Court Reform for Progressives: A Primer on Constitutional Considerations

PDF linkDownload PDF

Mark Tushnet

Introduction and Preliminaries

This brief and basically unannotated essay lays out some constitutional considerations associated with prominent (and some not so prominent) proposals for Supreme Court reform circulating among progressives.1 I confess to the feeling that writing this in the summer of 2024 resembles analysis of policy proposals for the German Social Democratic Party in January 1933. Here are two “defenses.” Saying something useful about the obviously more important questions on the table is outside my area of expertise. Legal progressives have typically, though not universally, focused on policy proposals for adoption in the short term; legal conservatives successfully played the long game. Legal progressives might learn something from that.

Two preliminary points: first, actually accomplishing Court reform by statute, one might think, requires (at least) what’s come to be called the “trifecta:” control of the House, Senate, and the Presidency by a single party whose members are generally committed to the idea of Court reform. The idea of “control” is a bit complicated. Court reform can be accomplished by constitutional amendment, of course, which requires supermajorities in the House, the Senate, and the states—but not the presidency. I generally put Court reform via constitutional amendment to one side in this essay, though the existence of the amendment process comes up in connection with some proposals for statutory reform. “Control” of the House and the Presidency is straight-forward, that of the Senate less so. There it might be a majority large enough to overcome a filibuster, or it might be a majority willing to enact legislation by simple majority, ignoring the filibuster on this issue or in general.

Second, I focus on proposals that would affect the Court immediately upon adoption. Ordinarily politicians have incentives to use their political capital to work for adoption of legislation that will give them immediate political benefit. Members of Congress typically don’t spend their capital on projects that will pay off fifteen years down the road. So, for example, a proposal to limit the terms of those Supreme Court justices appointed hereafter has little in it to recommend to a politician who believes that voters want change in the Supreme Court now.2 There’s one qualification: proposals with long-term payoffs might be part of a comprehensive institutional reform package—call it a “good government” program—with some immediate payoffs and some deferred payoffs. At the moment, though, Supreme Court reform doesn’t seem to be part of such a broader reform movement.

It turns out, though, that focusing on immediate effects might not mean focusing on statutes that are actually enacted. There’s some evidence from political scientists that the introduction of court-curbing proposals—not their enactment—affects the Court’s overall behavior.3 The introduction of such proposals in sufficient volume signals to the Court that a non-trivial number of legislators think the Court has overreached, and the Court appears to react by pulling back from its more advanced positions. The mechanism by which this effect occurs is quite obscure and not, I think, well described in the political science literature. It might have to do with individual justices’ desires to get approval from “attentive” audiences—and if so the effect might be muted in today’s political world where conservative and liberal justices have separate circles of social approval.4

As already indicated, I keep annotations to a bare minimum.5 As a primer this essay doesn’t deal with every wrinkle in the arguments, just those at the core. Readers who wonder about some point that I don’t address should feel free to contact me.

The essay has four parts, dealing successively with low-hanging fruit (about which there are few constitutional questions—though not none), statutory term limits including “bells-and-whistles” proposals for accomplishing effective term limits, jurisdiction-stripping, and Court expansion. The conclusion: proposals for Court reform raise rather deep questions about the kind of democratic self-governance system we want to have—as does resistance to such proposals (that is, saying that Court reform is a bad idea raises deep questions about the kind of democratic self-governance system we want to have and which the objector believes to be close to what we actually have).

I. Low-Hanging Fruit

Any Court reform can be achieved through constitutional amendment, of course. Impeachment is another constitutionally authorized method for affecting the Court’s composition.6 It requires a majority in the House and a supermajority in the Senate (and no participation by the president). The failed impeachment of Samuel Chase is conventionally taken to stand for the proposition that judges cannot be impeached merely for disagreement with their rulings. More formally, I suppose, the Chase impeachment (and the fact that no federal judges, including Justices, have been impeached for anything other than ethical lapses) is taken to be a congressional interpretation of the term “good behavior.” “Good behavior” means, roughly, “ethical behavior” rather than “behavior that advances a good vision of the law.”7

