Reprint From SSRN, How to Save the Supreme Court by Daniel Epps and Ganesh Sitaraman



How to Save the Supreme Court


Daniel Epps*  & Ganesh Sitaraman**


[DRAFT November 21, 2018]  

The consequences of Brett Kavanaugh’s confirmation to replace Justice  Anthony Kennedy on the Supreme Court are seismic.  The new conservative majority that Kavanaugh completes represents a stunning victory for the Republican party after decades of effort by the conservative legal movement. The result is a Supreme Court whose justices—on both sides—are likely to vote along party lines more consistently than ever before in American history. That development presents a grave threat to the Court’s legitimacy. If in the future roughly half of Americans lack confidence in the Supreme Court to render impartial justice, the Court’s ability to reach settlements of important questions that all Americans can live with is serious jeopardy.  Raising the stakes even higher, many  Democrats are already calling for changes like court-packing to prevent the new conservative majority from blocking progressive reforms. Even if justified, such moves could provoke further tit-for-tat escalation that would leave the Court’s image, and the rule of law, badly damaged.

The coming crisis can be stopped. But preserving the Court’s legitimacy as an institution above politics will require a complete rethinking of how the Court works and how the Justices are chosen.  To save what is good about the Court, we must reject and rethink much of how the Court has operated for more than two centuries. In this Essay, we outline a framework for thinking about saving the  Supreme Court,  evaluate existing proposals,  and offer two distinct reform proposals of our own, which we call the Supreme Court Lottery and the Balanced Court.  Whether policymakers adopt these precise proposals or not, however, it is imperative that they search for some kind of reforms along these lines. Saving the Court—by transforming the Court—is our best hope.

* Associate  Professor of Law, Washington University in St. Louis.

** Professor of Law and Chancellor’s Faculty Fellow, Vanderbilt Law School. We would like to thank Will Pugh for helpful research assistance. The proposals developed here first appeared in Daniel Epps & Ganesh Sitaraman, How to Save the Supreme Court,  VOX, Sept. 6, 2018 (updated Oct. 10, 2018), idea/2018/9/6/17827786/kavanaugh-vote-supreme-court-packing.


INTRODUCTION …………………………………………………………………………………………………………………. 2

I.       THE LOOMING THREAT …………………………………………………………………………………………. 4

II.     WHY SAVE THE COURT? …………………………………………………………………………………………. 9

III.   HOW (NOT) TO SAVE THE COURT…………………………………………………………………… 11

A.      Desiderata for Reform ………………………………………………………………………………………….. 11

  1. How Existing Proposals Shape Up ……………………………………………………………………….. 1
  2. IV.    SAVING THE COURT: TWO PROPOSALS…………………………………………………………. 16
  3. A.      The Supreme Court Lottery …………………………………………………………………………………. 17
    1. The Plan and its Benefits ……………………………………………………………………………… 17
    2. The Constitutionality of the Supreme Court Lottery ………………………………… 18
    3. (a)    Dual Appointments …………………………………………………………………………….. 19
    4. (b)    The Vesting Clause and “One Supreme Court” ………………………………… 21
    5. (c)    Supermajority Voting Requirements ………………………………………………….. 23
    6. (d)     Historical Practice ………………………………………………………………………………. 24
  4. The Balanced Bench …………………………………………………………………………………………….. 25
  5. The Plan and its Benefits ……………………………………………………………………………… 25
  6. Constitutional Objections……………………………………………………………………………. 30
  7. (a)    Appointments Clause Challenges ………………………………………………………. 30
  8. (b)     Partisan Balance Requirements…………………………………………………………… 32

CONCLUSION ……………………………………………………………………………………………………………………. 33



The consequences of Brett Kavanaugh’s confirmation to replace Justice Anthony Kennedy on the Supreme Court are seismic. The new conservative majority that Kavanaugh completes represents a stunning victory for the Republican party after decades of effort by the conservative legal movement—and, by the same token, a significant defeat for Democrats and the American left. Republicans, to be sure, look like the big short-term winners. But the ultimate loser here isn’t just their Democratic opponents. It’s the Supreme Court itself—and, eventually, the American people as a whole.


Recent events have already taken a heavy toll on the Court. Kavanaugh’s confirmation vote was one of the closest in American history. The vote came after a process that deeply divided the country when the Republicans stuck with the nominee after serious accusations of sexual misconduct—and even after Kavanaugh gave testimony to the Senate Judiciary Committee that many viewed as “nakedly partisan.”1 President Trump’s first nominee to the Court, Neil Gorsuch, was able to join the Court only because of unprecedented hardball tactics by Senate





1     Zack   Beauchamp,   The    Supreme   Court’s    Legitimacy   Crisis    Is   Here,    VOX,   Oct.   6,   2018, court-legitimacy.




Majority Leader Mitch McConnell, who, after Justice Scalia’s death, stonewalled President Obama’s nominee, Merrick Garland, to leave the seat open. But these debacles were only the latest in an increasingly politicized fight over justices. The predictable result is a Supreme Court whose justices—on both sides—are likely to vote along party lines more consistently than ever before in American history. Soon, “it will become impossible to regard the court as anything but a partisan institution,” Lee Epstein and Eric Posner warn.2


That development presents a grave threat to the Court’s legitimacy. If in the future roughly half of Americans lack confidence in the Supreme Court to render impartial justice, the Court’s ability to reach settlements of important questions that all Americans can live with is serious jeopardy. Raising the stakes even higher, many Democrats are already calling for ideas like court- packing to prevent the new conservative majority from blocking progressive reforms. Even if justified, such moves could potentially provoke further tit-for-tat escalation that would leave the Court’s image, and the rule of law, badly damaged.


Can this coming crisis be stopped? Or, to make what’s at stake starker: can the Supreme Court be saved? We think so. But preserving the Court’s legitimacy as an institution above pol- itics will require a complete rethinking of how the Court works and how the Justices are chosen. To save what is good about the Court, we must reject and rethink much of how the Court has operated for more than two centuries. The Court must radically change—or die.


And the Court is, we think, worth saving. American democracy could probably still func- tion if the Supreme Court had so little political capital that it could rarely, if ever, stand up to the political branches. But there are good reasons to want an institution like the Court that can sometimes provide a check on the political process and hold us to our deepest commitments. But more importantly, public confidence in the Supreme Court is impossible to untangle from public confidence in the very idea of law itself as an enterprise separate from politics. And a democracy that loses its confidence in law may not long survive.


In this Essay, we offer a framework for thinking about saving the Supreme Court. We ex- plain how only Supreme Court reforms—and only the right kind of reform—can preserve the Court’s role as a neutral arbiter of important questions of law. We begin in Part I by discussing why the Court’s legitimacy faces significant peril in the near term. Several factors—such as in- creased polarization in society in general, the development of polarized schools of legal inter- pretation that line up with political affiliations, and increased interest-group attention to the Supreme Court nomination process—have conspired to create a system in which the Court becomes a political football and in which each party’s nominees can be expected to predictably vote along ideological lines. Justice Kennedy—even though he was for the most part a reliable conservative—may well be the last Justice to vote differently against his partisan affiliation in some of the highest-profile cases. With his replacement by a more ideologically predictable Jus- tice, the notion of the Court as an institution above the political fray may soon vanish.





2 Lee Epstein & Eric Posner, If the Supreme Court  Is Nakedly Political, Can It Be Just?, N.Y. TIMES, July 9,





Next, in Part II, we consider what kinds of reforms would protect the Court’s perceived role as a legitimate and nonpartisan arbiter of important questions of law. Any solution must have at least three components. First, it must be constitutionally plausible, even if not constitu- tionally bulletproof. Second, it must be capable of implementation via statute, given the near impossibility of a constitutional amendment in an age of severe polarization, Finally, even though overwhelming bipartisan support for reform might not be possible at the time of re- form, the proposal needs to be at least plausibly stable going forward. That is, it has to be some- thing that both sides might be able to live with, leading to a fair, stable equilibrium. Unfortu- nately, some of the most prominent prior reform proposals do not satisfy these criteria; and in some cases, they could make the politicization of the Court worse.


Finally, and most importantly, in Part III we offer two distinct reform proposals of our own, which we call the Supreme Court Lottery and the Balanced Court. We offer two alternative approaches because policymakers might have different views about the viability of either ap- proach at the point that Congress takes up Supreme Court reform. For each, we discuss the plan and its benefits and then assess the constitutionality of the proposal. We think either pro- posal would be an excellent framework for reform. Though neither would perfectly solve all the problems we identify with the way the Supreme Court currently works, both would solve many of those problems and at the very least would be a marked improvement over how things cur- rently work.


Whether policymakers adopt these precise proposals or not, however, it is imperative that they search for some kind of reforms along these lines. Doing nothing is not a good option. The Court will be gravely damaged by clashes between the conservative majority and progressive politicians once Democrats regain power. But nakedly political reforms—even if a justified re- sponse to Republican escalation—may not lead to a stable equilibrium and could end up dam- aging the rule of law. Saving the Court—by transforming the Court—is our best hope.




As many observers have noted, the Supreme Court is facing an unprecedented legitimacy crisis in the wake of Justice Kennedy’s retirement and his replacement with Justice Kavanaugh.3

Observers note several serious dangers facing the Court going forward. First is the seemingly undeniable fact that the Court will be more polarized along party  lines than at any point in recent memory—probably ever. As Professors Epstein and Posner explain, Justice Kennedy was the last Justice on the Court to regularly vote against the ideology of the President who ap- pointed him.4 He has been replaced by a much more reliable conservative (Kavanaugh). Thus, “[f]or the first time in living memory, the court will be seen by the public as a party-dominated institution, one whose votes on controversial issues are essentially determined by the party af- filiation of recent presidents.”5  Indeed, even when Democratic President Franklin Roosevelt




3 See Beauchamp, supra note 1.

4 Epstein & Posner, supra note 2.

5 Id.




proposed his famous court-packing plan in the 1930s, his antagonists on the Supreme Court were not all of the opposing Republican party. One of the “four horsemen,” Justice James McReynolds had been appointed by Democratic President Woodrow Wilson.6  It is hard to know with certainty what the rise of a partisan Court portends, but it is hard to imagine that the Court will continue to enjoy public confidence if half the country sees the majority of Jus- tices as nothing more than political agents working for the other team.


Others raise different legitimacy concerns. A distinct problem is the Supreme Court’s lack of democratic pedigree. Of course, the “countermajoritarian difficulty” posed by the Court has been a topic of conversation for decades among constitutional theorists.7 Today, though, the Court has become particularly countermajoritarian. The problem is not just that the Justices themselves are insulated from politics through life tenure; it is that the political actors selecting them suffer from serious democratic deficits. As Michael Tomasky notes, the two most recent additions to the Court were selected “by a president and a Senate who represent the will of a minority of the American people.”8


These more general concerns are no doubt exacerbated by the particular circumstances of how the two newest Justices joined the Court. As noted, Justice Gorsuch was only able to join the Court after Senate Republicans’ unprecedented blockade of President Obama’s nominee, Judge Merrick Garland. The Court was left with only eight justices for more than a year after Justice Scalia’s death; Senate Republicans refused to even hold a hearing for Garland, despite his incontrovertible qualifications and relative centrism—and in the face of majority support for his confirmation among the American people.9 The inescapable conclusion was that the party affiliation of Supreme Court justices matters—and that the politicians will go to great lengths to capture control of the Court. The Senate’s treatment of Garland has generated sig- nificant and lasting outrage on the left, leading many to view Justice Gorsuch as illegitimate.


