The most-awaited decision left in this Supreme Court term is … still awaited after this morning. We will not find out about whether the EPA’s governing statute allows it to treat carbon as a pollutant under the Clean Air Act until tomorrow, when the court will publish its last two opinions from this term. Today, however, the court released what may be Justice Stephen Breyer’s valedictory decision, and gutted one of Justice Neil Gorsuch’s signature cases.

In Torres v Texas Department of Public Safety, the issue of sovereign immunity for states ran up against the constitutional authority of the federal government to field the US military. A Texas state trooper, Le Roy Torres, suffered injuries and maladies related to his reserve service in Iraq. When he returned home, those maladies proved too burdensome for patrol duty, but Texas DPS refused to employ him elsewhere. In a 5-4 decision, Breyer concluded that state sovereign immunity is inferior to the federal authority on national defense, and that the Uniformed Services Employment and Reemployment Rights Act of 1994 (USERRA) validly burdened the state to accommodate Torres:

The Constitution vests in Congress the power “[t]o raise and support Armies” and “[t]o provide and maintain a Navy.” Art. I, §8, cls. 1, 12–13. Pursuant to that authority, Congress enacted a federal law that gives returning veterans the right to reclaim their prior jobs with state employers and authorizes suit if those employers refuse to accommodate them. See Uniformed Services Employment and Reemployment Rights Act of 1994 (USERRA), 38 U. S. C. §4301 et seq. This case asks whether States may invoke sovereign immunity as a legal defense to block such suits.

In our view, they cannot. Upon entering the Union, the States implicitly agreed that their sovereignty would yield to federal policy to build and keep a national military. States thus gave up their immunity from congressionally authorized suits pursuant to the “‘plan of the Convention,’” as part of “‘the structure of the original Constitution itself.’” PennEast Pipeline Co. v. New Jersey, 594 U. S. ___, ___ (2021) (slip op., at 14) (quoting Alden v. Maine, 527 U. S. 706, 728 (1999)). …

PennEast defined the test for structural waiver as whether the federal power at issue is “complete in itself, and the States consented to the exercise of that power—in its entirety—in the plan of the Convention.” 594 U. S., at ___ (slip op., at 22) (internal quotation marks and citation omitted). Where that is so, the States implicitly agreed that their sovereignty “would yield to that of the Federal Government ‘so far as is necessary to the enjoyment of the powers conferred upon it by the Constitution.’” Id., at ___ (slip op., at 16) (quoting Kohl v. United States, 91 U. S. 367, 372 (1876)). By committing not to “thwart” or frustrate federal policy, the States accepted upon ratification that their “consent,” including to suit, could “never be a condition precedent to” Congress’ chosen exercise of its authority. 594 U. S., at ___, ___ (slip op., at 8, 10) (internal quotation marks omitted). The States simply “have no immunity left to waive or abrogate.”

It’s worth noting that PennEast got decided after Torres filed suit against Texas. While this is a very personal and important ruling for Torres himself and similarly situated veterans, the reach of this case is somewhat limited and its character mainly technical. At least in Torres’ case, it appears specific justice is done with this decision. However, Justice Clarence Thomas writes in his dissent that Torres and the recent PennEast decisions do significant damage to the constitutional order:

More than two decades ago, this Court found it “difficult to conceive that the Constitution would have been adopted if it had been understood to strip the States of immunity from suit in their own courts and cede to the Federal Government a power to subject nonconsenting States to private suits in these fora.” Alden v. Maine, 527 U. S. 706, 743 (1999). Accordingly, we held—without qualification—that “the powers delegated to Congress under Article I of the United States Constitution do not include the power to subject nonconsenting States to private suits for damages in state courts.” Id., at 712 (emphasis added).

No longer. Today, by adopting contrived interpretations of Alden and the recent decision in PennEast Pipeline Co. v. New Jersey, 594 U. S. ___ (2021), the Court holds that at least two (and perhaps more) Article I “war powers” do, in fact, include “the power to subject nonconsenting States to private suits for damages in state courts,” Alden, 527 U. S., at 712, and that Congress has exercised that power by enacting the Uniformed Services Employment and Reemployment Rights Act of 1994 (USERRA), 38 U. S. C. §4301 et seq. Alden should have squarely foreclosed that holding. As the Court there already explained, constitutional text, history, and precedent all show that when the States ratified the Constitution, they did not implicitly consent to private damages actions filed in their own courts—whether authorized by Congress’ war powers or any other Article I power. Because the Court today holds otherwise, I respectfully dissent. …

During the Nation’s first 200 years, this Court recognized only two instances in which the States had surrendered their sovereign immunity in the constitutional plan, both of which involved suits prosecuted by other sovereigns. The States had agreed to be sued by other States in this Court, see Principality of Monaco v. Mississippi, 292 U. S. 313, 328 (1934), and by the United States in federal court, see United States v. Texas, 143 U. S. 621, 644–645 (1892); Franchise Tax Bd., 587 U. S., at ___–___ (slip op., at 9–10). When it came to private litigation, however, this Court long maintained that “the Framers thought it an impermissible affront to a State’s dignity to be required to answer the complaints of private parties in federal courts,” Federal Maritime Comm’n, 535 U. S., at 760, and that “the Convention did not disturb States’ immunity from private suits,” id., at 752.