Congress also controls the Supreme Court’s budget.8 I have regularly, and without any success, raised the possibility that Congress could reduce the number of law clerks each justice is authorized to hire. The rationale is that the Court has substantially reduced the number of cases it decides on the merits while retaining and even expanding the number of law clerks each justice has—and that it is good public policy for justices to write more of “their” opinions (“more of” meaning both a larger number written personally by each justice and a greater portion of each opinion issued under the justice’s name). We know enough about the way justices employ law clerks to believe that reducing the number of law clerks will give incentives to older justices to retire (that is, we know that as justices serve more years on the Court they delegate more opinion-drafting responsibility to their law clerks—in my view, I should add, an entirely sensible course of action). They might respond to the reduction in the number of law clerks by reorganizing their offices, which is to say that the incentive effects of this proposal are only incentive effects, not mandates.

This proposal has several features that deserve specific mention. It is framed as a “good government” measure—people who do less work than they used to should get fewer resources than they used to. It might have an effect similar to term limits, though the choice to retire is left to each justice. And the increased incentive to retire might be an intended effect. This third feature raises a question I address in more detail in connection with Court expansion. Suppose a proposal would have the effect of changing the Court’s composition in a setting where it’s reasonable to believe that Congress disagrees with the substantive decisions the justices—and in particular disproportionately the older justices who will feel the incentives most strongly—have rendered. Is it sufficient that there be a “good government” rationale for the proposal (that is, is the proposal insulated from constitutional invalidation by the existence of such a rationale)? Or must the “good government” rationale actually motivate the proposal’s adoption (is that rationale a mere fig-leaf badly concealing an intention to alter outcomes)? Perhaps there’s a convention, similar to that associated with impeachment, that mere “fig leaf” rationales are inadequate.

There is then a further question: what if the rationale is not a “mere” fig leaf? Suppose we have (as we almost inevitably would) a “mixed motive” case. Members supporting the proposal do think that there’s a good-government rationale for decreasing the number of law clerks but some (or all) of them also think that the changed incentive structure will lead in the relatively short run to a change in the Court’s composition that will mean that the Court will render fewer decisions with which they disagree. I’m reasonably confident that the answer you give will be entirely parasitic on your views about the decisions supported by the justices likely to respond to the incentives by retiring.

II. Statutory Term Limits

Statutory term limits typically operate as follows: a person is nominated for a position on the United States courts. The person’s commission would read, “During good behavior, on the Supreme Court for [eighteen] years and thereafter on a court of appeals to be designated by [the Chief Justice or some other person or body within the judicial branch].” Note of course that the commissions of currently sitting Supreme Court justices do not read that way. Statutory term limits, then, have only prospective effect, which means that, despite their popularity, no member of Congress has much reason to expend much effort to get them enacted.

One possible constitutional difficulty with this strategy is that the two-part commission might not be permissible. One way to see this involves noting that Congress is under no obligation to create federal courts of appeals. Suppose a person receives a two-part commission and Congress then abolishes the federal courts of appeals. The person’s tenure during good behavior is replaced by a term of years. Alternatively, the only court specifically identified in the Constitution is the Supreme Court.9 That might be taken to imply that a person’s commission must refer only to that Court. Or, the Constitution refers to “one Supreme Court,”10 implying that a person can be appointed to only that Court (otherwise the person would serve on “one Supreme Court” and a court of appeals). Or, yet again, at least for the transition period before currently sitting justices depart we would have two Supreme Courts, one consisting of Justices with life tenure on the Court, the other of justices with the two-part commission.

None of these are knock-down arguments, of course, and I’m inclined to think that they ought to be rejected on the merits. But neither you nor I are the relevant decision-makers here; the justices of the Supreme Court are, and I would guess that they’d be unsympathetic to efforts to alter the tenure status of those who would join them on the Court.

