How (Not) to Reform the Supreme Court, Part One

President Joe Biden has just this morning published an op-ed in the Washington Post sketching out his plan for Supreme Court reform. I want to say right off the bat that, completely aside from the details of his plan, this is great to see. We absolutely need court reform, and now Joe Biden of all people is throwing his weight behind the cause.

That being said, I'm a constitutional philosopher and a huge nerd so of course I want to critique the details of Biden's proposal. For reasons that will become clear, to some extent this is an occasion for a broader conversation about what Congress can and cannot do vis-a-vis the Court, since Biden himself backs down from one of the most controversial claims. Also, there's a lot to go over here, so I'm going to split this piece in two: this first Part will discuss the stuff Biden did propose, and the second will discuss the stuff he didn't propose but should have.

Biden's plan has three components:

1. A constitutional amendment to overrule Trump v. United States and clarify that "there is no immunity for crimes a former president committed while in office."
2. Adopting (through unspecified means; this is important) 18-year term limits for the Supreme Court, with each president making two appointments in the course of their four-year term.
3. Adopting a "binding code of ethics" for Supreme Court Justices.

I'm going to save discussion of the first proposal for the end, so let's start with the term limits thing.

As noted, Biden does not endorse any particular means of imposing term limits, he just says he would "support a system" of 18-year terms. This is good, because Congress has no power to impose term limits on the Supreme Court by ordinary statute. This point has been hotly contested on Bluesky when I have asserted it, so I want to provide a really thorough explanation here.

To some extent this gets into questions of interpretive philosophy. That's why I recently wrote about sophomoric textualism. The thing about that "there's nothing in the rules that says a dog can't play basketball" approach is that there's an awful lot of stuff that the Constitution doesn't expressly talk about. There's nothing in the rules that says people have a right to abortion, or to travel from one state to another. Hell, there's nothing in the rules that says the president can't abridge freedom of speech! Despite which it's correct to say that the Constitution is best read as saying all that stuff.

But in order to get from the text, which says fairly little in so many words, to that conclusion, you need to adopt a holistic interpretive approach that strives in good faith to make sense of the thing as a whole. This is where the family of interpretive modalities comes in. The intentions of those who enacted the Constitution (including the amendments) matter; so does historical practice and precedent. So does the ethos of the thing, its character, the overall orientation and philosophy.

The goal, then, is not to come up with the strongest argument in favor of statutory term limits for the Supreme Court. It's to take an honest look at the Constitution, and Article III in particular, and discern whether the best reading thereof would or would not permit such a thing. And I think it's abundantly clear that the answer is "no."

We can start with the text. Here's all the relevant textual stuff:

Article I, § 8, cl. 9: The Congress shall have Power ... To constitute Tribunals inferior to the supreme Court.

Article I, § 8, cl. 18: The Congress shall have Power ... To make all Laws which shall be necessary and proper for carrying into Execution the foregoing Powers, and all other Powers vested by this Constitution in the Government of the United States, or in any Department or Officer thereof.

Article III, § I: The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish. The Judges, both of the supreme and inferior Courts, shall hold their Offices during good Behaviour, and shall, at stated Times, receive for their Services, a Compensation, which shall not be diminished during their Continuance in Office.

That's it, basically.

A lot of the debate centers around the meaning of "good Behaviour." I contend that this means "life tenure, subject only to removal via impeachment." A lot of people seem to disagree, suggesting for example that Congress could define by statute what "good behavior" means. I had at least one person suggest to me that "staying on the Court for more than 18 years" could be defined as bad behavior.

All of this is wrong. It's quite clear that it does, in fact, mean life tenure. This was the original understanding of the provision. In Federalist No. 78, Hamilton speaks of federal judges as having "permanent tenure;" in No. 79 he says that they "will be secured in their places for life." The phrase, moreover, has a history. English judges had since time immemorial served at the pleasure of the Crown; they were originally, in like the twelfth century, just royal deputies. One of King James II's provocations that led to the Glorious Revolution in 1688 was removing any judge who ruled against him. So in the Act of Settlement 1701 (most famous for barring Catholics from the royal succession) Parliament abolished the Crown's power to remove judges, instead stating that they would hold their offices during good behavior, subject only to removal by Parliament, i.e. through impeachment.

This history illustrates a deeper truth: the fundamental policy of Article III is one of judicial independence. This is plain to see if we imagine a world where Article III just didn't exist. We would still have federal courts in that world, but they would be altogether creatures of Congress, much as the English courts were for centuries creatures of the Crown. What Article III does is afford the judiciary an independent foundation, an independent existence. This policy is plain to see in the language immediately following "good Behaviour," which ensures that Congress can never use judges' salaries as leverage over them. Construing "good Behaviour" to mean "life tenure, subject only to impeachment" is thoroughly consonant with this overall philosophy.