One might have hoped that the nomination of Judge Kavanaugh to replace Justice Kennedy might have been less damaging. To be sure, the nomination was high stakes; Justice Kennedy had been the “swing” justice for many years, and the chance to replace him with a more reliable conservative gave Republicans a chance to reshape the law. Yet Justice Kennedy’s seat wasn’t stolen; under pre-Garland norms, the vacancy was President Trump’s to fill by right, given that it became open during his Presidency. Most expected a swift, relatively uneventful confirmation process.



PEOPLE (2009).

8  Michael Tomasky, The  Supreme Court’s  Legitimacy Crisis,  N.Y. TIMES, Oct. 5, 2018, https://www.ny-

9  See Supreme Court,  GALLUP, (noting results of a

March 2016 survey showing 52% support for Garland’s confirmation).

That was not to be. Days before the Senate Judiciary Committee was to vote on Ka- Kavanaugh’s nomination—and after hearings had concluded—Dr. Christine Blasey Ford came forward to allege a serious sexual assault by Kavanaugh during high school. The Ford allegations captured public attention. More allegations emerged, and the Judiciary Committee was forced to delay its vote and hold a hearing at which both Ford and Kavanaugh testified. At that hear- ing, Kavanaugh offered testimony that is fairly described as shocking. He lambasted the “two- week effort” surrounding the allegations as “a calculated and orchestrated political hit,” that Kavanaugh blamed on, among other things, “[r]evenge on behalf of the Clintons.”10  He went on to address Democratic committee members with contempt and disrespect. Observers condemned Kavanaugh’s performance as highly improper for a judge, with many saying that his testimony disqualified him for the Supreme Court even if the underlying allegations were not true.11  Some even alleged that he lied under oath about his behaviors as a student and other matters.12 Going forward, it will be hard for many Americans to believe that Kavanaugh is a fair, impartial jurist.


Given all this, it is not surprising that many think the Court’s legitimacy faces its most se- rious challenge in many years. Concerns about legitimacy are by no means limited to the liberal commentariat; but have been voiced by mainstream political figures. Former Attorney General Eric Holder, for example, suggested that “[w]ith the confirmation of Kavanaugh and the pro- cess which led to it, (and the treatment of Merrick Garland), the legitimacy of the Supreme Court can justifiably be questioned.”13  Even a sitting Supreme Court Justice, Elena Kagan, re- cently warned that it was “a dangerous time for the Court” because “people increasingly look at us and say ‘this is just an extension of the political process.’”14


And of course, we haven’t even discussed the legitimacy concerns that will be raised by the actual decisions the Supreme Court will render in the coming years. There’s every reason to expect the new conservative majority will assert its power in high-profile cases that will provoke significant controversy. Most obvious is the possibility—though not the certainty—that the Court will overturn Roe v. Wade15 and thereby permit state legislatures to criminalize abortion. Abortion has many opponents in American society the opinion in Roe itself has been heavily

10 Kavanaugh Hearing: Transcript, WASH. POST, Sept. 27, 2018, tional/wp/2018/09/27/kavanaugh-hearing-transcript/?utm_term=.d576f6f00485.

11 See, e.g., Laurence H. Tribe, All the Ways a Justice Kavanaugh Would Have to Recuse Himself,  N.Y. TIMES, Oct. 1, 2018 (describing Kavanaugh’s “intemperate personal attacks” and “his partisan tirades” as “display[ing] a strikingly injudicious temperament”).

12 James Roche, I was Brett Kavanaugh’s College Roommate, SLATE, Oct. 3, 2018 (“Brett Kavanaugh stood up under oath and lied about his drinking….”).

13  Eric Holder (@EricHolder), TWITTER, Oct. 6, 2018 1:10  p.m., tus/1048666766677876738.

14 Ian Millhiser, Kagan Warns That the Supreme Court’s Legitimacy is in Danger,  THINKPROGRESS, Sept. 17,



15 410 U.S. 113 (1973).


criticized by commentators across the political spectrum.16 Nonetheless, Roe has become deeply embedded in our culture and its explicit rejection by the Court would no doubt provoke mas- sive outrage on the left and would make the Court even more of a focal point of our politics than it is already.


Even if the Court doesn’t pull the trigger on overturning Roe outright, there is little doubt that the justices will intervene in many hot-button issues over the coming years. The intersec- tion of gay rights and religious liberty; the rights of corporations; constitutional challenges to federal legislation under the Commerce Clause; thorny issues of free speech, and many other possibilities. And there’s every reason to expect that in at least some instances, the Court will opt not for Thayerian deference to the political branches but instead will take strong conserva- tive positions.17 Last Term’s decision in Janus v. American Federation of State, County, and Mu- nicipal Employees,18 which dealt a crippling blow to public-sector unions, may provide a blue- print for how an emboldened majority might advance conservative interests using aggressive new doctrines—in that case, what Justice Kagan called the “weaponiz[ed]” First Amendment.19


In a world where the public had great confidence in the Supreme Court’s fairness and im- partiality, it is possible that Americans would accept many controversial decisions like this even if they did not agree with the outcome. Indeed, social science research has found some evidence for the proposition that the Supreme Court is more effective than other institutions at legiti- mizing unpopular decisions.20 Yet in a world where the public has lost faith in the idea that the Justices are fair and impartial—rather than just being political hacks—it is far from clear that the public will accept unpopular decisions. Though the point is contested, there is support for the view that Supreme Court legitimacy is strongly tied up with perceptions of how the Court makes decisions—particularly, the perception that the Court is reaching decisions impartially and using fair processes.21 Moreover, if the Court’s most salient decisions are almost universally


16 See, e.g., John Hart Ely, The Wages of Crying Wolf: A Comment on Roe v. Wade, 82 YALE L.J. 920 (1973); Ruth Bader Ginsburg, Some Thoughts on Autonomy and Equality in Relation to Roe v. Wade, 63 N.C. L. REV. 375 (1985).

17 James Bradley Thayer, The Origin and Scope of the American Doctrine of Constitutional Law, 7 HARV. L. REV.

129 (1893).

18 138 S. Ct. 2448 (2018).

19 Id. at 2501 (Kagan, J., dissenting).

20 See James L. Gibson, Understandings of Justice: Institutional Legitimacy, Procedural Justice, and Political Tol- erance, 23 LAW & SOC. REV. 469, 480–81 (1989) (finding, based on responses to surveys, “some evidence of the Court’s capacity to engender compliance with unpopular political decisions”).

21 See Tom R. Tyler & Kenneth Rasinski, Procedural Justice,  Institutional Legitimacy, and the Acceptance of Unpopular U.S. Supreme Court Decisions: A Reply to Gibson, 25 LAW & SOC. REV. 621, 627 (1991) (concluding that the “legitimacy of the U.S. Supreme Court is based on the belief that it makes decisions in fair ways, not on agreement with its decisions.”).




wins for one side—Republicans—the Court’s legitimacy may be affected much more than if its controversial decisions sometimes favored Democrats.22


The Court’s legitimacy may also be threatened by potential responses by Democrats to Re- publicans’ aggressive tactics. Facing the realistic prospect that the conservative majority willl block progressive legislative efforts, many on the left are already trying to identify potential re- sponses that would either reduce the Court’s power or disrupt the Republican stranglehold on its decisionmaking. Perhaps most prominently, Court packing is being discussed as a serious proposal after being seen as beyond the pale for decades.23 Although Congress has enlarged and decreased the size of the Court at various points in American history, often for nakedly political reasons,24 the Court’s membership has been set at nine for more than a century. Famously, Pres- ident Franklin Roosevelt advanced a plan to add additional justices to the Court after promi- nent losses for his New Deal agenda at the hands of a 5-4 conservative majority. Although the threat alone may have been sufficient to push the Court away from striking down more New Deal programs, Roosevelt’s court-packing plan was defeated.25 The proposal’s failure has re- sulted in many thinking of court-packing as a political third rail—making the Court’s current size look like an entrenched, quasi-constitutional norm.26 Yet the prospect of a young conserva- tive majority striking down Democratic legislation for decades has called that conventional wis- dom into question. Moreover, given that the Senate Republican majority effectively practiced Court-shrinking by refusing to give President Obama a chance to fill the seat that, by all rights, was his to fill, it is hard to justify taking court-packing entirely off the table as a justified re- sponse.





22 Cf. James L. Gibson & Michael J. Nelson, The Legitimacy of the US Supreme Court: Conventional Wisdoms and Recent Challenges Thereto, 10 ANN. REV. L. & SOC. SCI. 201, 209 (2014) (noting that “[l]ack of polarization [in peceptions of Supreme Court legitimacy] may also reflect the fact that the Supreme Court is currently making about 50% of its decisions in a conservative direction and 50% in a liberal direction”).

23  See, e.g., Michael Klarman, Why Democrats Should  Pack the Supreme Court,  TAKE CARE BLOG, Oct. 15,

2018; Aaron Blake, Pack the  Supreme Court?  Why  We  May  be Getting Closer,  WASH. POST, Oct. 9, 2018, closer/?noredirect=on&utm_term=.902a9c07cfe6, Ian Samuel, Kavanaugh Will be on the US Supreme Court for Life.  Here’s  How  We  Fight   Back,  GUARDIAN, Oct.  9,  2018, free/2018/oct/09/kavanaugh-us-supreme-court-fight-back-court-packing.

24 In 1863, in the midst of the Civil War, Congress expanded the size of the Court from nine to ten justices, a move which helped shore up support for Republican, pro-Union interests on the Court. Timothy Huebner The First Court-packing Plan,  SCOTUSBLOG, July 3, 2013, packing-plan/. Then, during the Presidency of Andrew Johnson, Congress reduced the Court’s membership to seven—preventing Johnson from appointing any Justices—before expanding it back to nine after Johnson left office. Id. The size of the Court has remained at nine since then. Id.

25 For a fascinating history of this episode, see JEFF SHESOL, SUPREME POWER: FRANKLIN ROOSEVELT VS. THE


26  See Tara Leigh Grove, The  Origins  (and Fragility) of Judicial Independence, 71 VAND. L. REV. 465, 505 (2018); Curtis A. Bradley & Neil S. Siegel, Historical Gloss, Constitutional Conventions, and the Judicial Separation of Powers, 105 GEO. L.J. 255 (2017).




Alternatives to court-packing are also under active discussion. Professor Samuel Moyn has argued that the left should “stand up for reforms that will take the last word from” the Court.27

He specifically points to jurisdiction-stripping statutes as well as “[o]ther changes in customs and precedent” that could “weaken judicial supremacy,” such as efforts to push the Court to “evolve into an advisory body, especially when the justices disagree.”28 Professor Mark Tushnet has been advancing arguments for abolishing judicial review for a number of years,29 and his proposals are receiving renewed interest.30


We cannot be certain that ideas like jurisdiction-stripping or court-packing will actually get off the ground—or even that Democrats will obtain the necessary control over Congress and the Presidency to make it even a possibility. But the fact that people are even discussing such ideas tells us how serious the situation is. The Court’s legitimacy could be questioned in the coming years—perhaps as it never has before. Indeed, even those who think the threat might be overblown still think coming challenges to the Court’s legitimacy need to be taken seriously.31




We think there is reason to be concerned about the looming threat to the Supreme Court legitimacy. A Supreme Court that is viewed as illegitimate by a significant chunk of the popu- lation will be one that is less able to settle important questions, and particularly less able (or, perhaps, willing) to exercise the power of judicial review to declare laws unconstitutional. Of course, for many on the left today, that may seem like a desirable goal. Those who favor Moyn’s critique of “juristocracy,” for example, or who are drawn to Tushnet’s arguments against judicial review, would likely welcome developments that would weaken the Court’s ability to stand up to the other branches of government.