The “complex commission” problem attends the most prominent “bells and whistles” alternatives to statutory term limits, such as Epps and Sitaraman’s “lottery court,”11 in which each court of appeals judge would receive a commission to sit on the Supreme Court for two weeks at random intervals. And “bells and whistles” proposals have their own political drawbacks. Ordinary people can understand what’s going on when the proposals on the table are term limits or Court expansion or (maybe) even jurisdiction stripping. Politicians can explain why those proposals are good ideas (if they are). “Bells and whistles” proposals are different. They’re so complicated that all a politician can say, really, is something like “Trust me, this is going to fix the Court.”

III. Jurisdiction Stripping

Consider a federal campaign finance statute that’s plainly inconsistent with Citizens United12 and the overall jurisprudence currently governing campaign contributions and independent expenditures (I’ll call this a CUI statute—for “Citizens United inconsistent”) or a federal statute mandating affirmative action in the admissions policies of the nation’s military service academies. Congress attempts to insulate the statutes from invalidation by denying courts jurisdiction to consider whether they are consistent with the Constitution. Here I consider only proposals to strip all courts of jurisdiction to consider the constitutionality of specific federal statutes or categories of statutes.13

A bit of background, again quite simplified: most scholars think that Congress need not create any lower federal courts. Having the power to create them or not, Congress, most scholars think, has essentially unlimited power to “define their jurisdiction,” that is, to say which kinds of cases they are authorized to hear (and which they can’t). The Constitution requires that there be a Supreme Court,14 but it also (scholars generally think) gives Congress the power to limit the categories of cases it hears on appeal from the lower federal courts and from the state courts.

Actually putting the idea of jurisdiction-stripping into statutory language that does the job is quite difficult. The CUI statute limits contributions and independent expenditures. How are the limitations to be enforced? Typically (if the drafters don’t get extremely creative) by injunctions or criminal prosecutions. Now suppose the CUI statute has a jurisdiction-stripping provision stating “No court, federal or state, and including the United States Supreme Court, shall have jurisdiction to determine whether or not the limitations herein are constitutional.”15

An action for injunctive relief is brought in a federal or state court to prohibit a banned independent expenditure. That court has to decide whether to issue the injunction. Maybe it will say, “I still have equitable discretion here. Because Congress has withdrawn one obviously relevant consideration from my overall calculus, I’m going to exercise that discretion to deny the injunction.” Perhaps that seems an overreach, with equitable discretion used as an indirect way of saying that the substantive provisions are or probably are unconstitutional. Still, one reading of United States v. Klein16 is that Congress can’t tell the courts that they have to rule in one direction only, when there’s a constitutional issue lurking behind the scenes.

Perhaps we can get around the equitable-discretion obstacle. The court will be asked next to hold the jurisdiction-stripping provision unconstitutional. That provision, as written, doesn’t eliminate jurisdiction to consider that question (Hart’s version is that a federal court will rely on the general federal question jurisdiction to consider that question.17)

I note, not entirely incidentally, that those who actually sit down to write jurisdiction-stripping statutes rarely anticipate this problem, perhaps because doing so introduces a “bells and whistles” difficulty in explaining what’s going on. Suppose, though, that anticipating it the CUI statute not only denies jurisdiction to consider the constitutionality of the regulations of contributions and independent expenditures but also denies jurisdiction to consider the constitutionality of the jurisdiction-stripping provision (in Hart’s version, Congress removes challenges to the jurisdiction-stripping provision from the general federal question jurisdiction18).

Is the two-level jurisdiction-stripping provision unconstitutional? The CUI statute (arguably) violates the First Amendment. Does the jurisdiction-stripping provision?19 Probably the most that can be said is that existing First Amendment doctrine isn’t well-designed to answer that question. It’s hard to say, for example, that the jurisdiction-stripping provision restricts speech based upon its viewpoint, content, or subject matter—or at least that it does so in the way the Court has used those terms heretofore. Maybe the Court could develop something like a “disparate impact” theory of viewpoint regulation. Regulations not dealing directly with speech might be viewpoint based if they disproportionately have viewpoint effects. It’s hard to know what the contours of such a doctrine would be and in particular whether the doctrine would be more than an ad hoc way to deal with this specific problem.