Now it's true that Article III does not place the courts wholly beyond the realm of political contestation; nor should it! Congress does have plenty of power over the judiciary, and even over the Supreme Court. While the judges' salaries are constitutionally ensured, the Court's budget is not, and indeed the 1802 Supreme Court term was famously canceled by Thomas Jefferson and the Jeffersonian Congress. (Cancel culture amirite?) Congress has a broad power to regulate the Court's functioning, its docket, even its jurisdiction. (I am a bit of a jurisdiction-stripping skeptic, for basically the reasons stated by Akhil Amar in this piece, but precedent since Ex parte McCardle (1868) has wholeheartedly confirmed the power.)

The through-line, I think, is that "judicial independence" means that Congress is not meant to be able to influence how the judges themselves behave. It has a lot of power to control when and how, and even whether, they are able to exercise their authority. But it cannot control how they use that authority, when given the opportunity. This is the position taken in Henry Hart's famous Dialogue about federal jurisdiction, and I think it's absolutely essential for the Court to perform its legitimating function (see my Liberal Currents piece for more on this). People need to know that, when a judge votes to uphold a federal statute, that's because they actually think the law is constitutional, not because they're worried Congress is looking over their shoulder.

Okay, so "good Behaviour" means life tenure. But so what? There are other ways, people contend, to effectively establish term limits. Congress gets to set the size of the Supreme Court. So why couldn't it say that, for instance, each seat on the Court is abolished after it has been occupied for 18 years, and a new one created? Or that the Court will consist of the nine most recently appointed Justices, with a new appointment made every two years? The older Justices would still "hold their office" in some technical sense, but wouldn't get to vote on cases – except perhaps to pinch-hit for other Justices who had to recuse from a particular case.

There are a few problems with this. One is the extremely consistent historical practice that federal courts are only shrunk by attrition. That is to say, Congress can specify that the next vacancy that opens up on a particular court will not be filled, thereby shrinking the court. But it cannot remove existing members of a court as a means of shrinking it. At the Supreme Court level this was last done when Andrew Johnson was president: Congress shrank the Court by attrition from 10 to 7 so as to deny Johnson any appointments, and then once Grant got into office it went back up to 9, where it has remained to this day. The outgoing Federalist Congress in 1801 had also specified that the Court would shrink from 6 to 5, denying Jefferson an appointment, though the Jeffersonians reversed that before it could go into effect.

At the lower court level things get a little bit wonky, because Congress can just abolish whole courts at a time. "The Fifth Circuit? What Fifth Circuit, there is no Fifth Circuit anymore." What happens to the judges on an abolished court is an interesting question. When Congress abolished the old "circuit courts" and created the modern "courts of appeals," which we all still call circuit courts but that's slightly inapt, all of the existing circuit judges were carried over to the new appellate courts. I tend to think that reassigning these judges is required, since they have a constitutional entitlement to their offices. Of course you can reassign them to the United States Court of Claims So Small You Can't See 'Em With An Electron Microscope that sits in northern Alaska. But you can't just disappear their commissions altogether.

In any case, this doesn't apply to the Supreme Court, which Congress did not create and cannot abolish. That same fact rules out any scheme along the lines of, okay, you're still a Justice, you just can't sit on cases anymore. Congress can tell the circuit courts how to operate, because it created those courts from whole cloth, but what "one supreme Court" means is a question of constitutional law, and the unbroken practice says that every member of the Supreme Court sits on every case. This, again, relates to the very reason why we have a Supreme Court, which is to harmonize the law. Because the same body has the final word on all questions of federal law, every court in America is bound to follow that understanding of the law, i.e. the same understanding as every other court. This avoids the situation where the meaning of federal law varies from one state to the next. Letting Congress tell the Supreme Court to sit in panels or whatever interferes with that uniformity.

There's also the matter of the whole "fraud on the Constitution" thing. If Congress could stop Supreme Court Justices from being able to vote on cases, that would give it a lever of influence not at all unlike what it could do by threatening to cut the Justices' salaries. Of course the thing is that the people proposing this are effectively trying to do a fraud on the Constitution; the feeling is that the constitutional policy of judicial independence and life tenure has become an obstacle to our desired goals and we are looking for clever ways to get around it. But that's not good-faith constitutional interpretation, and the Constitution is by no means obliged to just let you come up with clever workarounds.