At one level, we are sympathetic to some of the arguments offered by the critics. Judicial review is inescapably anti-democratic. And while it has served some important purposes at var- ious points in American history, it is also a power that the Court has unquestionably abused, as the Justices have taken on responsibility for resolving thorny questions that might have been better left to resolution by elected officials. Nonetheless, we have deep reservations about the long-term consequences of a powerless Supreme Court.


First, consider what it might look like if the Supreme Court suddenly became totally una- ble, or at least significantly less able, to exercise the power of judicial review. Certainly, such a






27  Samuel Moyn, Resisting  the  Juristocracy, BOSTON REV., Oct. 5, 2018, tice/samuel-moyn-resisting-juristocracy.

28 Id.


30      See    Sean   Illing,   The    Case    for    Abolishing   the    Supreme   Court,    VOX,   Oct.   12,   2018,

31 See Ilya Somin, Is the Supreme Court Going to Suffer a Crisis of Legitimacy?, VOLOKH CONSPIRACY, Oct. 10,





development would not spell the end of American democracy. Indeed, a number of other coun- tries either lack written Constitutions, do not permit courts to enforce their written constitu- tions through judicial review, or have mechanisms by which the legislature can reenact laws that the courts have struck down.32 Far from being tyrannical, these countries—places like England, the Netherlands, and Canada—are functional democracies. These examples suggest that it is certainly possible to have a well-functioning democracy that respects individual rights without giving courts the ability to strike down legislation as unconstitutional. Moreover, the United States itself barely had a system of judicial review of federal statutes during its early years, as the Supreme Court only exercised that power twice before the Civil War.33


Nonetheless, there is still reason to worry that a significant change to, or outright elimina- tion of, the Court’s power of judicial review could have bad consequences. Even if other democ- racies function well without judicial review, it doesn’t follow that our own system would func- tion equally well if judicial review were abolished or significantly curtailed. Whatever its merits, judicial review has been an integral part of the American constitutional system for a long time at this point. No one can know what would happen if it disappeared tomorrow. Perhaps the political branches would, more or less, safeguard basic rights, the way that legislatures do in other democracies. But perhaps not; perhaps political actors in our system have become so ac- customed to being reined in by courts that, once set free, they would trample over important rights. It is hard to know for certain.


The implications for judicial review, however, are in our view ultimately secondary con- cerns when it comes to the legitimacy challenge faced by the Supreme Court. The larger prob- lem is this. The Supreme Court plays a significant role in the public imagination as a citadel of justice. For many Americans, given the Supreme Court’s salience, faith in the Court may be deeply intertwined with feelings about the very idea of law more generally.34 And in a world where the Supreme Court is widely seen as nothing more than another political institution— where Supreme Court justices are considered nothing more than politicians in robes—how will people think about law itself? Our fear—and we think it is well grounded—is that in such a world, the very idea of law as an enterprise separate from politics will evaporate.


If that happens, our fears for democracy start to become much more concrete. Can a dem- ocratic society long survive if the citizenry loses faith in law? The rule of law is a critical element of a healthy democracy. Will the notion of the rule of law survive if people stop believing that judges are doing something other than exercising political will when deciding cases? Will polit- ical partisans cease to give credence the results of any legal proceeding that does not validate their preexisting political beliefs? We do not know the answers to these questions. But we are not eager to run the experiment required to answer them. Instead, we think it is important to




32 For a discussion, see Mark Tushnet, Dialogic Judicial Review, 61 ARK. L. REV. (2009); Mark Tushnet, Alter- native Forms of Judicial Review, 101 MICH. L. REV. 2781 (2003).

33 The cases were Marbury v. Madison, 5 U.S. (1 Cranch.) 137 (1803) and Scott v. Sandford, 60 U.S. (19 How.)

393 (1857).

34 Cf. Richard J. Pierce, Jr., Reconciling Chevron and Stare Decisis, 85 GEO. L.J. 2225, 2242 (1997) (noting that the Supreme Court is “the most salient symbol of the rule of law in our society”).




try to save the Supreme Court—particularly, to save the idea of the Court as an institution above the political fray.


Saving the Court, however, will unavoidably require changing the Court. Our current sys- tem is deeply flawed, and events from 2016 on have only exposed problems that were long lurk- ing below the surface. The problem is that the consequences of individual Supreme Court ap- pointments are so significant that it is impossible to expect political actors to do anything other than what they have been doing in recent years—fighting tooth and nail to take control of the Court at all costs. These flaws were less apparent in an age when the leading political parties were less polarized. But now, with the extreme ideological sorting that has taken place, politi- cians of both parties now realize the stakes of Supreme Court appointments, and are firmly committed to doing whatever it takes to staff the Court with devoted ideological comrades.


Some conservatives will no doubt argue that the solution to the legitimacy crisis we have laid out is to simply reject the challenge, and to treat the Court as legitimate. Yet things are not so simple. The new Supreme Court majority is the most conservative in history, and there is every reason to expect it to strike down all sorts of laws that Democrats favor using all sorts of novel theories not firmly grounded in the Constitution’s text. And that conservative majority was earned using underhanded tactics. We do not think Democrats should be required to ac- cept perpetual defeat of their favored policies for a generation or more simply to maintain the legitimacy of the Court. But even if we did, what we think on this score is unimportant. Others would disagree, and the stakes will make it inevitable that the Court’s legitimacy will be chal- lenged head-on. To avoid that collision, we need to change course—radically.




Saving what is good about the Court will require significant reform of the way the Court operates and how the justices are selected. But not just any reform will do. In this Part, we first explain what exactly reform would need to look like to be successful. We then go on to discuss how previous reform proposals would not accomplish the goals we think are necessary to save the Court but could make some of the problems we are most concerned about worse.


  1. Desiderata for Reform


The reform that we envision would have multiple goals. First, it would be designed to en- sure that the Court would remain an institution that is seen as above politics. Second, it would significantly reduce the political stakes of nominating individual justices, to avoid spectacles like we have seen over the last couple years. Related to that point, it would need to significantly lessen the importance of individual justices. In our current system, far too much turns on essen- tially random events. When any one of the justices dies or retires can have massive consequences for the law, depending on when the vacancy occurs and, it seems, whether the President’s party controls the Senate. This is not a sensible way to run a constitutional democracy. Whatever your views on abortion, free speech, or the powers of Congress, these issues should not depend on the health of individual octogenarians. No one would design such a system from scratch, and any good set of reforms would try to make the Court less sensitive to the choices and health of individual justices.




Third, a better system would, we think, preserve some ability for the Justices to strike down laws while perhaps pushing them at least a bit in the direction of deference to the political branches. In our view, some role for judicial review is important, so that the Court can hold the nation to its deepest commitments and check some of the worse injustices. But there are good arguments that the justices (on both sides of the ideological divide) have become too eager to exercise this power in recent decades. A sensible reform might provide one small thumb on the scale in the direction of deference.


These are the goals of the reforms we hope to lay out. But sensible reforms would also have other necessary criteria as well. Any significant change to the way the Supreme Court works will create immediate winners and losers. Given that the Republicans are currently enjoying the benefits of a strongly conservative Supreme Court, we have little reason to think they would support efforts to meaningful reform the Court or to rein in its worst impulses. For this reason, we think any reform proposal would have to be capable of implementation via statute rather than constitutional amendment in the event that Democrats are able to capture control of Con- gress and the White House at some point in the future. That limitation is significant but nec- essary; given the polarization of society, the stakes of control over the Supreme Court, and the relative distribution of partisan affiliation within society and geographically across the United States, we think that it is close to impossible that any kind of constitutional amendment chang- ing the way the Supreme Court works could pass in the near term.35


Related to that point, we also think it is important that any statutory reform proposal be plausibly constitutional. Not obviously constitutional, or undebatably so, but at least plausibly. Indeed, for the right kind of reform we are willing to accept constitutional arguments that are less than bulletproof. There is, to be sure, a significant risk that the Supreme Court itself would strike down reform on constitutional grounds, and for that reason one might think only con- stitutionally rock-solid reform proposals should be put forward. The conservative majority on the Court would, no doubt, be motivated to reject any reform that would have the effect of reducing its power. Yet this argument ignores the fact that if the Supreme Court is called upon to consider the constitutionality of reform along these lines, some much more serious threats to the Court’s power and legitimacy—jurisdiction stripping, court-packing, and perhaps even outright defiance by the political branches—will be lurking in the background if the Court re- jects more moderate reform. Such threats may be merely implicit, but they could be explicit— such as a reform statute with a severability clause that included a provision stating that the Court would be packed with five new justices, or its jurisdiction would be removed, in the event that the reform proposal were struck down. Under such circumstances, we think the Court might blink before overturning a reform package based on less-than-certain constitutional rea- soning.







35 We recognize that even a statutory proposal may be difficult to pass politically, but it remains far easier than a constitutional amendment. For a discussion of the political constraints on Supreme Court reform, see Adrian Vermeule, Political Constraints on Supreme Court Reform, 90 MINN. L. REV. 1154 (2006).




The final criterion for reform is that the resulting system must be potentially stable—it must be an arrangement that both political parties could live with going forward. This may seem quite inconsistent with what we have said thus far, given that we have suggested that reform would need to be enacted via statute, largely along party lines, and using fairly aggressive tactics in order to dissuade the Supreme Court from declaring it unconstitutional. How could such a reform lead to any kind of stable equilibrium going forward?


Here, we think it important to distinguish between means and ends. As David Pozen has explained, it is possible to imagine “hardball” tactics (defined as conduct that “violates or strains constitutional conventions for partisan ends” or that “attempts to shift settled understandings of the Constitution in an unusually aggressive or self-entrenching manner”) in order to accom- plish what he calls anti-hardball goals.36 “Anti-hardball policies” in Pozen’s account “forestall or foreclose tit-for-tat cycles and lower the temperature of political disputes.”37  Even if aggres- sive hardball tactics are used as means, it is at least possible to imagine them creating a system that has no obvious ideological valence going forward and which both sides could live with. What is necessary, though, is to design reforms that reflect “‘good-government’ rules that both sides would prefer to adopt, if they had to write the rules under a veil of ignorance.”38 Properly designed reforms could satisfy this criterion, even if they were initially  adopted by being rammed through on party lines using hardball tactics.


  1. How Existing Proposals Shape Up


On the criteria identified above, many of the proposals that commentators and scholars have offered do fall short. A number of commentators, for example, have proposed that Su- preme Court justices serve an 18-year term instead of a lifetime term.39 First proposed in a stu- dent note ,40 the plan is most famously associated with Roger Cramton and Paul Carrington.41

Under this plan, each president would get two appointments to the Court during a four-year term. This plan would make  appointments would be predictable, removing the pressure to stack the Court with younger and younger justices.






36   David Pozen, Hardball and/as Anti-Hardball, BALKINIZATION, Oct. 11, 2018,

37 Id.

38 Id.

39 See Roger C. Cramton & Paul D. Carrington, The Supreme Court Renewal Act: A Return to Basic Principles, in REFORMING THE COURT: TERM LIMITS FOR SUPREME COURT JUSTICES 467 (Roger C. Cramton & Paul D. Carrington, ed., 2006); Roger C. Cramton, Reforming the Supreme Court, 95 CALIF. L. REV. 1313 (2007); Steven G. Calabresi & James Lindgren, Term  Limits for the Supreme Court: Life Tenure Reconsidered, 29 HARV. J. L. & PUB. POL’Y 769 (2006); James DiTullio & John Schochet, Note, Saving  this Honorable Court: A Proposal  to Re- place Life Tenure on the Supreme Court with Staggered, Nonrenewable Eighteen-Year Terms,  90 VA. L. REV. 1093 (2004); see also Linda Greenhouse, New Focus on the Effects of Life Tenure, N.Y. TIMES, Sept. 10, 2007.