It’s sometimes suggested that there’s something like an equal protection problem here (putting aside the problem that the Equal Protection clause doesn’t apply to congressional action). The idea is that the First Amendment rights implicated by the CUI statute are treated less favorably than other constitutional rights. One response is, so what? Rights don’t have rights, people do. As a regulation of the jurisdiction of the lower federal courts, the jurisdiction-stripping provision is a docket-control measure: “Leave these questions to state courts.” You can gin up an equal-protection-like argument against doing that but here too existing equal protection doctrine isn’t well-designed to deal with the problem.

One possibility is this: laws adopted for the purpose and with the likely effect of denying someone her constitutional rights are unconstitutional no matter what their form. Maybe so, but proponents of the CUI state don’t say that they want to deny anyone their true First Amendment rights. In their view, the Supreme Court has seriously misinterpreted the Constitution and the CUI statute is a way of correcting that misinterpretation. According to proponents, if the Court can overrule its own decisions as wrong on the day they were decided, why can’t Congress? More on this below.

Note, though, that the CUI statute refers to both federal and state courts. Where does Congress get the power to deny state courts the power to determine the constitutionality of a federal statute? Start with the CUI statute. Election law specialists say that Congress has the power to enact it as a regulation of commerce among the several states. Sending money to a candidate or election committee headquartered in another state probably does count as interstate commerce even under post-Lopez jurisprudence, though we might expect cognitive pressure on the justices to revive limits on the commerce power when confronted with a statute relying on it to preclude state court jurisdiction over constitutionally problematic legislation.

Perhaps we could try to deal with this by making jurisdiction over all questions associated with the CUI statute exclusive in the federal courts (and then, of course, eliminating that jurisdiction). Consider here the “affirmative action for service academies” statute. Suppose it has a provision saying that jurisdiction over constitutional questions associated with the statute is exclusive in the federal courts. Exclusive jurisdiction might be permissible because of the congeries of constitutional provisions dealing with the armed forces of the United States (there are too many to list here). I don’t think something similar could be said about the power to regulate elections; indeed, the relevant clause gives states the first move in designing regulations, with a congressional power to displace or supplement them. This suggests that the power can’t support exclusive federal jurisdiction. And, once again, we have the recursion problem: can Congress prevent the state courts from considering the constitutionality of the provision conferring exclusive jurisdiction on the federal courts?

I’m going to stop proliferating problems here. It turns out to be really difficult to draft a statute that will prevent state courts (at least) from considering the constitutionality of the CUI statute’s substantive provisions. Suppose, though, that all the difficulties were overcome. With the state courts in the picture, we can now consider what seems to me the scholarly consensus about jurisdiction-stripping. According to that consensus, our constitutional system rests on the presupposition—not written anywhere in the Constitution (but we know that there are such presuppositions—see Trump v. United States20)—that somewhere there’s a court with the power to determine the constitutionality of federal legislation,21 at least absent truly compelling reasons.22

So, sure, maybe Congress can use its power to regulate the lower federal courts (and make exceptions to the Supreme Court’s appellate jurisdiction) to prevent a federal court from considering the constitutionality of the CUI statute’s substantive provisions. But it can’t do that for the state courts. State courts can always unwrap the layers of protection Congress has wrapped around the substantive provisions. A state court invokes the presupposition to hold the provision giving federal courts exclusive jurisdiction unconstitutional, then holds the provision purporting to strip state courts of jurisdiction unconstitutional because, for example, it’s not within the power to regulate commerce among the several states, and then holds the substantive provisions unconstitutional.

