On July 30, the U.S. Court of Appeals for the Fifth Circuit ruled that the district court should not have granted the United States a preliminary injunction ordering Texas to remove a barrier lying in the Rio Grande River. The case was United States v. Abbott, and it was decided on the issue of navigability. However, the case also has implications for states' power to wage defensive war—and particularly defensive war against illegal immigration.

Andrew T. Hyman and I recently published an examination of those issues in the British Journal of American Legal Studies. We focused mostly on Founding-era evidence of the kind probative of the Constitution's original meaning. Our article played a role in the case—but, as described below, a rather unusual one.

As the post recounts, Texas relied in part on the state's power to repel invasions, which the majority did not reach but which Judge Ho discussed in concurrence: 

Judge James C. Ho wrote a concurring-and-dissenting opinion focusing on the state right of self defense. He argued that the U.S. government's request for a preliminary injunction should have been dismissed because when a state, in good faith, claims it has been invaded and invokes its war powers, the legality of its decisions are non-justiciable political questions:

Supreme Court precedent and longstanding Executive Branch practice confirm that, when a President decides to use military force, that's a nonjusticiable political question not susceptible to judicial reversal. I see no principled basis for treating such authority differently when it's invoked by a Governor rather than by a President. If anything, a State's authority to "engage in War" in response to invasion "without the Consent of Congress" is even more textually explicit than the President's.

In Judge Ho's view, however, "good faith" decision making is a prerequisite to non-justiciability. In this respect and in some other respects, his analysis is similar to ours. We wrote:

"Insurrection" and "invasion" not only trigger the federal government's duty under the [Guarantee] Clause, but also trigger exercise of state war powers. If the terms are too vague for courts to define for federal purposes, then they also are too vague for courts to define for state purposes. If [Guarantee] Clause cases are held to be non-justiciable because the Constitution commits the decision of whether and how to protect states against invasion to the political branches of the federal government, then the Constitution even more clearly commits (as demonstrated by the Self-Defense Clause) the determination of whether a state has been "Invaded" or in "imminent Danger" to the state government.  If redressibility issues impede justiciability in [Guarantee] Clause cases, then they could also impede justiciability when a state has gone onto a war footing and raised an army.

To be clear: If federal officials are proceeding in good faith to crush an insurrection or repel an invasion, the courts should not second-guess their tactics.  But judicial intervention is appropriate when federal officials utterly neglect their duty or adopt measures so plainly insufficient as to demonstrate a lack of good faith effort.

Judge Ho's concurrence matched our conclusions in another respect as well: Both he and we doubted whether a federal law, even if clearly contradicting the right to state self-defense, could take priority over that right. ("[F]ederal statutes," he wrote, "ordinarily must give way to federal constitutional rights.") This makes sense: Self-defense is inherent in sovereignty, and the Supreme Court has defended less important aspects of state sovereignty from otherwise-valid congressional action. Examples include the protection of a state's decision on where to locate its capital and protection of state officials from federal "commandeering."

And on the role of the Natelson/Hyman article: 

Judge Dana M. Douglas's dissent challenged the majority's evidentiary conclusions on navigability, maintaining that the federal government had presented ample evidence that the Rio Grande near Eagle Pass qualified as navigable. She also concluded that once Congress has an opportunity to respond to an invasion, state war powers cease:

Clause 3 provides that a state may engage in war without consent of Congress only when it is "actually invaded, or in such imminent Danger as will not admit of delay." . . .  See, e.g., Articles of Confederation of 1781, art. VI, para. 5 (limiting a state's power to engage in war "till the united states in congress assembled can be consulted"); Robert G. Natelson & Andrew T. Hyman, The Constitution, Invasion, Immigration, and the War Powers of States, 13 Brit. J. Am. Legal Stud. 1, 17 (2024) (noting that, in regard to state war powers, the Constitution resulted in "a balance between federal and state prerogatives roughly similar to that under the Articles of Confederation") . . . .

In other words, because the scope of state war power under the Constitution is roughly equal to the scope under the Articles, and because the Articles required consultation and/or consent by Congress, then state war power under the Constitution is similarly limited.

Unfortunately, the publication she relied on—ours—directly contradicted her conclusions. We wrote that under the Articles of Confederation, states "retained virtually unlimited flexibility to engage in defensive land war—even after Congress had been consulted—except for power to strike pre-emptively at non-Indian enemies."

More importantly, we found that the Constitution had removed the Articles' constraints on state defensive war:

[O]n the land side, the Constitution preserved general state control over their militias while providing that "No State shall, without the Consent of Congress . . . keep Troops . . . in time of Peace . . . or engage in War, unless actually invaded, or in such imminent Danger as will not admit of delay." This limitation omitted the Articles' contingent requirement of consultation with Congress. (Italics added.)

We have written to Judge Douglas advising her of the discrepancy.