmative action or gun control.”73

Supermajority-voting rules (versions of which we included in our proposed reforms) would require the Court to have a specified supermajority in order to accomplish certain outcomes—such as declaring a federal statute unconstitutional. Doerfler and Moyn argue that such a rule “would effectively implement a Thayerian ‘clear error’ standard for judicial review.”74 Unlike jurisdiction-stripping, which cuts the Court out of the picture entirely, this reform would “preserve but severely constrain the Supreme Court’s ability to intervene in federal policymaking.”75

Finally, Doerfler and Moyn briefly discuss legislative overrides, such as a rule “letting Congress override the Supreme Court’s judgment that federal legislation is unconstitutional with a majority or supermajority vote.”76 We also think legislative overrides are a particularly useful form of external constraint, and that they deserve greater discussion. Examples of such external constraints can be found in other countries. The Canadian Charter of Rights and Freedoms, for example, has had a provision since 1982 that enables Parliament to override a constitutional decision by the high court that overturns a statute.77 The “notwithstanding” clause allows the re-enacted statute to operate in spite of the ’court’s decision. Mark Tushnet has called this arrangement “weak form judicial review,” because it preserves a role for the judiciary in reviewing statutes for constitutionality, but does not necessarily let the judiciary have the last word.78

Under current arrangements, Congress has no power to override constitutional decisions. But it has the power to override statutory decisions, and it has often used that power to express disapproval of the Supreme Court’s work.79 It did so, for example, when Congress passed the Lilly Ledbetter Fair Pay Act of 2009, which overturned Ledbetter v. Goodyear Tire & Rubber Co.,80 a decision that imposed strict time limits for filing employment-discrimination claims. The power to override statutory decisions is exercised on an occasional, ad hoc basis, and in recent years has been in decline.81 But one of us has proposed a mechanism for regularizing that power: a Congressional Review Act for the Court.82 Under the existing Congressional Review Act,83 Congress may review a regulation promulgated by a federal agency and overturn it through a fast-tracked bicameralism and presentment process. Because regulations are issued pursuant to statutory authority, Congress and the President are well within their powers to make such a revision to agency action. Using that act as a model, Congress could pass a Congressional Review Act to overturn statutory interpretation decisions from the Supreme Court. The CRA for the Court would create a similar fast-track process requiring Congress to vote on whether to overturn the Court’s statutory decisions.84

In addition to the reforms that Doerfler and Moyn highlight, many other provisions or potential reforms are properly understood as external constraints. Consider removals from office. The Constitution unquestionably permits Congress to impeach, try, convict, and remove a Supreme Court Justice from office.85 But it may be permissible to remove judges more easily. Saikrishna Prakash and Steven Smith have argued that “good-behavior tenure” under Article III does not mean that impeachment is the only mechanism for removal.86 They argue that this provision is far broader than commentators have recognized and allows for removal based on a judicial finding of misconduct.87 Even more broadly, we could imagine a constitutional system in which Supreme Court Justices were removable without cause through a simple majority vote in Congress,88 or in which they served at the pleasure of the President, as in the hypothetical discussed above.

In all cases, removal powers are an external threat to continuation in office, and will have some disciplining effect on the behavior of Justices. The extent to which they rein in the Justices will, of course, be linked to the difficulty of the removal process and the criteria for removal, both of which speak to the likelihood the threat will be carried out. Under current law, the high threshold for removal in Senate impeachment trials makes removal an idle threat. Nonetheless, it is much easier to impeach a Justice than remove them from office, as only a majority vote in the House of Representatives is necessary for the former. And, it is possible the impeachment power could be used to rein in the Supreme Court even if actual removal remains unlikely. The process of being impeached may be unpleasant and stigmatizing enough, regardless of the outcome. Those who call for Congress to impeach Justice Kavanaugh may have this goal in mind, given how hard it is to imagine two-thirds of the Senate ever supporting his removal.89

Consider also a reform that is the flipside of jurisdiction-stripping: docket filling. Today, the Court enjoys almost total discretion over its caseload; it has a small number of cases per year arising under mandatory appellate and original jurisdiction, but otherwise it has total discretion to choose cases from among the many thousands of certiorari petitions filed each year.90 It was not always thus; the Court for much of its history had mandatory appellate jurisdiction over a broader swath of cases.91 One could imagine restoring more of the Court’s mandatory appellate jurisdiction. A Court that had to decide, say, more routine criminal appeals every year might have less time and energy to devote to grand constitutional rulings that intrude on the prerogatives of the other branches. Perhaps the mere threat of such a reform might cause the Justices to dodge particularly controversial issues by denying review or even voting differently in cases before the Court. Another similar possibility is restricting the Court’s ability to choose its own cases, thus limiting its agenda-setting power and accordingly its ability to shape the law.92