Here it’s worth mentioning what’s basically a policy consideration. What’s a state court to do about the merits of the challenge to the CUI’s substantive provisions? Does the statute freeze the law in place, leaving Citizens United as controlling precedent? If so, the substantive provisions are unconstitutional. If not, state courts are free to examine the substantive constitutional question as a matter of first principles. Some state courts will say that Citizens United remains controlling law, others will disagree and consider the First Amendment issues ab initio. Among the latter, some will end up agreeing that Citizens United correctly interpreted the First Amendment and others will end up disagreeing. Limits on contributions and independent expenditures will be constitutional in some state and not in others. Some (not me) will be troubled by this disuniformity, which might affect (“distort,” opponents of jurisdiction-stripping will say) the pattern of contributions.23 These policy concerns will affect politicians’ incentives to vote for the CUI statute, particularly because they’re not going to be able to predict well what the courts of their states will do and so won’t be able to predict well what the statute’s effects on their constituents (and their candidacies) will be.

There’s another wrinkle. Suppose the CUI statute is enforced via a criminal prosecution. I’m going to be quite telegraphic about the arguments here. The defendant says, “You can’t convict me because the statute’s unconstitutional.” The trial court and appeals court say, “Sorry, we don’t have jurisdiction to consider that defense.” The defendant, now convicted, seeks habeas corpus relief in a federal court. The habeas court says, “Sorry, etc.” The defendant/habeas petitioner says, “Wait a minute, that means that Congress has suspended the habeas clause. First of all, there (ought to be) a clear statement rule here: if Congress wants to suspend habeas it’s got to say so explicitly. And second, this isn’t a case of ‘Rebellion or Invasion’ so Congress doesn’t have the power to suspend.” The habeas court says, “Interesting arguments, but we think the statute’s clear enough and denies us jurisdiction to consider your ‘Rebellion or Invasion’ argument.”

The defendant now goes to state court seeking habeas relief. The federal government says, “Independent of the CUI statute, state courts don’t have habeas jurisdiction over federal prisoners.” The state court (subject to the considerations in the preceding paragraph about variation among state courts) responds, “The cases you rely on do say ‘exclusive jurisdiction,’ but they all assume that there’s habeas jurisdiction in the federal courts. And the presupposition about the availability of some court kicks in here. So we’re going to decide the case on the merits.”

What happens when the case gets to the Supreme Court? Again stripping away all the preliminaries, either the Supreme Court says, “We don’t have jurisdiction per the CUI statute (so the state court decision stands),” or it says, “We have jurisdiction and we have to consider the defendant’s claims on the merits. Otherwise, Congress would be telling us that we have to, in effect, affirm a conviction that’s defensible only if the underlying statute is constitutional. We know that Congress can tell us how to resolve statutory questions (even by using jurisdictional language to specify what the statute really means), but it can’t do that for constitutional questions.”

At the end of the road, then, we have two issues to resolve for jurisdiction-stripping to “work.” In reverse order of their appearance: “Why not?” That is, why can’t Congress tell the courts what the Constitution means? True, it’s emphatically the province and duty of the courts to say what the law is, but that doesn’t have mean that only the courts can say what the law is. And second (perhaps the same question in different words), where does the presupposition about there being a court somewhere with the power to resolve constitutional questions come from? And, perhaps more pointedly, even if there is such a general presupposition, why can’t we acknowledge a power in Congress selectively to disregard the presupposition? As the preceding material shows (and there are lots of details I’ve omitted that would drive the point home even more forcefully), it’s really hard to exercise such a power effectively. There certainly would be political costs associated with doing so, which suggests that any slippery slope here would have a lot of friction and resistance. And, maybe, sliding gradually down the slope wouldn’t be so bad for democratic self-governance (“slowly” matters because we’d want to allow people to adapt their political behavior as the presupposition gradually fades away).

