Interlude on Jurisdiction Stripping

Today the Senate Democrats introduced what they style the "No Kings Act," which purports to overrule Trump v. United States. Now, you may be saying, wait a minute, Trump v. U.S. was a constitutional case, how can Congress overrule it by ordinary legislation? The answer is that the bill contains a jurisdiction-stripping provision that prevents the Supreme Court from hearing any appeal of the conviction of a former president. Basically everyone on my Bluesky feed is very enthusiastic about this; I on the other hand hate it with a passion. So let's talk about jurisdiction stripping.

(I don't quite consider this a part of my series on How (Not) To Reform the Supreme Court, though it's certainly related.)

First, the basics: what is jurisdiction stripping? Where does it come from?

It comes from Article III, § 2, clause 2 (emphasis mine):

In all Cases affecting Ambassadors, other public Ministers and Consuls, and those in which a State shall be Party, the supreme Court shall have original Jurisdiction. In all the other Cases before mentioned, the supreme Court shall have appellate Jurisdiction, both as to Law and Fact, with such Exceptions, and under such Regulations as the Congress shall make.

"All the other Cases before mentioned" refers to the enumeration of federal jurisdiction in the first clause of § 2. It's not very important for our purposes what exactly the full scope of federal jurisdiction is. Also not exactly relevant is the last clause, about "Regulations." That's where you get, e.g., Congress's power to remove mandatory appeals to the Supreme Court and switch to the discretionary certiorari system used today. Regulations, in other words, don't change whether the Court has jurisdiction, only the details of how it exercises that jurisdiction.

The idea behind jurisdiction-stripping is that the "with such Exceptions" language permits Congress to say that, in some cases in which the Court would otherwise have appellate jurisdiction, it doesn't, actually. This power was expressly recognized by the Supreme Court itself in Ex parte McCardle (1868), a case in which it was feared that the Court would hold various aspects of Reconstruction unconstitutional and therefore Congress stripped the Supreme Court's jurisdiction as the case was being argued before it. But also, the Supreme Court didn't have general appellate jurisdiction over federal criminal cases until after the Civil War. So it's fair to say that there's a completely unbroken tradition of recognizing Congress's jurisdiction-stripping power since the early days of the Republic.

The text, however, could bear another reading. Just look at the structure of the clause. The first sentence lists certain cases in which the Court has original jurisdiction. The second sentence says that in all other cases – with such exceptions as Congress may make – the Court has appellate jurisdiction. Wouldn't it be kind of natural to read that as saying that, when Congress makes an "exception" to the appellate jurisdiction, it does so by shifting those cases to original jurisdiction? If that's what it means, then maybe there isn't a power of jurisdiction-stripping at all.

This construction is ruled out, however, by Marbury v. Madison (1803). These days we remember Marbury as the first case in which the Court struck down an Act of Congress as unconstitutional, and for its lengthy discussion of the proprietary of judicial review. But in order to reach that issue, of course, the Court (a.k.a. Chief Justice John Marshall) had to say that a specific federal statute conflicted with a particular part of the Constitution. Here's the actual holding of Marbury:

1. A provision of the Judiciary Act of 1789 authorizing the Supreme Court to issue writs of mandamus authorized people to bring an original action for mandamus in the Supreme Court, therefore expanding the Court's original jurisdiction.
2. The list of original jurisdiction cases in Article III, § 2, clause 2 is exhaustive, and Congress has no power to increase the Court's original jurisdiction.

The first part is pretty clearly wrong: that provision of the Judiciary Act was better construed as authorizing the Court to issue writs of mandamus in aid of its existing jurisdiction, not as creating a whole new category of case as part of the Court's original jurisdiction.

The second part is... at least debatable. Marshall very clearly went out of his way to create the constitutional conflict, which today we would consider poor form. (Where possible you're supposed to interpret a statute so as to avoid constitutional problems.) But once that holding is in place, there's no other way to construe the "Exceptions" language than as authorizing jurisdiction-stripping. So here we are.

So far I've been speaking entirely in terms of text and doctrine. The text can bear either meaning, and historical practice/precedent is entirely on one side of the issue. But if we want to assess whether that precedent is right, we need to delve a bit deeper – particularly because in Marbury itself, Marshall was pretty clearly using motivated reasoning to set up the occasion for judicial review. If his construction of the relevant language is right, it's not for any reason he himself gave us. (For what it's worth, Marshall's actual reasoning is that, if Congress had the power to determine the boundary between original and appellate jurisdiction, then the enumeration of original jurisdiction would be "mere surplusage." This is clearly wrong, since that enumeration could provide a minimum as easily as a maximum.)

Akhil Amar has argued that we should understand the Supreme Court's original jurisdiction as being essentially a venue issue. The Supreme Court would sit in the national capital; litigating a case there would require the parties to travel to the capital, which was no mean feat back in the late eighteenth century. The listed original jurisdiction cases are all ones in which hearing the case in the capital makes sense. Expanding original jurisdiction, however, would impose a great burden on litigants for whom this would be very inconvenient.