40 See DiTullio & Schochet, supra note _.

41 See Cramton & Carrington, supra note _.




This is a well-intentioned proposal. But in our view it does not satisfy the criteria for re- form—most importantly, because it is unlikely to depoliticize the Court. Indeed, if anything, it will make the politicization of the Court even worse.


An initial problem, though, is that it may not be possible to implement the 18-year proposal via statute alone. Constitutional scholars—and even some who wish to eliminate lifetime ten- ure—have argued that the clause in Article III that gives justices a term for “good behavior” indicates a lifetime appointment.42 While we can imagine arguments that “good behavior” can coexist with a term-of-years appointment, we think the argument that a statute implementing this proposal is constitutional rests on comparatively weak grounds. For these reasons, the orig- inal proponents of the plan, James DiTullio and John Schochet, explicitly framed their proposal as requiring constitutional amendment.43 Such an amendment would almost certainly need sig- nificant Republican support, which seems highly unlikely for the foreseeable future.


Cramton and Carrington, though, offer a version of the plan which they contend could be implemented via statute. In their proposal, Congress would pass a statute giving each President one Supreme Court appointment after each federal election. Justices who served longer than

18 years would not lose their commissions, but would instead effectively serve in a senior status role, sitting only when one of the nine-most junior Justices (i.e., those appointed within the last

18 years) was unable to participate in a case.44 This version of the proposal strikes us as more constitutionally plausible than a true term-limit requirement implemented by statute, though it is far from certain that it would pass muster.


Constitutional issues aside, however, the deeper problem is that the proposal would likely make the Supreme Court more  political, rather than less so. The proposal guarantees that the Supreme Court will be a campaign issue in every single presidential election because each pres- ident would get to shape the Court with two nominees. During the appointment process, movement activists on both sides would still jockey to make sure only the purest ideologues would be appointed. And once on the bench, the justices themselves might actually become more political: A term-limited justice might see the Court as the perfect jumping off point for a presidential run, decide cases in hopes of retiring into a lucrative lobbying gig, or play to the public to secure a future on Fox News or MSNBC.45 This is a major, severely underappreciated, and we think disqualifying drawback to the 18-year proposal.







42 See Saikrishna Prakash & Stephen D. Smith, How to Remove a Federal Judge, 116 YALE L.J. 72 (2006); David R. Stras & Ryan W. Scott, Retaining Life Tenure: The  Case for a Golden  Parachute, 83 WASH. U. L.Q. 1397 (2005); Calabresi & Lindgren, supra note _.

43 See DiTullio & Schochet, supra note _, at 1097 (“Ending life tenure would require a constitutional amend- ment.”).

44 See Cramton & Carrington, supra note _, at 471.

45 Cramton and Carrington’s proposal would not solve this problem, because even if effectively term-limited justices were entitled to remain on the Court, they might well choose not to.




Another proposal, from Professor Tracey George and Dean Chris Guthrie, is to expand the Supreme Court to the size of one of the courts of appeals, and then have this larger Supreme Court to hear cases in panels, with the opportunity for en banc review.46 Professors George and Guthrie’s stated aim is to solve the problem of the Court hearing too few cases, and they see this reform as a way to expand the Court’s docket.47 This proposal could potentially tamp down the politicization of the Court, in that the Court would have many more justices and panels making initial decisions would be randomly selected. But Court appointments—particularly in the transition period to this system—would remain be highly politicized.


The most prominent proposal is court-packing—expanding the size of the Court to include two or more new justices. One of the virtues of this proposal is that it is almost certainly capable of implementation by statute without any constitutional problem, as the size of the Supreme Court has always been set by statute and has varied over American history. It has, however, remained at nine members since 1870,48  and President Franklin Roosevelt’s failed attempt to expand the Court’s size in the 1930s has led many to conclude that the Court’s size is now a settled Constitutional norm. For example, Professor Richard Primus (responding to a proposal for Republicans to pack the lower courts for nakedly political reasons49) argues that such a meas- ure are “not constitutional in the small-c sense of the term” because they “depart[] from long- settled norms and understandings about how American government is conducted.”50


Yet from another perspective, court packing could be the appropriate response by Demo- crats for Republicans’ violation of norms. Professor Michael Klarman recently argued the case for court-packing, stressing not only the circumstances of the recent Republican Supreme Court nominations, but also the fact that Republicans are systemically “abrogat[ing] a basic principle of democracy—when you lose in politics, sometimes you have to just admit defeat.” 51

Instead, Klarman argues, they are changing the rules of politics—from voter suppression to re- stricting the powers of Democratic governors.52 Klarman thus contends that Democrats should





46 Tracey E. George & Chris Guthrie, Remaking the United States Supreme Court in the Courts’ of Appeals Im- age, 58 DUKE L.J. 1439 (2009).

47 Id. at 1442.

48 See, e.g., RICHARD H. FALLON, JR. ET AL., HART AND WECHSLER’S THE FEDERAL COURTS AND THE FED- ERAL SYSTEM 318 (7th ed. 2015); AKHIL REED AMAR, AMERICA’S UNWRITTEN CONSTITUTION: THE PRECE- DENTS AND PRINCIPLES WE LIVE BY 354-55 (2012) (noting that changing the size is constitutional if done for good government reasons). Some, however, are not convinced. Professors Bradley and Siegel, for example, suggest that court-packing might violate a norm derived from historical practice. See Bradley & Siegel, supra note 26; Grove, supra note 26. Others think that court-packing violates a separation of powers convention. David E. Pozen, Self-Help and the Separation of Powers, 124 YALE L.J. 2, 34 (2014).

49 See Steven G. Calabresi & Shams Hirji, Proposed Judgeship Bill,

50  Richard Primus, Rulebooks,  Playgrounds, and  Endgames: A Constitutional Analysis of the Calabresi-Hirji Judgeship Proposal,  HARV. L. REV. BLOG, Nov. 24, 2017, grounds-and-endgames-a-constitutional-analysis-of-the-calabresi-hirji-judgeship-proposal/.

51 Klarman, supra note 23.

52 Id.




not “unilaterally disarm,” but instead need to pack the courts in order to restore and protect the basic infrastructure of democracy.


At first glance, court-packing plans appear to be the kind of reform that might lead to greater politicization and delegitimization of the Court. If Democrats pack the Court, the ar- gument goes, when the Republicans are next in power, they will return the favor and pack the Court further in a tit-for-tat response. On this approach, court-packing both makes politics worse and is an unstable reform. Yet as Mark Tushnet has observed, “there are numerous diffi- culties with this informal game-theoretic argument” given the difficulty of determining what the different “rounds” of the game are, and given the fact that “when rounds of play are sepa- rated by long periods of time, the actual people who play against each other can be quite differ- ent.”53


More concretely, we can imagine conditions  under which court-packing could lead to a stable equilibrium that does not invite an ever-escalating cycle of political retaliation. Through- out American history, there have been moments in which major upheaval have realigned poli- tics (and constitutional politics) to a new equilibrium.54 If Democrats engaged in court-packing and were able to hold power for long enough to implement policies to revive basic principles of democracy (e.g. voter access and anti-gerrymandering reforms), it is possible that this polarized era would give way and a new, more progressive equilibrium might emerge.


That said, it is certainly conceivable that no such new equilibrium would emerge, and in- stead each party would expand the size of the Court whenever it controlled both the Presidency and both houses of Congress. If court packing produced that result, it would delegitimize the Court but also, most likely, the entire enterprise of law. Thus, while court-packing’s great strength is that it is unquestionably constitutional, it could make end up making our predica- ment worse.




Comprehensive reform is the key to saving the Supreme Court. In this Part, we explain how that might be accomplished. We offer two different plans for reform. In Section A, we propose the “Supreme Court Lottery,” a plan in which the Court would sit in panels selected at random from a large pool of potential justices who would also serve as judges on the U.S. Courts of Appeals. In Section B, we propose the “Balanced Bench,” in which the Supreme Court would be composed of an equal number of Democratic- and Republican-selected Justices, plus addi- tional justices drawn from the circuit courts whom the “partisan” justices would have to agree





53 Mark Tushnet, The Pirate’s Code: Constitutional Conventions in U.S. Constitutional Law, 45 PEPP. L. REV.

481, 500 (2018).

54 The classic account is BRUCE ACKERMAN, WE THE PEOPLE: FOUNDATIONS (1991); BRUCE ACKERMAN, WE THE PEOPLE: TRANSFORMATIONS (2000). Drawing on his idea of constitutional time, Jack Balkin has argued that President Trump represents the end of one era of politics and that a new era could be on the horizon. Jack Balkin, What  Kind  of President Will  Trump Become, Part II — Donald  Trump and  the Politics  of Disjunction, BALKINIZATION, Nov. 14, 2016, trump.html.




on unanimously. While neither proposal eliminates every problem we have identified, either would be a major improvement over the status quo.


  1. The Supreme Court Lottery


  1. The Plan and its Benefits


We call our first proposal the Supreme Court Lottery. Under this reform, every judge on the federal courts of appeals would also be appointed as an Associate Justice of the Supreme Court. The Supreme Court would hear cases, but through a panel of nine justices selected, at random, from all the justices. Once selected, the justices would hear cases for only two weeks, before another set of judges would replace them.55 In addition, a two-thirds supermajority of the Court, rather than a simple majority, would be needed to overturn a federal statute.56


This approach would have significant benefits. First, it would de-politicize the appoint- ments process. The lottery approach takes the Supreme Court out of the electoral and political realm. It would also mean that Court appointments would no longer be as significant an issue during elections, and that the future of the Court and of American public policy would no longer depend on random occurrences, such as the unexpected death of a justice. This would also free up presidents and congresses to do the work of governing, instead of occasionally put- ting that work aside for Court appointments.


This approach would, however, make appointments to the federal court of appeals more significant than they currently are, as these judges would constitute the minor leagues for the Supreme Court. But we think this concern is relatively minor, and nonetheless has desirable side-effects. Appointments to the federal courts of appeal are already polarized, with Republi- cans in the Senate currently working at high-speed to fill vacancies with young, ideological ap- pointees.57 This is precisely because they understand the importance of the courts of appeals.





55 Our proposal is similar to that offered in John O. McGinnis, Justice without Justices, 16 CONST. COMMENT.

541 (1999). McGinnis calls his proposal “Supreme Court riding,” and it differs from ours in a few important ways. First, McGinnis imagines abolishing the office of Supreme Court justice overall (a proposal that requires consti- tution amendment). We instead propose expanding the number of associate justices, a reform that we think is constitutional because it is simply deciding the size of the Court. Second, McGinnis suggests that the term of ser- vice for “riding” be six months. We propose two weeks, a far shorter term that would intensify many of the argu- ments made here, as compared with a six-month rotation.

56 This last change would also require establishing that if a lower court strikes down a federal statute, the Su- preme Court would have to hear the case, and it would take a two-thirds vote for the statute to be deemed uncon- stitutional, regardless of the lower court’s decision. This would solve the problem of a federal court of appeal strik- ing down a statute and the Supreme Court needing only a bare majority to affirm that ruling, when it would oth- erwise need two-thirds to overturn the statute itself. Without this change, the proposal would perversely aggran- dize the power of lower courts. For a discussion, see Jed Handelsman Shugerman, A Six-Three Rule: Reviving Con- sensus and Deference at the Supreme Court, 37 GA. L. REV. 893, 957 (2003).

57 See Charlie Savage, Trump is Rapidly  Reshaping the Judiciary. Here’s How., N.Y. TIMES, Nov. 11, 2017. Cf. Joseph Fiskin & David E. Pozen, Asymmetric Constitutional Hardball, 118 COLUM. L. REV. 915 (2018) (noting that polarization is largely a Republican phenomenon on issues of constitutional hardball).