IV. Court Expansion

Why isn’t Court expansion a low-hanging fruit? The Court’s size is set by statute and it has varied from time to time. Indeed, proposals surface from time to time to amend the Constitution to set the Court’s size, which suggests that under current law Congress does have the power to set the Court’s size wherever it wants.24

Here’s a specific proposal grounded in historical experience. There used to be one Supreme Court justice “for” each circuit court of appeals. The practice of one circuit, one justice arose when the justices actually “rode circuit,” that is, sat on courts within those circuits. This “good government” rationale for changing the Court’s size doesn’t stand alone, though. Justin Crowe has shown that the timing of the creation of new circuits, and so of new justices, was often driven in substantial part by the expected effect of a new appointee on the court’s decisions.25 And, the temporary reduction in the Court’s size after the Civil War was almost entirely driven by politics—to deny Andrew Johnson the opportunity to name justices to the Court.26 Today’s proponents of Court expansion would add the Senate’s inaction on the nomination of Merrick Garland to the list of actions taken for purely political reasons to affect the Court’s size. Overall, then, we have a history of modifying the Court’s size for plausible good government reasons tempered by politics—and one or two examples of purely political changes in the Court’s size.

Today there are thirteen circuit courts, so the Court’s size “should be” expanded to thirteen. Though the justices don’t ride circuit, even today they do some circuit-specific work. Most notably, they rule on emergency motions coming from the circuits for which they are responsible. So, the “one circuit, one justice” rule has some modest functional justification.

But, of course, there’s more than that going on. If Democrats won the trifecta and added four justices, the Court’s composition would change from six justices nominated by Republican presidents and three by Democrats, to seven nominated by Democrat and six nominated by Republicans. That effect opens up the possibility of constitutional challenges to Court expansion.

Here everything turns, I believe, on whether we give constitutional significance to the defeat of FDR’s Court-packing plan in 1937 (and on ignoring the Andrew-Johnson era reduction in the Court’s size). Was that defeat an ordinary political loss or did it somehow establish or consolidate some sort of constitutional rule limiting the occasions for changing the Court’s size? The best recent scholarship establishes that neither Congress nor “the country” decisively repudiated the Court-packing plan. It was a close-run race up to the very end. FDR packaged the plan as a good-government measure, pretending that the Court needed new members to do work that elderly justices were unable to do effectively. His disingenuousness impaired the political effort to get the plan adopted. Yet, constitutional language, mostly about judicial independence, pervaded the arguments of the plan’s opponents.

The “ordinary politics versus constitutional deliberation” question is difficult to answer because we have only one data point. Or, more precisely, we have only one serious effort in the past hundred years or so to change the Court’s size. Recent discussions of the new “history and tradition” approach to constitutional interpretation have brought to attention a systematic problem when evidence of practice is so thin. Perhaps there’s a tradition of not changing the Court’s size for purely political reasons, Or perhaps there’s no such tradition, because no one tried to do so, thinking that such action wasn’t necessary. To use a cliché, absence of evidence (of purely politically driven changes in the Court’s size) isn’t evidence of absence (of congressional power to make such changes).

Note the formulation: purely politically driven changes. Suppose proponents of change offer some modest good-government rationale for the change—in my example, the value of one justice per circuit. If that’s a pure fig-leaf, meaning here, not even a decent rationale on its own terms (which it might well not be), we have the purely politically driven question. What if the good-government rationale gets over some probably low bar of plausibility, but that isn’t what motivates its supporters? That is, they support the change mostly because it would likely lead to changes in the decisions the Court hands down but believe as well that it would have some modest good-government effects.

At this point the precedent of the failed Chase impeachment might matter. Impeaching a justice for “wrong” decisions changes the Court’s composition with the likely effect of changing outcomes (on the assumption that the impeached justice’s replacement would vote differently). If you can’t use a supermajority requirement (conviction in the Senate) to do that, you shouldn’t be able to use ordinary majorities to do so by manipulating the Court’s size.

On this view, we can change the Court’s size for good-government reasons even if we know that doing so will have some effects on outcomes. Otherwise, we have to rely on the ordinary processes of retirement and death to replace justices, backed up by the possibility of a constitutional amendment with a strong supermajority requirement.

It’s important to see what this (supposed) restriction on congressional power to alter the Court’s size rests on: giving a great deal of weight to the failure of the Court-packing plan, ignoring the Johnson-era events, treating the absence of efforts to change the Court’s size as an indication of a normative view about congressional power, and some inferences from the structurally similar issue of impeaching for erroneous decisions.