I've never been altogether persuaded by this. We can just as easily understand the grant of original jurisdiction as being a matter of respect, the respect due to foreign nations (and the agents thereof) and to the states. Amar takes pains to emphasize that all Article III judges are equally vested with the judicial power of the United States, which is correct. But you can still imagine a foreign power feeling a bit slighted that its minister was made to appear in some lesser court. We might also think that the Supreme Court would be somewhat more sensitive to the foreign policy nuances that would arise in these kinds of cases.

Even if Amar's geographical interpretation isn't super compelling, though, what's on the other side? What's wrong with the conventional interpretation, why is jurisdiction-stripping a problem? My argument here is (surprise, surprise) basically structural in nature. The function of the Supreme Court in the scheme of the American government is to harmonize the law, to ensure that (within reason) every court in the land is applying the same understanding of federal law. And jurisdiction-stripping greatly impedes the Court's ability to create this uniformity. This is expressly the point of the No Kings Act: the bill would openly invite lower federal courts to defy the Supreme Court's construction of federal law. This bothers me, and I think it would bother everyone else a lot more if the particular Supreme Court decision we were trying to nullify weren't itself so odious.

It's interesting to me that Amar isn't bothered by this. He has a whole Theory of federal jurisdiction that is all about the importance of having the federal courts control the construction of federal law. Thus, he says, whereas Congress is not obliged to give any federal court jurisdiction over the so-called "diversity" cases (where federal jurisdiction is premised on the fact that the parties come from different states), it must always give some federal court jurisdiction over all "federal question" cases. One nightmare scenario that comes to mind with jurisdiction-stripping is that Congress could remove the Supreme Court's ability to hear criminal appeals from the state courts, cutting off the main channel for constitutional review of state criminal practices. Amar has a solution to this problem, however, for he would say that, if Congress wanted to remove this jurisdiction from the Supreme Court, it would be obliged to place it in some other federal court.

Generally, however, Amar is comfortable with letting Congress make some other federal court the final authority on particular questions of federal law. I'm not, even though I very much agree with him that all federal judges have the federal judicial power in full measure. Because, well, the Article III vesting clause really does say "one supreme Court, and ... such inferior Courts as Congress may from time to time ordain and establish." This, it seems to me, is a plain statement that the court of final resort is to be the same in all cases. You could design a court system differently from that! Texas for example has separate courts of final resort for civil matters (the so-called Supreme Court of Texas) and for criminal cases (the Texas Court of Criminal Appeals). I take the "one supreme Court" language as ruling out any such scheme: what else could it really mean? If Congress by statute created a United States Court of Criminal Appeals, with power to review criminal law decisions by the federal courts of appeals and the state supreme courts and no possibility of review by the Supreme Court, could we really term such a court "inferior"? This seems to me a stretch.

Uniformity isn't the only issue here, as this example demonstrates. For the new Court of Criminal Appeals would be able to maintain legal uniformity in criminal matters throughout the whole country, just as the Supreme Court currently does. Letting Congress create new mini-supreme courts would also interfere with judicial independence. It would make the federal courts that much more a creature of Congress. Imagine that the Supreme Court had taken a view of criminal law matters that Congress disliked. This is not a hypothetical, of course: it basically happened during the Warren Court years. If Congress could create a rival Court of Criminal Appeals, it could then fill that court with its lackeys, chosen specifically for their more palatable views on criminal law. And it could potentially do this for every legal issue under the sun.

Amar is right to say that we should not identify the Supreme Court with the whole of the judicial branch. But the Supreme Court's supremacy plays a really important role in placing the federal judiciary on an independent and coequal footing. Every other federal court is essentially a creature of Congress, subject only to the limitation of life tenure. But final resort is always had to the one court that is not a creature of Congress. And because the final word rests with a single generalist court rather than a multiplicity of specialist courts, the appointment process cannot be overly particular; there's a limit on how much Supreme Court Justices can be chosen for their views on individual legal issues, because their views on all legal issues matter. (This is not, of course, to say they can't be chosen for their overall judicial philosophy; they clearly can, and this is good and important.)

You can try to defend against this kind of attack on the Supreme Court's supremacy even while recognizing the power of jurisdiction-stripping. It would be pretty plausible to say, for example, that a bill stripping all of the Supreme Court's appellate jurisdiction and vesting it in some new Pseudo-Supreme Court would not count as making "exceptions" to the appellate jurisdiction: the exceptions would have swallowed the rule. A doctrinal jurist like Oliver Wendell Holmes might prefer to just craft a test that would draw the line between valid "exceptions" and abuses of that power. The structural modality, which I favor, tends more toward absolutes. (I'll get around to writing about constitutional theory one of these days, I promise!)

But in any event, current law recognizes essentially no limits to the jurisdiction-stripping power, and would I think affirm Congress's authority to set up a Pseudo-Supreme Court. I think that's a problem, and I think there are real issues with jurisdiction-stripping that we should at the very least be mindful of.