I relied a bunch in this last section on the idea of historical practice, so it's worth saying a few words about how to treat that kind of practice. Essentially I view it as the same species of thing as judicial precedent. It's not the same thing as argument from original understanding, a.k.a. "historical argument;" that is exclusively concerned with the state of the world prior to and at the moment of ratification. (Sometimes post-ratification precedent can illuminate the pre-ratification understanding, but that's a separate maneuver.) Now, precedent can be overruled sometimes! But when you're dealing with a long, unbroken historical practice, you really shouldn't do that lightly. There's an awful lot of stuff where I think the unbroken practice is probably wrong as a matter of first principles (jurisdiction stripping, say!) but I kind of accept that it's just settled at this point. If you want to try to overrule the historic understanding of judicial tenure, you better have a damn good argument that it's wrong. And there just isn't one.

So, judicial term limits require a constitutional amendment. I tend to think that such an amendment should be a broader regularization of Article III, although to be clear that's not without some cost. The degree of power that Congress has over the Court helps to maintain the link of popular sanction for the Court's business. But l'Affair Merrick Garland in 2016 really highlighted how vulnerable to abuse the nomination process is, and I think an amendment fixing the size of the Court and providing for term limits and regular appointments also needs to come up with some way to prevent similar abuses going forward. I don't know exactly what that should be; it's not a trivial problem, because someone has to have the final word, and I don't just want to cut Congress out of the process altogether. But that's the work that's cut out for us.

Next up: this notion of an enforceable ethics code! I confess that I'm not altogether certain what kind of enforcement mechanism Biden has in mind. The existing judicial code of conduct, so far as I can determine, does not specify a means of enforcement. The judicial conference has a certain amount of power to discipline the lower court judges, removing their caseloads and the like. I don't know that those powers could be extended to the Supreme Court; again, the fact that the Court is created directly by the Constitution matters a lot here.

I've seen suggestions that the ethics code should be understood as a construction of the "good Behaviour" clause, with the penalty for violation being automatic removal from office. That's not correct, for reasons more or less already stated. The only constitutionally sanctioned mechanism for removing a judge for bad Behaviour is impeachment. You can't vest any other body with the power to determine that a judge has behaved badly and hence should be removed from office. You just can't. Also, except in its capacity as a court of impeachment, Congress has no power to define what "good Behaviour" means. Such a power would be wildly open to abuse and would undermine the constitutional policy of judicial independence.

So what can you do? A few things. First of all, you can make ethics violations into crimes! We already do this! Bribery is a crime! The criminal penalty does not and cannot include removal from office, but it can include, you know, imprisonment, or fines. And it can be enforced by Department of Justice investigation and prosecution.

Speaking of investigation, a thing Congress could easily do is create an oversight/watchdog body to monitor the Court's ethics. This body would not be able to remove the Justices from office for violations, but it could make impeachment referrals to Congress when it discovers sufficiently severe problems. And presumably the mere existence of such a body would have a strong in terrorem deterrent effect.

Now, the impeachment mechanism is currently broken. In my view this is an enormous problem with the entire constitutional structure. I'm honestly not really sure what to do in the absence of a functional impeachment process. It's actually true that this is the remedy the constitution prescribes for this stuff, and that it often precludes any other kind of remedy. It's also true that there are good reasons for that! Removal from office is a big deal, and it's also a political matter. There's a reason we don't automatically remove presidents from office when they're convicted of crimes, for instance: that would invite political prosecutions as a device to change control of the White House. Congress has a unique authority to speak for the Nation in these immensely sensitive matters. But Congress itself is broken. Which... makes the whole Constitution kind of non-functional. But that's a much bigger problem; we're here to talk about court reform, so I'm going to basically take the existence of the impeachment mechanism as a given.

There's another question worth asking about ethics reform specifically: can Congress impose a statutory recusal mechanism on the Court? Right now, the decision that a Justice ought to recuse from a particular case is left to..... that Justice themselves. No review, not even by the full Court. This approach depends utterly on the basic integrity of the Justices, which obviously we can't trust anymore.

I think it's possible that Congress could provide by law that motions to recuse must be heard by the entire Court, perhaps even that the Justice in question can't vote on the motion to recuse and that a tie among the other Justices means the motion to recuse is granted. It also might be possible to create, say, a body of appellate judges separate from the Supreme Court that could hear these motions? I dunno. Again the prospect for abuse looms large here. We can imagine a spurious recusal motion designed to change a 5-4 case to a 4-4 split, for instance. But it's not immediately obvious to me that this sort of thing would be unconstitutional.

Long story short, then, there's plenty you can do in this space, so long as you're a bit careful about what the enforcement mechanism is, i.e. not automatic removal from office.