What about the USERRA statute? Clearly, Congress intended for returning veterans to have some sort of potential action against states. Thomas agrees, but points out that Congress never included any language that binds non-consenting states to waive sovereign immunity:

To be clear, I am not disputing whether USERRA speaks clearly enough to express a congressional intent to “abrogate” the States’ sovereign immunity in their own courts; plan-of-the-Convention waiver asks whether the States surrendered that immunity when the Constitution was ratified and thus “agreed . . . not to assert that immunity” in particular contexts. Katz, 546 U. S., at 373. But even if the Constitution itself partially strips state sovereign immunity, it would still fall to Congress to decide whether, and on what terms, to render States amenable to suit, or to permit States to assert immunity. Cf. id., at 379 (“Congress may, at its option, either treat States in the same way as other creditors insofar as concerns ‘Laws on the subject of Bankruptcies’ or exempt them from the operation of such laws”).

The Court should not casually consider the constitutionality of USERRA’s supposed subjection of nonconsenting States to damages actions in state court when it is not clear the statute does any such thing. By doing so, the Court gives short shrift to the “well-established principle governing the prudent exercise of this Court’s jurisdiction that normally the Court will not decide a constitutional question if there is some other ground upon which to dispose of the case.” Northwest Austin Municipal Util. Dist. No. One v. Holder, 557 U. S. 193, 205 (2009) (internal quotation marks omitted).

Again, this is mostly technical, but it again points up the problem with poorly written statutes and cases that fall into the gaps they create. Thomas clearly would prefer that the court kick this back to Congress to fix rather than continue to reverse its 200-year track record on sovereign immunity.

The other case released today gutted Gorsuch’s blockbuster ruling in McGirt. On that case, Gorsuch reimposed the terms of a federal treaty with Oklahoma’s Indian tribes that prevented the state from prosecuting crimes that fell within clearly delineated Native lands — including Tulsa. Today, however, Justice Brett Kavanaugh writes for a 5-4 majority in Oklahoma v Castro-Huerta that both the federal and state governments have jurisdiction to enforce the law, the latter only subject to whether their jurisdiction has been “preempted” in some specific manner:

To begin with, the Constitution allows a State to exercise jurisdiction in Indian country. Indian country is part of the State, not separate from the State. To be sure, under this Court’s precedents, federal law may preempt that state jurisdiction in certain circumstances. But otherwise, as a matter of state sovereignty, a State has jurisdiction over all of its territory, including Indian country. See U. S. Const., Amdt. 10. As this Court has phrased it, a State is generally “entitled to the sovereignty and jurisdiction over all the territory within her limits.” Lessee of Pollard v. Hagan, 3 How. 212, 228 (1845).

In the early years of the Republic, the Federal Government sometimes treated Indian country as separate from state territory—in the same way that, for example, New Jersey is separate from New York. Most prominently, in the 1832 decision in Worcester v. Georgia, 6 Pet. 515, 561, this Court held that Georgia state law had no force in the Cherokee Nation because the Cherokee Nation “is a distinct community occupying its own territory.”

But the “general notion drawn from Chief Justice Marshall’s opinion in Worcester v. Georgia” “has yielded to closer analysis.” Organized Village of Kake v. Egan, 369 U. S. 60, 72 (1962). “By 1880 the Court no longer viewed reservations as distinct nations.” Ibid. Since the latter half of the 1800s, the Court has consistently and explicitly held that Indian reservations are “part of the surrounding State” and subject to the State’s jurisdiction “except as forbidden by federal law.” Ibid.

That at least limits the reach of McGirt. Gorsuch’s lengthy dissent — 42 pages, 17 pages longer than the opinion — makes it clear that he feels that the court just gutted it. And in doing so, Gorsuch accuses the majority of intellectual dishonesty:

It acknowledges that, at the Nation’s founding, tribal sovereignty precluded States from prosecuting crimes on tribal lands by or against tribal members without congressional authorization. See ante, at 5. But the Court suggests this traditional “‘notion’” flipped 180 degrees sometime in “the latter half of the 1800s.” Ante, at 5, 21. Since then, the Court says, Oklahoma has enjoyed the “inherent” power to try at least crimes by non-Indians against tribal members on tribal reservations until and unless Congress preempts state authority.