A somewhat related proposal is Stephen Calabresi and David Presser’s argument for reintroducing circuit riding for Justices.93 They argue that the Justices should be required to spend four weeks each summer hearing trials in the circuits for which they serve as circuit justices. Far from an innovation, this proposal would essentially restore a practice from the very early days of the Republic, when riding circuit was an important part of each Justice’s duties. This proposal would, they argue, prevent the Justices from spending their summers in Europe (a source, in the authors’ eyes, of pernicious influence on their jurisprudence) while also providing greater motivation for elderly justices to retire.94

Ethics rules are another external constraint that deserve greater attention. Currently, Supreme Court Justices are not subject to the judicial code of conduct, nor other common-sense rules on conflicts of interest (e.g., prohibitions against trading individual stocks or taking paid trips).95 Justices also routinely omit items from their financial disclosures without penalty.96 More stringent ethics requirements might constrain Justices from hearing certain cases by creating clear recusal standards (such as in cases where there is potential for public skepticism of a Justice’s objectivity97) or from being corrupted by soft forms of influence such as being feted by outside organizations.98

Disclosure and transparency requirements might be another way to constrain the Court. Currently, no rules govern what the Justices must make public about their internal decisionmaking. Each Justice decides for themselves what papers to make public after retirement, and when. Justice Souter, for example, will not make his own papers public until fifty years after his death.99 Rules mandating quicker disclosure could enable the public to better understand how the Court makes decisions. And, to the extent that the Court’s tight secrecy enables it to maintain its aura of mystery, and thereby its prestige, mandatory disclosure rules might serve as a small but nontrivial check on the Court’s power.100

Other measures also fall into this category. Some have proposed mandating that the Supreme Court publicly broadcast oral arguments via television.101 This might have a restraining effect on the Justices’ behaviors; they might be more fearful of majority opinion if publicity increased awareness of the Court generally. Or it could have a more activist effect, by causing Justices to play to partisan bases if publicity mainly caused the most ideologically extreme portions of the public to understand the Court’s work. But in either case, it is a legislative restriction placed on the Justices’ ability to control how the Court is perceived (as would be the opposite, legislation barring public broadcast of oral arguments). Suzanna Sherry has argued for mandating that the Court issue only unsigned opinions.102 She contends that the emergence of celebrity Justices is partly a function of signed opinions, in which Justices can show their styles and play to polarized publics. Omitting signatures might lead to greater restraint.103

Other possibilities include Congress’s ability to control the Supreme Court’s budget. Though Justices’ salaries cannot be reduced under the Constitution,104 Congress otherwise has plenary control over all other appropriations related to the Court. Congress could, for example, decline to provide funds for some of the Justices’ law clerks or other staff. Or it could make the Court vacate the “Marble Palace” at One First Street. Indeed, some scholars have argued that Congress’s decision to build a separate Supreme Court building in the 1930s—rather than have the Court remain in the basement of the Capitol—pushed the Court toward judicial supremacy.105 Whether this is true as a historical matter or not, the mere threat of losing their stately quarters might deter the Court from going too far in challenging Congress.

There is thus a wide variety of external constraints, and a wide range in how constraining they are. At one extreme is jurisdiction-stripping, which completely deprives the Court of the ability to decide certain kinds of cases. At the other are gentler constraints, like Sherry’s requirement of unsigned opinions, which might effect more marginal change. Many other reforms likely fit into this category of external constraints. While we cannot enumerate all the possibilities here, defining the category can help guide further discussions on reforms.