Both sides, we expect, would engage in this behavior. At the same time, however, the lower sa- lience and higher volume of these appointments means they are less likely to become central to public debate. This would be a positive development, as it would make the courts less of a po- litical football in elections and also prevent the creation of cults of personality around the jus- tices. Instead, the Court would be what it should be—a relatively anonymous group of skilled, thoughtful judges.58


Second, we expect this approach would also decrease the ideological partisanship of Court decisions and the idiosyncratic nature of Court decisions. No judge would be able to advance an ideological agenda over decades of service, and no judge would be the single swing voter over a period of years, such that the lion’s share of advocacy aims at that individual’s idiosyncratic views. In addition, it would be very difficult for any judge to be too much a judicial activist on any given case because the next panel—arriving as soon as two weeks later—might have a dif- ferent composition and take a different tack. This would push judges towards more minimalis- tic, narrow, deferential decisions.59 While there is some chance that a random selection of court of appeals judges might lead to radical swings in judicial review, the two-third supermajority requirement should mitigate this concern.


Cases would also be chosen behind a veil of ignorance. While serving their two weeks, the justices would consider petitions for Supreme Court review. But with such short terms of ser- vice, the justices could not pick cases with an agenda in mind; another slate of justices would hear the cases they select.60 Activist lawyers would also not be able to game the system, bringing legal arguments and cases based on their prediction of which way the Court is likely to decide. In the run of cases, the court’s decisions would likely be far less aggressive in reversing Congress and far more tightly linked to precedent.


  1. The Constitutionality of the Supreme Court Lottery


We think the Supreme Court Lottery could be implemented by statute, without a consti- tutional amendment. It is generally uncontested that Congress has the power to change the size of the Supreme Court and to set the basic procedures governing the Court.61  The Court has






58 McGinnis, supra note 55, at 542 (“Vested for life with the awesome power to make final decisions with wide- ranging consequences for the nation, Supreme Court Justices generally cannot help but come to see themselves as statesmen rather than as humble arbitrators of legal disputes.”).

59 See id. at 544 (“Supreme Court riders would have been less able to instantiate their political vision and would therefore be more likely to follow precedent. Moreover, because the riders would have come from inferior courts, which operate under the threat of reversal, they would have had more practice in following precedent.”).

60 For a brief discussion, noting this possibility, see Adrian Vermeule, Veil of Ignorance Rules in Constitutional

Law, 111 YALE L.J. 399, 424 (2001); see also McGinnis, supra note 55, at 545.

61 See, e.g., Michael Stokes Paulsen, Checking the Court, 10 N.Y.U. J. L. & LIBERTY 18, 64 (2016) (“Nothing in the Constitution specifies the size of the membership of the Supreme Court. … The size and details of the Supreme Court’s membership are up to Congress.”). Indeed, the proof of the point is that the most notable arguments against altering the size of the Court state that there is “a strong norm” or “convention” against reforms to “pack




gotten bigger and smaller over the centuries, and Congress sets many basic provisions of the Court’s operations through statutes.62 Statutes give the Chief Justice various powers, required justices to “ride circuit” for more than a century, and organize the Court in a variety of other ways.63


This proposal works from that constitutional baseline. The proposal formally expands the size of the court to some 180 judges, and it then provides for how the Court would hear cases.64

The president would still nominate every justice, and the Senate would confirm them. The jus- tices would serve for life, assuming good behavior, as is current practice. The current Supreme Court Justices would not lose their positions or their lifetime appointments; they would simply enter the lottery, like all the other associate justices.65 If they wanted, they could also be ap- pointed to the federal courts of appeal, as the other associate justices would be. And the current Chief Justice would retain his lifetime position and additional statutory duties, including his constitutionally-prescribed role presiding over an impeachment trial of the President in the Senate.66


Still, there are a variety of potential constitutional issues that the proposal raises. While we think we have solid arguments to possible objections, we stress again that our goal is only plau- sibility. There are reasonable constitutional counterarguments to our proposed reforms, espe- cially for those working within the highly formalistic methodology favored by the current con- servative majority. Given, however, that these reforms would likely be litigated against a com- plex political backdrop of popular sentiment directed against the Court, and the threat of more radical reform, we think slam-dunk constitutional arguments are not strictly essential.


(a) Dual Appointments


Some might argue that it is unconstitutional for judges to effectively have two appoint- ments—as a federal court of appeals judge and as an associate justice on the Supreme Court. Article III of the Constitution, on this argument, contemplates the existence of a Supreme Court and additionally, inferior courts. In addition, the appointments clause recognizes that the President can appoint justices of the Supreme Court, treating that as a distinct position from other, inferior, appointments.





the court” by changing the size, not that any change is manifestly unconstitutional. See Tara Leigh Grove, The

Origins (and Fragility) of Judicial Independence, 71 VAND. L. REV. 465, 505 (2018).

62 See Paulsen, supra note 61, at 64 (discussing the size changes of the Court).

63 See, e.g., 50 U.S.C. 1803 (vesting the Chief Justice with authority to designate members of the FISA Court); Judiciary Act of 1789, 1 Stat. 73 (providing for circuit riding).

64 There are 179 authorized federal court of appeals judgeships. See Administrative Office of the U.S. Courts, Judicial Vacancies,

65 Note that this proposal does not run afoul of arguments that the Constitution mandates life-tenure for federal judges. For a discussion of Article III’s good behavior clause, see Saikrishna Prakash & Stephen D. Smith, How to Remove a Federal Judge, 116 YALE L.J. 72 (2006).

66 U.S. Const. Art. I, sec. 3.




This argument is not persuasive. Unlike other proposals that do away with the Court, jus- tices in the Supreme Court Lottery would be appointed and confirmed to their position on the Supreme Court, in full accordance with the terms of the Appointments Clause.67  More im- portantly, the text of the Constitution does not have any bar on judges serving in two judicial positions, or two-commissioned positions of any kind, at the same time. In fact, the Constitu- tion is naturally read to allow it. Article I specifically bans members of Congress from serving in another role under the Constitution.68  Thus, as Professors Steven Calabresi and Joan Larsen have noted, “the Constitution contains an express legislative Incompatibility clause but no comparable provision exists to bar joint service in the judicial and executive departments.”69

The framers of the Constitution understood the possibility of conflicts arising from holding multiple posts, accounted for it in one part of the Constitution, but chose not to provide such a bar for justices on the Supreme Court.


In addition, historical and contemporary practice suggests that judges can have multiple roles at once. Foremost, the Judiciary Act of 1789 created federal circuit courts, but it did not create circuit judgeships. Instead, it required Supreme Court justices to “ride circuit,” acting as judges on the nascent federal courts.70 The first Congress thus directed Supreme Court justices to effectively serve on two courts at once. This practice was upheld in the 1803 case of Stuart v. Laird, even though the justices had not been separately appointed to the lower federal courts, and it persisted throughout the Nineteenth century.71












67 For a discussion, see Steven G. Calabresi & James Lindgren, Term Limits for the Supreme Court: Life Tenure

Reconsidered, 29 HARV. J. L. & PUB. POL’Y 769, 859-63 (2006)

68 U.S. CONST. art. I, sec. 6 (“No Person holding any Office under the United States, shall be a Member of either House during his Continuance in Office.”). There are, in fact, two other similar clauses. Article I, Section 9 prohibits holding “any Office” while also “accept[ing] any [other] office” from foreign states, and Article II, Sec- tion 1 prohibits “Senator[s] or Representative[s], or Person[s] holding an Office of Trust or Profit… [from being] appointed an Elector.” The omission in Article III is thus particularly noteable.

69 Steven G. Calabresi & Joan L. Larsen, One Person, One Office: Separation of Powers or Separation of Personnel,

79 CORNELL L. REV. 1045, 1122 (1994). The founding generation was also aware of this omission. The Virginia Ratifying Convention urged the First Congress to adopt an amendment stating that “The Judges  of the federal Court shall be incapable of holding any other Office, or of receiving the Profits of any other Office, or Emolument under the United States or any of them.” It was not adopted. Id. at 1125.

70 Judiciary Act of 1789 (“[T]he before mentioned districts… shall be divided into three circuits, and… that there shall be held annually in each district of said circuits, two courts, which shall be called Circuit Courts, and shall consist of any two justices of the Supreme Court, and the district judge of such districts, any two of whom shall constitute a quorum.”). See generally Joshua Glick, Comment: On the Road: The Supreme Court and the His- tory of Circuit Riding, 24 Cardozo L. Rev. 1753 (2003).

71 Stuart v. Laird, 5 U.S. 299 (1803).




In addition, some judges have had multiple commissions simultaneously. Chief Justice John

Marshall was, for a time, simultaneously commissioned as Secretary of State and Chief Justice.72

Judge Claria Horn Boom currently serves as a federal district judge for both the Eastern and Western Districts of Kentucky.73 Justices on the Supreme Court have also taken on additional roles, apparently without concern. Chief Justice John Jay was dispatched to negotiate the Jay Treaty in 1795, while still serving on the Court. Justice Robert Jackson took up the role as Chief Prosecutor at Nuremberg after World War II. Chief Justice Earl Warren chaired the commis- sion tasked with investigating the assassination of President Kennedy.


Judges also serve on separately constituted courts and commissions from those that they were initially confirmed. Some federal district court judges serve a seven-year term on the For- eign Intelligence Surveilliance Court, while simultaneously fulfilling their district court du- ties.74 Judges serve on the United States Sentencing Commission, a practice upheld by the Su- preme Court.75 And, as discussed in more detail below, judges and justices sit by designation on inferior courts, lateral courts (e.g. a different circuit or district), and superior courts.76 While each of these examples differs from holding a dual appointment, they do suggest that as a matter of historical and contemporary practice, judges have had multiple roles simultaneously and Americans have accepted that variation as at least legitimate, and often desirable.


(b) The Vesting Clause and “One Supreme Court”


Article III of the Constitution vests the judicial power in “one Supreme Court.” Some con- tend that this provision mandates that the Supreme Court be comprised of a single set of per- sons and cannot include a rotating group of justices.77 But this argument suffers from serious




72 The Senate confirmed Marshall’s appointment as Chief Justice on January 27, 1801, yet he did not resign his position as Secretary of State until March 4 of that year. See ALBERT J. BEVERIDGE, THE LIFE OF JOHN MAR- SHALL 558 -59 (1919); 1 CHARLES WARREN, THE SUPREME COURT IN UNITED STATES HISTORY 178, 184 – 85,

200 – 01 (1922).

73  See Vote Number 65, United States Senate, April 10, 2018, tive/LIS/roll_call_lists/roll_call_vote_cfm.cfm?congress=115&session=2&vote=00065.

74 50 U.S.C. §1803.

75 See Mistretta v. United States, 488 U.S. 361 (1989).

76 For example, retired Supreme Court justices sit on the courts of appeals. Cramton, supra note 39, at 1327. For a brief discussion of “upward designation,” see David R. Stras & Ryan W. Scott, Retaining Life Tenure: The Case for a Golden  Parachute, 83 WASH. U. L.Q. 1397, 1418 (2005). For a broad discussion of judges on other courts, see Marin Levy, Visiting Judges, CAL. L. REV. (forthcoming 2019) (draft on file with the authors).