Finally, we come to the proposed amendment abrogating Trump v. United States. As it happens I'm possibly the world expert on "redundant amendments," i.e. amendments that only enact what their supporters believe to be the correct construction of the Constitution as it stood previously. And I think Trump v. US is a good candidate for abrogation by amendment! But, as was done with Dred Scott and the Fourteenth Amendment, I think the amendment in question shouldn't just be tailored to the holding of the case being abrogated, which is what Biden proposes.

I would rather have a broader "republicanism" amendment about placing the government firmly under law. This can sweep in a handful of thematically related issues that prevent government accountability:

• Qualified immunity. This is the doctrine, originating in Harlow v. Fitzgerald (1982), that says that most government officials (e.g. police officers) are liable for constitutional violations only when they violate "clearly established" rights. This has evolved into an absolute monstrosity that effectively says that if the local federal court of appeals hasn't had a case on identical facts before, you can't get damages. It was never very clear whether this was a constitutional rule or just a construction of 42 U.S.C. § 1983, or what. After Trump v. U.S., which is plainly a constitutional case, I think that's been cleared up. I favored abrogating qualified immunity by statute before Trump v. U.S., and I favor abrogating it by constitutional amendment now. You can do all of this at once by specifying the "ultra vires" standard for claims of official immunity. As I discuss in my post on Trump v. U.S., this means that immunity extends exactly as far as the officer's actual legal authority to act, and not an inch further.
• Causes of action for constitutional violations. In Marbury v. Madison, John Marshall famously proclaimed that "for every right, there is a remedy." Current practice makes mockery of that idea. The Court is increasingly hostile to so-called "implied causes of action." A cause of action is basically a legally-cognizable claim that you can sue someone for: breach of contract, for example, is a cause of action, or negligence, or products liability or whatever. But what if there's a statute that says "X is unlawful" but doesn't specifically say that anyone can sue over it? That's what implied causes of action are for, and the courts have become increasingly unwilling to find them in statutory cases. Sure, whatever, I think that's probably wrong but if Congress wants to make a cause of action it can always do so explicitly.
  But that attitude has also extended to constitutional cases. There's a 1971 case called Bivens that found an implied cause of action for a Fourth Amendment violation. (Getting a bit technical, the reason it mattered in Bivens is that the property owner consented to a search conducted under a claim of legal authority that was later found to be invalid. The consent means it wasn't technically common law trespass, so if there isn't a claim for the constitutional violation itself then there's just no remedy.) But in the intervening fifty years the Court has narrowed Bivens to the point where it basically only stands for the proposition that Walter Bivens had a remedy that one time.
  The correct answer, of course, is "Bivens for literally everything." This is already the rule for constitutional violations at the state level, under 42 U.S.C. § 1983, which was adopted in the immediate wake of the Civil War to enforce the Fourteenth Amendment. We should put that in the Constitution and make it apply to state and federal officers alike.
• Sovereign immunity. Both the federal and state governments in our system enjoy this privilege, which means that the government itself cannot be sued for money damages (except by its own consent). State sovereign immunity is associated with the Eleventh Amendment, although Akhil Amar among others has argued that the Supreme Court has been basically wrong in construing the Amendment this way. (Amar's view can be found in Part II of this long and excellent article.) Federal sovereign immunity doesn't rest on any express provision of any written law, but has been recognized by the Supreme Court basically without exception; United States v. Lee (1882), a case amusingly about the federal government's seizure of Robert E. Lee's Arlington estates to make what is now Arlington National Cemetery, is a good early demonstration.
  This whole thing is nonsense and unrepublican. As Amar says in the linked article, and as many others have said, in a republic it is the People who are sovereign, not the government. The Eleventh Amendment ought to be repealed and the principles of sovereign immunity repudiated more broadly. In fact I think there's an argument to be made that the Bivens remedies I alluded to in the previous bullet should just run against the government as a corporate entity, cutting the official (who doesn't have deep pockets and will be indemnified anyway) out of the loop altogether.

Put all of this together and you get something like the following proposed amendment:

§ 1. Whereas the People of the United States are sovereign, and not the government of the United States or of any state, the eleventh article of amendment to the Constitution is hereby repealed. Both the United States and each of the several states shall be amenable to suit in the courts of the United States.
§ 2. Every person who, under color of the laws of the United States or of any state, shall deprive any person of any rights, privileges, or immunities secured by this Constitution shall be liable to the injured party.
Both the United States and the several states shall be liable to any party deprived of any rights, privileges or immunities secured by this Constitution by any officer thereof.
§ 3. Every officer of the United States or of any state who, acting beyond the scope of their lawful authority, commits any civil or criminal offense under the laws of the United States or of any state shall be fully subject to legal sanction for that offense.