Having done so, we will also briefly offer some thoughts about the promise, and limitations, of external constraints. Doerfler and Moyn argue that “[a]sking ‘how to save the Supreme Court’ is asking the wrong question. For saving it is not a desirable goal; getting it out of the way of progressive reform is.”106 That is, the threat they think progressives should be most concerned about is not that the Court will decline to exercise judicial review, and thus fail to protect certain rights; it is that the Court will inappropriately use the power of judicial review to block progressive legislation. This is a legitimate concern, and Doerfler and Moyn are right to emphasize it. This is especially so given liberals’ tendency to valorize the Court as a countermajoritarian defender of minority rights.107

Nonetheless, we worry that the concern that Doerfler and Moyn identify does not obviously justify their conclusions about which kinds of reforms are most appropriate. Suppose that in the future Democrats were to conclude that the biggest threat from the Court was the danger that it would strike down healthcare reforms (such as Medicare for All) and climate-change legislation (such as the Green New Deal). Doerfler and Moyn would thus recommend what they call “disempowering” reforms, such as jurisdiction-stripping, to prevent the danger that the Court would declare these reforms unconstitutional.108

But there are some significant practical problems with that strategy. Certainly, pairing a Medicare for All reform with a jurisdiction-stripping statute preventing the Supreme Court from ruling on constitutional challenges to the law might seem sensible on its face. But how would this play out in practice? First and foremost, a jurisdiction-stripping provision barring only the Supreme Court from hearing a case would be ineffective because it would leave decisions in the hands of the federal courts of appeals. Such a provision would not solve the problem Doerfler and Moyn identify; it would merely push it downward from the high court to the lower courts.

The alternative is to bar all federal courts from hearing constitutional cases. But this too has problems. As we already know from legal cases involving the Affordable Care Act, federal healthcare reforms are complex creatures with many moving parts. After the ACA survived its first major constitutional test in National Federation of Independent Business v. Sebelius,109 it faced a major statutory challenge in King v. Burwell,110 where the Court was asked to address the legality of the federal government’s practice (allegedly contrary to the statutory text) of providing subsidies to individuals who bought health insurance on federal exchanges in states that had not set up state health insurance exchanges. The Court agreed with the government’s reading of the statute as permitting the subsidies, but the case could easily have come out the other way—a result that would have led to major disruptions, perhaps causing millions to lose health insurance and severely undermining Congress’s goals in the ACA.111

Any major, complex piece of federal legislation will likely face its own challenges like King, in which the Court could interpret the statute in some textually plausible way that could cripple or undermine the reform. And a Court that has been deprived of jurisdiction to consider constitutional challenges to a reform might be less inclined to interpret a statute as charitably as the King Court read the ACA. If the Court has fundamentally different values from those of the Congress enacting the law, eliminating jurisdiction over constitutional challenges does not eliminate the threat.

Perhaps, then, the response of jurisdiction-stripping proponents might be: the solution is not merely to strip all of the federal courts of jurisdiction over constitutional challenges to the healthcare reform; it is to strip them of jurisdiction over all cases involving healthcare reform, constitutional or otherwise. Doerfler and Moyn propose such a solution when they discuss jurisdiction-stripping statutes that also “channel jurisdiction to exclusive executive branch adjudication.”112 Here, too, there are problems. Reforms like the Affordable Care Act and Medicare for All are not one-time government actions but instead long-term commitments by government to make certain services or resources available. And to function effectively, those commitments need to be durable.

But how, exactly, will any such reform be durable if entitlement holders cannot enforce their entitlements? Would the Affordable Care Act have meaningfully survived the Trump Administration if its guarantees were protected exclusively by some form of “adjudication” conducted by the Trump Administration itself? We are skeptical. Our legal system relies heavily on courts to make government promises binding over time, especially given the tendency of the Executive Branch to flip-flop between Democratic and Republican control. A jurisdiction-stripping approach needs to have a worked-out theory of how long-term commitments are possible if courts are taken out of the picture entirely.113

This point concerns jurisdiction-stripping, but it can be generalized to at least some other external constraints as well. External constraints can deter the Court from overstepping too far, but they cannot necessarily make the Court act as an effective partner in governance if it does not wish to play ball. And even if one shares Doerfler and Moyn’s view that courts impeding progressive legislative reform is a greater danger than courts refusing to interfere with legislation, there still will be situations in which the Supreme Court’s ability to enforce commitments is needed to make reform stable. External constraints such as jurisdiction-stripping alone cannot do that. Instead, to build the Court that we need, we must use the tools of structural reform—a topic to which the next Part turns.

III. structural reforms

Internal restraints and external constraints both seek to limit the Court’s power in some way, and thus ensure that the Court does not overstep its bounds in a democracy. Structural reforms are different; they seek to redesign the Court itself so that it is more likely to produce decisions that are in line with—or, at least, not too out of line with—democracy.