But of course there’s actually more to it: a view about the proper role of the Supreme Court in our system of democratic self-governance, as I put it in connection with jurisdiction-stripping. I suspect that the specific elements of the argument matter pretty much only because you might have a view that a system in which the Supreme Court has real power to implement its views about constitutional meaning (against the views of Congress) is better than one in which congressional views about constitutional meaning have substantial and sometimes overriding weight.

The case against congressional power to change the Court’s size for purely political reasons relies on rather expansive inferences from specific structural arrangements to deeper “postulates which limit and control” (Principality of Monaco v. Mississippi27) and “penumbras, formed by emanations” (Griswold v. Connecticut28), which one might think would be disfavored by the current Court. “Disfavored” doesn’t mean “always to be disparaged,” as Trump v. United States29 shows. That’s why this discussion isn’t in the “low-hanging fruit” section.30

Conclusion

All the proposals for statutory Supreme Court reform considered here run up against constitutional objections, some substantial, others merely plausible. My guess is that the same would be true of any other proposals for statutory Supreme Court reform.

The real lesson, though, is that we can’t resolve those questions through careful analysis of the Constitution’s text and structure and existing constitutional doctrine. Anyone who says either “obviously unconstitutional” or “obviously constitutionally permissible” is blowing smoke. What’s at stake—that is, how you think the questions should be resolved—depends upon your vision of how to run a well-designed system of democratic self-governance.

I’ll end with a common argument, seemingly policy-based, against Court expansion. It’s the “tit for tat” retaliation point. You do it when you have the trifecta and your opponents will do it when they do, ultimately giving a Court with 57 members. That might be an exaggeration. One paper models the “trifecta” scenario and concludes that sixty years from now the Court might have 27 members31—in my view not unwieldy. The Court would be so large that it wouldn’t be able to develop a consistent constitutional jurisprudence. The American people would come to see the Court as simply another actor in ordinary politics, not an institution dedicated to the principled interpretation of the Constitution. One perfectly sensible response to that argument (mine, I freely admit) is, “And your point is?”

William Nelson Cromwell Professor of Law emeritus, Harvard Law School.