77 See, e.g., STEPHEN M. SHAPIRO ET AL., SUPREME COURT PRACTICE § 1.1 (10th ed. 2013) (arguing that “the fact that the Constitution vests the judicial power ‘in only one Supreme Court . . . does not permit Supreme Court action by committees, panels, or sections” (quoting William J. Brennan, Jr., State Court Decisions and the Supreme Court, 31 PA. BAR. ASS’N Q. 393, 406 (1960) (alteration in original)). The authors cite a letter from Chief Justice Hughes and articles by Justice Harlan, Brennan, and Field to support the idea that the Court cannot hear cases as a panel. Id. They also argue that the rejection of an 1890 proposal for creating panels within the Supreme Court supports this position. Id. But it is not clear why that inference is reasonable. First, inferences from legislative in- action should be disfavored. Second, the 1890 moment was one of radical change in any event. The idea of panels within the Court, with full Court review, had been considered at least as early as 1869, gained the support of a




infirmities. First, Article III’s vesting clause was partly drafted and designed to address a variety of concerns on the balance between federalism and nationalism. The government of the Articles of Confederation did not have a national judiciary; the vesting clause established clearly that the new government would.78 In addition, during the debates at the Constitutional Conven- tion, much of the discussion over the creation of the federal courts was about whether there would be lower federal courts. Some members of the Convention preferred establishing lower federal courts in the Constitution, while others feared that lower federal courts would take power from the states. The compromise was to establish a Supreme Court and permit (but not mandate) Congress to create lower federal courts.79 The drafting history of the vesting clause was tied to this debate more than some theoretical sense of “oneness.”


Moreover, as Professor Klarman has shown, the debate over the Court was tied to the broader question of “enforcing federal supremacy.” The Convention rejected the option of a federal veto over state laws in favor of the Supremacy Clause and creating a Supreme Court.80

In Federalist No. 22, Alexander Hamilton pointed out that one of the core benefits of a single institution—which would still apply if personnel fluctuate—is finality amidst a federal system of multiple courts: “To avoid the confusion which would unavoidably result from the contra- dictory decisions of a number of independent judicatories, all nations have found it necessary to establish one court paramount to the rest, possessing a general superintendence, and author- ized to settle and declare in the last resort a uniform rule of civil justice.”81


Second, the vesting clause argument mistakenly assumes that a singular institution, which the Supreme Court is—and would continue to be under this proposal—cannot be composed of multiple people in rotation. But there is a difference between having a single institution, which the vesting clause clearly indicates, and having that institution with a fixed rather than variable membership. Textually, the clause itself does not specify that the number of justices, nor that the membership be fixed rather than rotational. When combined with Congress’s power in the necessary and proper clause to “carry into execution” “any other power vested” anywhere else in the government, Congress has authority under the Article III vesting clause to





number of prominent commentators and elected officials, and was one prominent option on the table. The other radical option, which was ultimately chosen, was the creation of intermediate courts and brought the eventual end of the century-long tradition of circuit riding. For a brief discussion of this proposal, see Tracey E. George & Chris Guthrie, “The Threes”: Re-Imagining Supreme Court Decisionmaking, 61 VAND. L. REV. 1854 (2008). On circuit riding, see Glick, supra note 70.

78  See THE FEDERALIST NO. 22 (Alexander Hamilton) (“A circumstances which crowns the defects of the

Confederation remains yet to be mentioned, the want of a judiciary power.”).


164–67 (2016).

80 Id. at 164.

81 THE FEDERALIST NO. 22 (Alexander Hamilton). See also THE FEDERALIST NO. 80 (“The mere necessity of uniformity in the interpretation of the national laws, decides the question. Thirteen independent courts of final jurisdiction over the same causes, arising upon the same laws, is a hydra in government, from which nothing but contradiction and confusion can proceed.”)




make rules for the creation, composition, and terms of the judiciary—including the Supreme Court.82   This includes deciding that the Court’s membership should rotate. Indeed, more broadly, singular institutions—include the current Court—always have a fluctuating member- ship. At present, justices recuse themselves from cases, quorum requirements contemplate that fewer than a full complement of justices will hear cases, and intertemporally, the Court as an institution changes its personnel with regularity. Institutions can be singular, even if their per- sonnel fluctuate.


(c)  Supermajority Voting Requirements


There are also a number of plausible constitutional challenges to a supermajority voting requirement for striking down federal statutes. One set of arguments is that Article III implic- itly either requires majority rule or gives the Court the power to decide how to resolve its own cases. Both suffer from an absence of textual support.83 A second set of arguments are structural: that supermajority rules would aggrandize congressional power or effectively mean that Con- gress is determining the outcome of cases.84 These arguments are also unmoored from any tex- tual provisions, and are effectively a version of free-form structural constitutional arguments.85

It is also worth noting that whatever normative strength such arguments have, there are also prominent constitutional thinkers who have questioned the case for majority decisions at the Supreme Court on normative grounds and noted that values like expertise, respect for consti- tutional structure, and fairness cut in favor of supermajority requirements.86


The constitutional case for setting supermajority requirements starts from the premise that Congress has the power to structure the judiciary. The source of this power is a combination of the necessary and proper clause, which gives Congress the power to “make all laws necessary and proper for carrying into execution … all other powers vested in this Constitution,”87  and the







82 The classic article on the general claim of the scope of the necessary and proper clause is William W. Van Alstyne, The Role of Congress in Determining Incidental Powers  of the President and of the Federal  Courts:  A Com- ment on the Horizontal Effect of the Sweeping Clause, 40 L. & CONTEMPORARY PROBS. 102 (Spring 1976). For more recent takes, see John F. Manning, Foreword: The  Means  of Constitutional Power,  128 HARV. L. REV. 1 (2014); John Mikhail, The Necessary and Proper Clauses, 102 GEO. L.J. 1045 (2014).

83 For an overview of these challenges, see Evan H. Caminker, Thayerian Deference to Congress and Supreme Court  Supermajority Rules: Lessons from the Past, 78 YALE L.J. 73, 77 n.12 (2003). There might be an argument that Article I gives Congress the power to structure its own rules and operations and this approach should be applied to Article III as well. But the opposite argument—that the Constitution contemplates such a provision but excludes it from Article III—seems at least equally persuasive.

84 Id. at 77 n. 12.

85 See Manning, supra note 82 (discussing the illegitimacy of free-form structural constitutional arguments).

86 Jeremy Waldron, Five to Four: Why do Bare Majorities Rule on Courts?,  123 YALE L.J. 1692 (2014); Jed Handelsman Shugerman, A Six-Three Rule: Reviving Consensus and Deference at the Supreme Court,  37 GA. L. REV. 893 (2003).

87 U.S. CONST., Art. I, sec. 8.




Exceptions Clause, which states that the Supreme Court has jurisdiction “with such Excep- tions, and under such Regulations as the Congress shall make.”88 From the Judiciary Act of 1789 onward, Congress has exercised these power. The first Congress not only established the size of the Supreme Court, but also required that “any four of [the justices] shall be a quorum.”89 In terms of potentially dictating judicial outcomes, a supermajority requirement is not so different from a quorum requirement. Both are restrictions on how many justices are needed for a judicial determination to be binding.


Supermajority requirements themselves also have a long history within debates over reform- ing Supreme Court. They were proposed at least as early as the 1820s, with another sixty pro- posals being offered between that time and the early 1980s.90 And some states, including Ne- braska and North Dakota, have adopted supermajority requirements.91 The fact that these pro- visions have been discussed over almost two centuries certainly does not establish their consti- tutionality, but it is worth noting that many have thought such a proposal would be constitu- tional if adopted.


(d)  Historical Practice


Commentators and scholars have also suggested that the long historical practice of having a Supreme Court constituted by a single set of justices prevents change.92 Both the Supreme Court and commentators have recognized the historical practice can inform constitutional meaning.93 This is an important argument, as historical practice is one way of fleshing out con- stitutional meaning. At the same time, however, taking historical practice too far prevents de- mocracy experimentation. Adherents to the historical practice school can fall into the trap of arguing that Congress always legislates to its maximal authorities and that it always explores and implements every possible strategy.94  Under our constitutional system, Congress has been granted significant powers under Article I—and there is no provision anywhere in the Consti- tution that suggests that Congress loses those powers if it chooses not to exercise them for a period of time. Indeed, the idea that Congress’s Article I powers disappear if Congress chooses not to use them flies in the face of Article I’s vesting clause and the Separation of Powers, which give to Congress legislative powers, whether or not they are exercised at any given moment.







88 U.S. CONST. Art III, sec. 2. For an extensive discussion making this argument, see Shugerman, supra note

86, at 972–81.

89 Judiciary Act of 1789, 1 Stat. 73.

90 Caminker, supra note 83, at 88.

91 NEB. CONST. art V, § 2; N.D. CONST. art. VI, §4; Caminker, supra note 83, at 91–92.

92 Bradley & Siegel, supra note 26.

93 See Curtis A. Bradley & Trevor W. Morrison, Historical Gloss and the Separation of Powers,  126 HARV. L. REV. 411 (2012); William Baude, Constitutional Liquidation, 71 STAN. L. REV. (forthcoming 2019) (draft on file with the authors).

94 Leah Litman, Debunking Antinovelty, 66 DUKE L.J. 1407 (2017).




  1. The Balanced Bench


  1. The Plan and its Benefits


Our second proposal, the Balanced Bench, looks quite different from the Lottery Court, but in our view it is a different strategy for addressing similar concerns. The proposal has several components. First, the Supreme Court would start with ten justices. Five would be affiliated with the Democratic Party, and five would be affiliated with the Republican Party. These ten justices would then select five additional justices chosen from current circuit court (or, possibly, district court) judges. The catch? The ten partisan-affiliated justices would need to select the additional five justices unanimously (or at the very least, by a supermajority requirement on each side).95 These additional justices would be chosen two years in advance, to serve for one year. And if the justices failed to agree on a slate of additional colleagues, the Supreme Court would lack a quorum and could not hear any cases for that year.


The idea behind this proposal is that it provides a mechanism to restore something im- portant that we fear has been lost: the notion that Supreme Court justices are deciding ques- tions of law, in ways that don’t invariably line up with their political preferences in the biggest cases. That was once true—even during periods of the most serious political conflict over the Supreme Court, the Justices were not strictly following party lines. As noted above,96 during the infamous Court-packing drama in the 1930s, the Justices were closely divided along ideological lines but not party lines.


Today, however, it seems like a quaint notion that Presidents would ever choose Supreme Court justices who would vote against their party’s interests in big cases. The Republicans made this mistake (if it is a mistake) in recent decades, which led them to vow to appoint “no more Souters.”97 Democrats, despite having far fewer opportunities to appoint justices in recent dec- ades, have done a reasonably good job of identifying ideologically reliable nominees. Given that both sides seem to realize the stakes of Supreme Court nominations, it is hard to imagine that there will be many more justices like Justice Kennedy, who would sometimes vote against “type” in the biggest cases.


This proposal brings back the possibility of a Supreme Court that isn’t strictly partisan. The idea would be that the “partisan” justices would agree on colleagues who have a reputation for fairness, independence, and centrism, and who have views that do not strictly track partisan affiliation. In short, the kind of judges who have essentially zero chance of being appointed to the U.S. Supreme Court today. The partisan justices would pick such colleagues not for public- regarding reasons, but out of self-interest: assuming that those justices want their own views to prevail on the Court, they would have an incentive to veto committed partisans on the other




95  The reason to opt for a supermajority requirement rather than unanimity is to avoid the risk of a single holdout who might prefer that the Supreme Court not operate. A supermajority of each side, rather than a ordi- nary majority, is necessary, though, to help ensure that the additional justices chosen are not ideological extremists.

96 See supra Part I.

97  See, e.g., Editorial, No More  Souters,  WALL ST. J., July 19, 2005, cles/SB112173866457289093.




side. But each side might be willing to compromise (and, really, to gamble) on other judges who seem open-minded and persuadable.