The insight is fundamentally Madisonian, in that Madison’s method of structural checks and balances in the original Constitution was intended to restrain the federal government without relying on a Bill of Rights that he and others thought would be little more than “parchment barriers.”114 In recent years, there has been significant scholarly commentary on the problems with the details of Madison’s approach—most notably, its failure to consider economic power and political parties.115 But the Madisonian insight nonetheless remains a powerful one: the design of institutions will invariably shape their scope and exercise of power.

This is another place where we object to Doerfler and Moyn’s framework. By juxtaposing personnel reforms with disempowering reforms, Doerfler and Moyn fail to recognize that “personnel” reforms (as well as other structural reforms) can be empowering or disempowering, depending on their details.116 In our earlier Feature, we proposed significant structural reforms to the Court, paired with some external constraints. The particular combination of reforms we chose was designed to accomplish the goals we laid out. We hoped to put “a thumb on the scale in the direction of deference” while also noting that “some role for judicial review is important, so that the Court can hold the nation to its deepest commitments and check its worst injustices.”117

Our two proposals sought to operationalize those goals in distinct ways. The Supreme Court Lottery proposed to expand the Supreme Court to include all of the federal appellate court judges, and then to constitute temporary panels of nine Justices for short periods.118 The panels would be limited by partisan-balance requirements, so that no more than five Justices on a nine-Justice panel could be appointed by presidents of one political party. The Balanced Bench offered a different set of structural reforms towards the same ends. It imagined expanding the Court to fifteen Justices, with five Justices affiliated with each of the two leading political parties and an additional five selected for temporary service by the other ten through unanimous or near-unanimous consensus.119

Our goal here is not to defend these proposals in particular. Instead, we wish merely to urge potential reformers—including those who share Doerfler and Moyn’s concerns—to see the value of structural reforms. We think there is great promise in the kinds of structural reforms we favor because they can disempower the Court so that it does not exceed its proper role in our democracy while also building the kind of Court we need to serve as an effective participant in democratic governance. To be clear, our argument is not that structural reforms alone are the path forward—after all, our own proposals incorporated both structural reforms and external constraints. Our point is that reformers must not dismiss structural reforms as a categorical matter.

We will briefly consider a few examples of structural reforms to illustrate their potential. Eric Segall has argued for partisan-balance requirements on the Court, with an even number of Supreme Court Justices.120 Taking inspiration from the Court’s 4-4 Term after the death of Justice Scalia, in which the Court reached more consensus than it had in the seventy years prior,121 Segall notes that an even split along party lines would force the Justices to find compromises and write narrower opinions. In cases with ideological salience, the partisan-balance structure builds in conflict and thus forces restraint, compared to the baseline of partisan majorities exercising their will.

Note that nothing about partisan-balance requirements formally prevents the Court from making broad, substantive decisions—including striking down federal statutes. The proposal is thus not the same as an external constraint like jurisdiction-stripping or a supermajority-voting rule. But at the same time, the design itself should—and indeed is intended to—create political dynamics on the bench that tend toward narrower and more restrained decisions. Similarly, our Balanced Bench proposal sought to prevent the Court from issuing decisions wildly out of step with democracy by combining partisan-balance requirements with a rule requiring consensus on the appointment of additional temporary Justices from the lower courts before the Court could function. With such a system, random chance no longer could produce a Court that bore little correspondence to the results of the political process.

Other reforms that are being actively considered would also change the Court’s structure in order to make it more responsive to democracy in some way. Consider proposals for staggered, eighteen-year term limits, of which there are many variations in the literature.122 These reforms would provide that each presidential term would be guaranteed two appointments to the Court, thus ensuring that the Court’s composition over time would more closely track the results of presidential elections. This reform would disempower individual Justices, but it would not necessarily disempower the Court itself. For that reason it may be understandably less attractive to those whose goal is Court disempowerment.

Consider also more traditional Court expansion (or Court “packing”), in which a party currently in power adds seats to the Court for the purpose of gaining partisan advantage. This approach would not appear to disempower the Court; it would instead simply keep its formal power constant, while changing the composition of its majority in order to shape how that power is exercised. But this strategy might have significant consequences for the Court’s power as a practical matter. To the extent that the move is seen as illegitimate, it could lead to pressure on officials to refuse to follow facially lawful Supreme Court rulings. It also could lead to a new set of norms in which each time Congress and the White House changed hands, new seats were added to the Court. That development might make the Court more democratically responsive, but it could potentially make it less powerful, if its decisions are seen as less permanent.