But exactly when and how did this change happen? The Court never explains. Instead, the Court seeks to cast blame for its ruling on a grab bag of decisions issued by our predecessors. But the failure of that effort is transparent. Start with McBratney, which the Court describes as our “leading case in the criminal context.” Ante, at 6. There, as we have seen, the Court said that States admitted to the Union may gain the right to prosecute cases involving only non-Indians on tribal lands, but they do not gain any inherent right to punish “crimes committed by or against Indians” on tribal lands. McBratney, 104 U. S., at 624. The Court’s reliance on Draper fares no better, for that case issued a similar disclaimer. See 164 U. S., at 247. Tellingly, not even Oklahoma thinks McBratney and Draper compel a ruling in its favor. See Brief for Petitioner 12. And if anything, the Court’s invocation of Donnelly, 228 U. S. 243, is more baffling still. Ante, at 14, n. 3. There, the Court once more reaffirmed the rule that “offenses committed by or against Indians” on tribal lands remain subject to federal, not state, jurisdiction. Donnelly, 228 U. S., at 271; see also Ramsey, 271 U. S., at 469.

That leaves the Court to assemble a string of carefully curated snippets—a clause here, a sentence there—from six decisions out of the galaxy of this Court’s Indian law jurisprudence. Ante, at 5–6. But this collection of cases is no more at fault for the Court’s decision than the last. Organized Village of Kake v. Egan—which the Court seems to think is some magic bullet, see ante, at 5, 14, n. 2, 21, 22– 24—addressed the prosaic question whether Alaska could apply its fishing laws on lands owned by a native Alaska tribal corporation. 369 U. S. 60, 61–63 (1962); see also n. 5, supra. Subsequently, the Court cabined that case to circumstances “dealing with Indians who have left or never inhabited reservations set aside for their exclusive use or who do not possess the usual accoutrements of tribal selfgovernment.” McClanahan, 411 U. S., at 167–168. Meanwhile, New York ex rel. Cutler v. Dibble allowed New York to use civil proceedings to eject non-Indian trespassers on Indian lands. 21 How. 366, 369–371 (1859). In Surplus Trading Co. v. Cook, the crime at issue did not take place on tribal lands but on a “supply station of the United States” sold by Arkansas to the federal government. 281 U. S. 647, 649 (1930). In New York ex rel. Ray v. Martin, this Court merely reaffirmed McBratney and held that States could exercise jurisdiction over crimes involving only non-Indians. 326 U. S. 496, 499–500 (1946). Both County of Yakima v. Confederated Tribes and Bands of Yakima Nation and Nevada v. Hicks issued holdings about state civil jurisdiction, not criminal jurisdiction striking at the heart of tribal sovereignty. See 502 U. S. 251, 256–258, 270 (1992); 533 U. S. 353, 361, 363, 374 (2001).

In the end, the Court cannot fault our predecessors for today’s decision. The blame belongs only with this Court here and now. Standing before us is a mountain of statutes and precedents making plain that Oklahoma possesses no authority to prosecute crimes against tribal members on tribal reservations until it amends its laws and wins tribal consent. This Court may choose to ignore Congress’s statutes and the Nation’s treaties, but it has no power to negate them. The Court may choose to disregard our precedents, but it does not purport to overrule a single one. As a result, today’s decision surely marks an embarrassing new entry into the anticanon of Indian law. But its mistakes need not—and should not—be repeated.

In response, Kavanaugh gets almost as chippy:

From start to finish, the dissent employs extraordinary rhetoric in articulating its deeply held policy views about what Indian law should be. The dissent goes so far as to draft a proposed statute for Congress. But this Court’s proper role under Article III of the Constitution is to declare what the law is, not what we think the law should be. The dissent’s views about the jurisdictional question presented in this case are contrary to this Court’s precedents and to the laws enacted by Congress.

Food fight on the Right! It seems that tempers may have worn more thin on Castro-Huerta than they did on Dobbs, if such a thing is possible. The upshot is that the massive shift of McGirt is at the very least heavily circumscribed if not entirely reversed. Oklahoma and other states where larger population centers exist within treaty boundaries will heave a sigh of relief over the apparent easing of obstacles that McGirt erected in criminal prosecutions of non-Indians on Indian territory, but Gorsuch’s poetic derision will likely sting for a long time to come.

The Supreme Court added another decision day for tomorrow. Only two cases remain — the much anticipated West Virginia v EPA and Biden v Texas, in which the Remain in Mexico policy hangs in the balance. Looks like we’ll end with a couple of bangs rather than a whimper after all.