To be sure, they might not pick independent-minded justices for all five of the slots. Per- haps the two sides would compromise on a couple of more ideologically reliable justices. But requiring the partisan justices to pick an odd number of justices means that, at the very least, they would likely have to pick one moderate and ideologically unpredictable justice whose vote could break ties. Our hope, though, is that they would pick more than one.


The partisan justices would select their colleagues with two years of lead time. This would reduce the risk of the justices brokering deals during the selection process to pick colleagues based on their expectations about individual cases or issues. For example, knowing that a gay marriage case was on the docket, perhaps the Democratic justices would accept a generally con- servative judge who had a reputation for voting in more liberal directions on important social issues (say, like Justice Kennedy). Even assuming the partisan justices had such granular infor- mation about their potential colleagues votes, we think delaying the start date of the new jus- tices would reduce this risk.


Once chosen, the independent justices would serve for terms of one year. Their terms would not be renewable. Although the prospect of renewal might serve as a powerful incentive for centrism, we think the threat of non-renewal would undermine the justices independence and damage the internal dynamics of Supreme Court decisionmaking. Morever, we think there are good reasons to have some justices with shorter tenures. As discussed above, the modern Court, with its nine life-tenured members, is too dominated by cults of personality (think of the “No- torious RBG”) and too focused on particular justices’ idiosyncratic views (think of the emphasis on “Kennedy briefs” in recent years). Throwing some less well-known, shorter-term justices into the mix would significantly reduce this problem. The short-term justices could also intro- duce a helpful perspective. They would introduce a greater diversity of educational, profes- sional, and geographic backgrounds to the Court and bring in the trenches experience to the bench. To the extent that long-time service on the Supreme Court changes one’s perspective, they would also not be affected by that bias.


Finally, the independent justices—and the explicit partisan-balance requirements—would significantly reduce the stakes of the Supreme Court nominations process. Because each politi- cal party would get a set number of seats, and because additional justices would join the Court no matter what, the fate of issues like abortion would never turn on the consequences of any one confirmation battle. Because the “independent’ justices would be drawn from the lower courts, this proposal would likely exacerbate the politicization of lower-court nominations. But as discussed above, that phenomenon is already happening on its own. Moreover, given the need for some independent-minded justices who could plausibly be chosen by the other justices to join the Supreme Court temporarily, the system might actually provide some incentives to ap- point some moderates rather than ideologues on the lower courts.


In order to produce some of the veil-of-ignorance benefits provided by the first proposal with respect to the case selection process, the Court’s internal processes could be designed to minimize the temporary justices’ ability to pick their own cases. For example, the new justices




could join the Court immediately after the “long conference,” in which the Court votes on a significant number of certiorari petitions that have built up over the summer, by which point a significant number of the Term’s cases would already have been selected.


With luck, a Court designed as we propose would issue rulings in big cases that wouldn’t be predictable based solely on party affiliation. Those rulings would have a greater chance of being seen as legitimate by the public. And thus this plan has a chance of saving the image of the Su- preme Court as an enterprise above politics—and thus of preserving the image of law as an en- terprise distinct from politics.


Given our interest in divorcing the Court from partisan politics, it is a fair question why we would want to explicitly build in partisan affiliation to the selection of justices. Well, first of all, someone has to select the independent justices. And if we could identify some actor in govern- ment would could be reliably trusted to always select justices without regard to partisan affilia- tion, we could just put that person on the Supreme Court. Given our inability to identify such a person, however, the best solution is to try to design a system that creates incentives for highly ideological and/or partisan government actors to select for non-partisan justices.98


But there are other arguments for building in some form of partisan balance. Indeed, Pro- fessor Eric Segal has argued for the institution of a Court permanently and evenly divided along partisan and ideological lines.99 He contends that such a Court would produce narrower, more consensus-based decisions, would “reduce the opportunities for five or more Justices to impose rigid ideological agendas over long periods of time,” and would eliminate the problem of the Court’s ideology being driven by random events like deaths or strategically timed retirements.


Indeed, our brief experiment with a Court that was evenly divided along and partisan and ideological lines showed that there was something to Segall’s idea. During the more than a year in which the Court was understaffed, the justices generally strove to reach consensus where possible, often deciding cases on narrower grounds. In fact, October Term 2016—in which the Court was down a Justice for almost the entire Term—displayed the most consensus among the Justices in more than 70 years.100 That said, that experiment also revealed downsides of the arrangement. Where the justices were unable to reach agreement—in the most ideological cases with the highest stakes—the Court was left powerless to make law, and the courts of appeals








98 There are other mechanisms we could implement to accomplish that result, perhaps. For example, a super- supermajority requirement for the confirmation of justices in the Senate—say, a 75-vote requirement—might only permit extremely moderate justices to be confirmed (although it is hard to imagine 75 senators today agreeing on anything important). Our chosen mechanism is different, though, because it directly incorporates the two partisan sides’ preferred decisionmakers directly into the Supreme Court decisionmaking process.

99 Eric Segall, Eight Justices Are Enough

100 See Adam Liptak, A Cautious Supreme Court Sets a Modern Record for Consensus, N.Y. TIMES, June 27, 2017,




effectivey became the Supreme Court.101  For this reason, at the very least, a proposal for a per- manent, equally divided Court would need to be accompanied by a set of other wide-ranging reforms, such as different rules about the consequences of a deadlock on the Court.102


But even if implemented appropriately, an evenly divided Court would not solve one of the most significant problems we see with the state of the Supreme Court: the widespread percep- tion that the Supreme Court is simply one more political institution, where votes in the biggest cases turn on party affiliation. Indeed, adopting explicit partisan balance requirements without making additional changes would only exacerbate this perception. For this reason, having the “partisan” justices select additional justices to join the Court is critical to the proposal’s success.


While having justices choose their colleagues might seem initially strange, this proposal bears a structural resemblance to the way civil arbitration often works. Under many bilateral arbitration agreements, the two sides each get to select one arbitrator. The two party-chosen arbitrators then agree collectively on a third, neutral arbitrator. Indeed, such provisions date back to at least the late Eighteenth Century.103   Their continued and widespread use likely re- flects the view that this method is a good way of procuring decisionmakers who will be unbiased







101 This happened in United States v. Texas, 136 S. Ct. 2271 (2016) (per curiam), the case addressing the con- stitutionality of President Obama’s Deferred Action for Parents of Americans and Lawful Permanent Residents (DAPA) program, which granted temporary work authorizations to certain undocumented immigrants who were the parents of U.S. Citizens or legal permanent residents. Texas v. United States, 809 F.3d 134, 147 (5th Cir.

2015). There, the Justices’ even split ended up making the Fifth Circuit—which enjoined the program—the Su- preme Court for all practical purposes. A similar result—though one with the opposite ideological valance—ob- tained in Friedrichs v. California Teachers Association. 136 S. Ct. 1083 (per curiam) (2016). That case involved a constitutional challenge to rules requiring non-union members to pay for collective-bargaining expenses by unions designated as the exclusive bargaining representative. The Ninth Circuit, relying on Supreme Court precedent, rejected the challenge. Friedrichs v. California Teachers Ass’n, 13-57095, 2014 WL 10076847 (9th Cir. Nov. 18,

2014). The Supreme Court split 4-4, leaving the Ninth Circuit’s ruling in place. Two years later, once Justice Gorsuch had joined the Court, the Justices overturned precedent and declared such arrangents unconstitutional in Janus v. American Federation of State, County, and Municipal Employees, 138 S.Ct. 2448 (2018).

102 Whereas current law gives lower courts the power to set the status quo—an equally divided Court results in automatic affirmance of the judgment below—one could imagine setting different default rules. For example, law might provide that an equally divided Court has the effect of overturning any judgment that strikes down a judgment overturning an Act of Congress, as a way to build in slightly more deference or that if the Supreme Court cannot reach a supermajority, the Act of Congress stands, notwithstanding the lower court decision. Depending on the design of these rules, a proposal for a permanent eight-member Court might need to be accompanied by limits on the ability of lower courts to issue so-called “nationwide” or “universal” injunctions, as they let individual circuits effectively set the law for the entire country. For a recent discussion of nationwide injunctions, see Samuel L. Bray, Multiple Chancellors: Reforming the National Injunction, 131 HARV. L. REV. 417 (2017).

103 See  Brian Winn & Earl Davis, Arbitration of Reinsurance Disputes:  Is There A Better  Way?, 59 DISPUTE RESOLUTION J. 22, 22 (2004) (noting a 1793 insurance contract with provided that “if any Dispute should arise relating to the Loss on this Policy; it shall be referred to two indifferent Persons, one to be chosen by the Assured, the other by the Assurer, who shall have full Power to adjust the same; but in case they cannot agree, then such two persons shall choose a third; and any two of them agreeing, shall be obligatory to both parties”).




and fair—or, perhaps better stated, decisionmakers who will appear to be unbiased and fair to both sides.


Commercial arbitration has many disanalogies with democratic politics, to be sure. But we think in both cases, there are important reasons to care about designing procedures that the eventual losers can live with. A concern for appearance is an important reason why we think it is necessary to incorporate the partisan justices into the decisionmaking process. Their presence ensures that both sides’ best arguments will be aired and considered. And thus they will help ensure that the losing side feels that the decisionmaking process was fair, even if it did not result in that side’s desired outcome.104


One other objection concerns our proposal’s emphasis on partisan balance. Why should the Court’s design evenly balance the two parties (and thus their respective judicial ideologies) no matter what, instead of allowing for more variability based on the results of the political process? We have a couple responses. First, as a comparative matter, we think our proposal would be an improvement over the status quo. Over the last half century, Democrats have controlled the presidency for 20 out of 50 years, but have appointed only four Justices; Republicans have ap- pointed fourteen (fifteen if you count moving William Rehnquist from an Associate Justice seat to being Chief Justice). That skew has been the result of deaths, strategically timed retire- ments, and other factors. The Balanced Bench would make each party’s power over the Court more regular and predictable.


Our proposal would also, however, not take into consideration a long string of political victories. Democrats controlled the Presidency from 1933 to 1949; during this time, Franklin Roosevelt and Harry Truman appointed thirteen justices to the Court. Under the Balanced Bench, the partisan composition of the Court would have looked exactly the same at the begin- ning of their tenure as it did at the end. Would it really be fair to have an evenly divided Supreme Court after so many years of Democratic control?


A couple points in response. First, regardless of who wins Presidential elections, it is still possible that the country as a whole might be divided along partisan lines in a close-to-even way, making a partisan balance requirement seem fairer than it might appear. Moreover, our pro- posal is focused on appearances, and an evenly divided Court has the best chance of seeming fair to the whole country, regardless of what the actual divisions of party affiliation within the country look like. Finally, to the extent that critics might have concern over this proposal’s











104 Cf. Chiara Giorgetti, Who Decides Who Decides in International Investment Arbitration?, 35 U. PA. J. INT’L L. 431, 443 (2013) (“For the parties [to an arbitration], having a say in deciding their case [by choosing one of the arbitrators] is both appealing and reassuring, and strengthens their support to the entire process.”).




seeming tendency to permit the minority to govern the majority (with the help of the inde- pendent justices) one solution would be to pair this reform with the supermajority voting role considered above.105


  1. Constitutional Objections


As with the Lottery Court, this proposal would be subject to some significant constitu- tional objections. And as with our first proposal, we think there are plausible responses to those objections. Some of the objections overlap with constitutional arguments against the Lottery Court—in particular, the argument that it would be impermissible for judges to serve as both circuit court judges and as Supreme Court justices.106


(a)  Appointments Clause Challenges


The Appointments Clause provides that the President “shall nominate, and by and with the Advice and Consent of the Senate, shall appoint . . . Judges of the supreme Court, and all other Officers of the United States, whose Appointments are not herein otherwise provided for, and which shall be established by Law: but the Congress may by Law vest the Appointment of such inferior Officers, as they think proper, in the President alone, in the Courts of Law, or in the Heads of Departments.”107  Under our proposal, some of the justices would be selected by other  justices,  an arrangement that is permissible for “inferior Officers” but not for so-called “principal” officers—and explicitly not for “judges of the supreme court.” Under a straightfor- ward reading of the Clause, this proposal thus seems unconstitutional.