Here, again, Doerfler and Moyn’s taxonomy leaves us without the right conceptual vocabulary to distinguish reforms. Some “personnel” reforms, like Court expansion, could potentially be disempowering or could keep the Court’s power constant. Yet Doerfler and Moyn’s approach looks only to the formal way in which a reform operates, without asking the deeper question of how the reform might actually affect the Court’s power in practice after implementation.

Importantly, structural reforms can also be combined with external constraints. Take our proposal for the Supreme Court Lottery. As noted, the idea is that all of the judges on the federal courts of appeals would be appointed Associate Justices of the Supreme Court, and the Supreme Court would hear cases in panels of nine randomly selected from its full membership. This design structure has much to offer: it would draw judges from all over the country, with greater professional diversity, and from a wider set of legal backgrounds. It would also lower the stakes significantly of individual confirmation battles, given that any one Justice would hear cases on the Supreme Court only occasionally.

But one might worry that this design would lead to wild fluctuations based on panel composition. The partisan-balance requirement would help prevent an 8-1 conservative court swinging to an 8-1 liberal court. Here is where external constraints can also come into play. Our proposal, for example, would also impose supermajority-voting rules to further rein in the power of a court to strike down federal statutes.123

Doerfler and Moyn’s approach, however, omits the possibility of using structural reforms to disempower the Court indirectly or combining them with external constraints. Indeed, they ignore the combination of external constraints and structural-reform strategies in our proposals. But as we have seen here, structural reforms undoubtedly can be restraining. The degree to which they are disempowering and the mechanism by which they disempower the Court will of course vary depending on the details of the reform. Moreover, structural reforms do not suffer from the central drawback we identified with focusing only on external constraints like jurisdiction-stripping. In our legal and constitutional context, we need a Court that can resolve important issues in order to ensure that policies are stable.

Conclusion: Why Democracy Needs To Reform—and Save—the Supreme Court

Our framework for understanding democracy-reinforcing Supreme Court reforms has important implications. First, as we have argued, continued debates over interpretive methodology are unlikely to be helpful from the perspective of advancing democratically-respectful judicial restraint. Versions of Court reform along these lines—such as the uniform adoption of a single interpretive methodology like originalism—have been hotly debated for more than a half-century. We have reached no consensus about which theory is correct; and even within single theories, there are often plausible arguments for various outcomes. Moreover, a realistic understanding of human nature provides strong reasons for skepticism that this is, by itself, a useful way to restrain the Court in a democracy.

External constraints are more promising. Yet they suffer from their own limitations—limitations that deserve greater attention. Even if one primarily hopes courts will “get[] . . . out of the way of progressive reform,”124 there is no way to get courts entirely out of the way in our constitutional structure without at the same time precluding the ability of government to credibly commit to the very kinds of policy initiatives that progressives support. As we have shown, proposals such as jurisdiction-stripping run up against severe problems when confronted with either a hostile Court engaged in manipulative statutory interpretation or a future administration bent on ignoring or undermining a policy initiative. Proponents of external constraints have not—so far as we are aware—offered practical solutions to these difficulties. While some external constraints can be useful parts of larger reform packages, they are themselves unlikely to be the entire solution.

The limitations of both internal restraints and external constraints leave us with needing to keep structural reforms on the table: changes that seek to design a Court that is appropriately deferential to democracy, while also serving as a useful partner in governance. There are many possible structural reforms, including but certainly not limited to those we have proposed. Each has its own advantages and drawbacks, and our goal here is not to argue for any one proposal. But as scholars, commentators, and policymakers think about reforming the Supreme Court, they should recognize that structural reforms are critical if the goal is ensuring that the Court stays within its appropriate place in a democracy. In other words, the path forward cannot simply be to ask judges to exercise restraint or to disempower the judiciary; it must be to save the Supreme Court.

Daniel Epps is Associate Professor of Law, Washington University in St. Louis. Ganesh Sitaraman is Professor of Law and Director of the Program in Law and Government, Vanderbilt Law School. The authors are grateful to workshop participants at Drexel Kline School of Law and to helpful suggestions from Anil Kalhan, Michael Klarman, David Pozen, and the editors of the Yale Law Journal Forum. Samantha Kobor provided outstanding research assistance.