  1. ↑ 1 That’s why the idea of adopting term limits to take effect gradually differs from the Twenty-Second Amendment, which imposed term limits on the presidency but exempted the sitting president. U.S. Const. amend. XXII. At the time of its proposal and adoption no one expected Harry Truman to serve more than two terms anyway. Giving it prospective effect wasn’t politically costly and might have helped build a cross-party coalition in its favor (and such a coalition might have existed no matter what because the parties were less ideologically unified than they are today).
  2. ↑ 2
    Tom Clark, The Limits of Judicial Independence 62–121 (2011).
  3. ↑ 3
    Neal Devins & Lawrence Baum, The Company They Keep: How Partisan Divisions Came to the Supreme Court 15–58 (2019).
  4. ↑ 4
    Readers who wonder about the support for my assertions about possible constitutional problems should regularly say to themselves, “For the analysis underlying this proposition, see Henry Hart, The Power of Congress to Limit the Jurisdiction of Federal Court: An Exercise in Dialectic, 66 Harv. L. Rev. 1362 (1953).”
  5. ↑ 5
    U.S. Const. art. II, §4.
  6. ↑ 6
    In July 2024, Representative Alexandria Ocasio-Cortez introduced a resolution of impeachment for Justices Clarence Thomas and Samuel Alito that referred to what the resolution described as their unethical behavior—and notably she did not introduce a resolution with respect to other members of the Court’s conservative majority. Press Release, Alexandria Ocasio-Cortez, Congresswoman, House of Representative, Ocasio-Cortez Introduces Articles of Impeachment Against Justice Thomas and Justice Alito (July 10, 2024), https://ocasio-cortez.house.gov/media/press-releases/ocasio-cortez-introduces-articles-impeachment-against-justice-thomas-and [https://perma.cc/8ADV-53PX]. The Abe Fortas episode deserves mention as well. He resigned in the fact of realistic threats of impeachment, the substance of which would have involved serious ethical lapses but the motivation for which certainly included disagreement with his substantive positions. The ethical lapses were more than a fig-leaf, though.
  7. ↑ 7
    U.S. Const. art. II, §4.
  8. ↑ 8
    U.S. Const. art. III, §1.
  9. ↑ 9
    Id.
  10. ↑ 10
    Daniel Epps & Ganesh Sitaraman, How to Save the Supreme Court, 129 Yale L.J. 181–92 (2019).
  11. ↑ 11
    Citizens United v. FEC, 558 U.S. 310 (2010).
  12. ↑ 12
    If you don’t like the Supreme Court and only the Supreme Court (the set of people with such a view is I think pretty close to empty), you could limit the Court’s jurisdiction to cases that fall within the constitutionally mandated original jurisdiction and—if you think that invoking the “exceptions” clause requires that there be something left to the appellate jurisdiction—cases “between Citizens of the same State claiming Lands under Grants of different States.” U.S. Const. art. III, §2, cl. 1.
  13. ↑ 13
    Id. at art. III, §1.
  14. ↑ 14
    The Supreme Court is specifically mentioned to avoid the possibility that, with respect to a statute referring generically only to “courts,” the Court would invoke a clear-statement rule and hold that the generic term refers only to lower courts.
  15. ↑ 15
    United States v. Klein, 80 U.S. (13 Wall.) 128, 146–48 (1871).
  16. ↑ 16
    Hart, supra note 5, at 1387–96.
  17. ↑ 17
    Id. at 1396–1401.
  18. ↑ 18
    In the literature this is phrased as, “Is there an external limit on Congress’s power to regulate jurisdiction?”
  19. ↑ 19
    Trump v. United States, 144 S. Ct. 2312 (2024).
  20. ↑ 20
    You can shoehorn the presupposition into the Fifth Amendment’s Due Process Clause but I’m pretty sure you’d do so only if you already thought there was such a presupposition—that is, you can’t find in the usual sources for interpreting the Due Process Clause enough material to establish that somewhere there has to be a court with the power to determine the constitutionality of federal legislation.
  21. ↑ 21
    The “truly compelling reasons” clause here might be invoked in connection with the service-academies statute.
  22. ↑ 22
    “Might,” because the distortion would consist of money flowing from states with limits to states without (unless the states with limits say that their limitations apply to contributions and expenditures outside their boundaries, which would raise another set of constitutional questions).
  23. ↑ 23
    I note simply in passing that the Constitution’s “numerical” provisions—two senators for every state, age qualifications, the twenty-dollar jury trial right—are among its most problematic precisely because the numbers were set against a background in which they probably made sense, but they have come into question as the background has changed.
  24. ↑ 24
    Justin Crowe, Building the Judiciary: Law, Courts, and the Politics of Institutional Development (2012).
  25. ↑ 25
    The best recent comparative scholarship makes it clear that court-packing and court-slimming are, under the right circumstances, similar ways of using changes in the court’s size to achieve substantive outcomes.
  26. ↑ 26
    Principality of Monaco v. Mississippi, 292 U.S. 313, 322 (1934).
  27. ↑ 27
    Griswold v. Connecticut, 381 U.S. 479, 484 (1965).
  28. ↑ 28
    Trump, 144 S. Ct. at 2346–47.
  29. ↑ 29
    See infra Section I.
  30. ↑ 30
    Adam Chilton, Daniel Epps, Kyle Rozema & Maya Sen, The Endgame of Court-Packing (May 3, 2021) (Working Paper), https://papers.ssrn.com/abstract=3835502 [https://perma.cc/8KC2-GLJJ].
  31. ↑ 31
    The President’s Commission’s report focused on policy considerations and touched on constitutional issues only in passing. Presidential Comm’n on the Sup. Ct. of the U.S., Final Report (2021), https://www.whitehouse.gov/wp-content/uploads/2021/12/SCOTUS-Report-Final-12.8.21-1.pdf [https://perma.cc/XS3U-YTCC].