Not so fast. As it happens, existing law and practice permits significant flexibility with the movement of Article III judges within the federal judiciary. United States District Judges regu- larly sit by designation on circuit courts; circuit judges regularly sit by designation in district







105 The proposal could be modified in order to give the President the power to appoint one additional Justice from his or her own party—a design choice that resembles the way many partisan-balance requirements are imple- mented in independent agencies. Many independent multimember government agencies have partisan-balance requirements governing the selection of their members. While some such agencies are designed to have an even split between the parties, more common are those whose organic statutes “provide for an odd number of members, no more than a bare majority of whom may be from the same political party.” Brian D. Feinstein & Daniel J. Hemel, Partisan Balance with Bite, 118 COLUM. L. REV. 9, 18 (2018).  That  requirement that effectively permits the party that controls the White House to direct more control over policymaking by appointing a majority of agency members. Likewise, one could imagine giving the President the power to appoint a bare majority of the partisan justices from her own party—say, six versus five of the other party. Whether this proposal is a good idea would turn on how it would affect the dynamics of the decision to add additional justices. That modification might reduce the benefits of our proposal if it turned out that the majority was able to bully the minority into appointing more new justices. If, however, the unanimity requirement meant that the independent justices remained truly open-minded and independent, the change would not make a tremendous difference because the independent justices could cancel out the vote of the additional partisan justice.

106 See supra § II.A.2(a).

107 U.S. CONST. art. II, § 2, cl. 2.




courts or other circuits;108  and retired Supreme Court justices regularly sit by designation on courts of appeals. (Justice Souter, for example, regularly sits with the First Circuit, on which he briefly served as a judge before joining the Supreme Court). When judges sit by designation on different Article III courts, they do not get newly nominated by the President and confirmed by the Senate. Instead, they are designated by the Chief Judge of the circuit in which they are visiting, or in some instances the Chief Justice or a Circuit Justice.109 Their initial President- and-Senate appointment seems to be doing all the relevant work.110


Our proposal functions similarly, letting Supreme Court justices invite lower court judges to sit with them for limited periods. If there is a problem with our proposal, then, there also are serious problems with these widespread practices in the lower courts. Some have, to be sure, criticized the status quo. Among somewhat related lines, David Stras and Ryan Scott, for exam- ple, argue that senior judges—who regularly sit by designation on other courts—violate the Ap- pointments Clause, and must instead be separately appointed and confirmed to the distinct office of “senior judge.”111  Such arguments seem to have, thus far, fallen on deaf ears in both the judiciary and Congress.


There is even precedent for a court being entirely constituted by judges chosen by a Su- preme Court justice. Under the Foreign Intelligence Surveillance Act of 1978, the Chief Justice of the United States designates “11 district court judges from at least seven of the United States judicial circuits of whom no fewer than 3 shall reside within 20 miles of the District of Colum- bia who shall constitute a court which shall have jurisdiction to hear applications for and grant orders approving electronic surveillance anywhere within the United States.”112 The judges of





108 See Marin K. Levy, Judging Justice on Appeal, 123 YALE L.J. 2385, 2415 (2014) (noting the regularity of the participation of visiting judges in the courts of appeals). For an in-depth analysis of the use of visiting judges, see Levy, Visiting Judges, supra note 76.

109 See 282 U.S.C. § 292(a) (“The chief judge of a circuit may designate and assign one or more district judges within the circuit to sit upon the court of appeals or a division thereof whenever the business of that court so requires.”); id. § 291(a) (“The Chief Justice of the United States may, in the public interest, designate and assign temporarily any circuit judge to act as circuit judge in another circuit upon request by the chief judge or circuit justice of such circuit.”); id. § 291(b) (“The chief judge of a circuit or the circuit justice may, in the public interest, designate and assign temporarily any circuit judge within the circuit, including a judge designated and assigned to temporary duty therein, to hold a district court in any district within the circuit.”). Designations also require the consent of the chief judge of the visiting judge’s home circuit. See 28 U.S.C. § 295 (“No designation and assignment of a circuit or district judge in active service shall be made without the consent of the chief judge or judicial council of the circuit from which the judge is to be designated and assigned.”).

110 Although the constitutional text does not make it explicit, it has long been thought that lower court judges are also principal officers requiring Presidential nomination and Senate confirmation. See Weiss v. United States,

510 US 163, 191 n.7 (1994) (Souter, J., concurring) (observing that “from the early days of the Republic ‘[t]he practical construction has uniformly been that [judges of the inferior courts] are not . . . inferior officers,’ and I doubt many today would disagree” (quoting 3 JOSEPH STORY, COMMENTARIES ON THE CONSTITUTION 456, n.1 (1833) (alterations in original)).

111 David R. Stras & Ryan W. Scott, Are Senior Judges Unconstitutional?, 92 CORNELL L. REV. 453, 518 (2007).

112 50 U.S.C. § 1803(a)(1).




this court—the Foreign Intelligence Surveillance Court (FISC)—are all Article III judges, but they are not formally nominated by the President or confirmed by the Senate to serve in their dual roles as judges of the FISC. Appointment by the Chief Justice is apparently sufficient. The Chief Justice has similar power to choose three judges to constitute an appellate court that re- views the decisions of the FISC.113


We think it would be similarly permissible for the Supreme Court justices to choose addi- tional Article III judges to visit the High Court. We also note that the Appointments Clause challenge could be further reduced by adopting the strategy endorsed for the first proposal— formal appointment of all circuit judges as Supreme Court justices. That approach would elim- inate the objection that the additional justices needed to be nominated and confirmed as jus- tices of the Supreme Court.


(b)  Partisan Balance Requirements


Another objection could be raised to our proposal’s explicit inclusion of partisan-balance requirements. Would requiring that the President appoint justices of particular parties uncon- stitutionally limit her appointment power or otherwise violate the Constitution? Well, if so, a wide range well-established practices would be called into question. Similar requirements first appeared in the Nineteenth Century.114  There are now dozens of agencies with some form of partisan-balance requirement.115 Such requirements have been largely acquiesced to by Presi- dents for many decades, and the courts have never held that they are unconstitutional.


Such requirements normally do not explicitly state that particular seats belong to Demo- crats or Republicans, but instead state that no more than a set number of members can come from the same political party—effectively forcing the President to choose members of the other party (or independents) for the remaining positions. Brian Feinstein and Daniel Hemel argue that such requirements have more “bite” today than they once did, as increasing partisan polar- ization has meant that cross-party appointees are more likely to have ideologies that strongly diverge from their appointing president’s.116 While in earlier periods it was easier for presidents to find more moderate opposite party members to appoint, that is less true today.


When it comes to appointing Supreme Court justices, it is not clear that a mere limit on the number of same-party appointees on the Court would be sufficient. Given the stakes, one might expect some number of qualified but highly ideological judicial nominees to simply change their party allegiance to independent in order to improve their chances of being selected for the Court. A related piece of gamesmanship occurred in the early 2000s in relation to the




113 Id. § 1803(b) (“The Chief Justice shall publicly designate three judges, one of whom shall be publicly desig- nated as the presiding judge, from the United States district courts or courts of appeals who together shall comprise a court of review which shall have jurisdiction to review the denial of any application made under this chapter.”).

114 Feinstein & Hemel, supra note 105, at 17.

115 See Ronald J. Krotozynski et al., Partisan Balance Requirements in the Age of New Formalism, 90 NOTRE

DAME L. REV. 941, 1009–15 (2015).

116 See Feinstein & Hemel, supra note 105, at 14.




U.S. Commission on Civil Rights, “when two Republican members of the Commission on Civil Rights changed their registration to independent. Their switches allowed President George W. Bush to name two additional Republicans to the commission, bringing the number of Republican or recently Republican members of the panel to six” out of eight members.117


For this reason, it might be necessary to impose further constraints on presidential deci- sionmaking—perhaps by requiring that the President choose nominees from a list prepared by leadership of the opposite party or by some kind of bipartisan commission. Such a restriction on presidential decisionmaking power would no doubt be criticized, but there are some analo- gies in existing law. Under District of Columbia Law, the President must select judicial nomi- nees to the D.C. court system from a list prepared by a multimember commission known as the District of Columbia Judicial Nomination Commission.118 Despite significant grounds for pos- sible constitutional objection,119 Presidents of both parties have generally abided by this sys- tem’s requirements rather than picking a legal fight.120 The stakes are higher here, and thus there is surely a greater chance that these kinds of restrictions would be challenged. The example proves, however, that it is possible to reach a settlement that both sides can live with even in the phase of some meaningful constitutional objections. And it is our hope that such a settlement is possible here, if both sides could be convinced this system would be better than the open partisan warfare into which our current system is degenerating.




Recent events have seriously called into question the legitimacy of the Supreme Court. Re- publicans engaged in unprecedented exercises of raw power in order to capture control of the judicial branch. The coming years will almost certainly pose even deeper challenges to the Court’s legitimacy. The left will struggle against a deeply conservative Supreme Court majority that will likely try to stand in the way of various progressive policy initiatives. Proposals like court-packing, that might have seemed unthinkable just a few years ago, could quickly go from fantasy to reality.


From one perspective, this conflict is a welcome one. Americans have not been sufficiently critical of the Supreme Court’s role in our society over recent decades. The Justices have arro- gated massive amounts of power to themselves, and it is high time for us to question as a nation how much power we should give to nine unelected and unaccountable judges to resolve our




117 Id. at 21.

118 D.C. CODE § 1–204.33.

119 The most obvious objection concerns the Appointments Clause. By limiting the President’s power to nom- inate whoever she wishes to a federal office, such a law might impermissibly encroach on the separation of powers. See, e.g., Note, Congressional Restrictions on the President’s Appointment Power and the Role of Longstanding Prac- tice in Constitutional Interpretation, 120 HARV. L. REV. 1914, 1919 (2007) (suggesting that “there is strong evi- dence that the original understanding of the Appointments Clause grants the President plenary appointment power contingent only on Senate confirmation”).

120 Just last year, President Trump selected a nominee for the D.C. Courts from the list prepared by the Com- mission. See Seventeen Nominations Sent  to the  Senate  Today, THE WHITE  HOUSE, June 29, 2017,




most divisive disputes. At the same time, however, there are serious dangers ahead. If the Court’s legitimacy is seriously called into question in the coming years, there are reasons to worry about a collapse of public confidence in the rule of law more generally. If it is possible to save what is good about the Court, while still recognizing how the Court must change, we should do so.


In this essay, we have offered two different proposals that could save the Supreme Court as an institution above politics. Neither solution is perfect; each would fail to address some of the problems with the Supreme Court. We are confident, however, that each proposal would be an improvement over the status quo—especially given how we expect our already broken system to deteriorate even further in the near Term. We would be happy if either proposal were taken as a blueprint for reform, or if ideas from each were combined in some way as a model for change.


But whether our particular proposals are adopted, in whole or in part, is not ultimately the most important thing. What matters most is recognizing the need for some kind of reform of this kind to address the coming threats to the Court’s legitimacy. Simply maintaining the status quo is no longer an option. Given present circumstances, saving the Court will require radically changing the Court.

Related posts

Leave a Reply