Conservatives are inclined to celebrate and defend the Roberts Court, and especially its young supermajority, for its constitutionalist achievements. In one crucial area, however, what the Court’s jurisprudence needs is an exorcism — and a laborious one at that.

A quarter century or so ago, the Rehnquist Court propounded what was then called a “federalism revolution.” The Court sought to protect the “states as states” against federal overreach by reimposing modest limitations on the scope of the commerce clause; by immunizing states and their officials against litigation under federal statutes; and with various other, more arcane doctrines. Such state-versus-federal power struggles have subsided in the era of the Roberts Court. Perhaps the justices think that they have achieved all that can be achieved on this front without running headlong into the premises and precedents that ground the New Deal Constitution and the civil-rights state; perhaps the federalism questions of old have been crowded out by the separation-of-powers cases that now dominate the Court’s civil docket. To be sure, many of those controversies have a federalism overlay. When federal agencies get sued over immigration policy, student loans, health care, or climate change, state attorneys general usually play a lead role. Those cases, though, typically turn on the scope of executive power, not on federalism’s “vertical” distribution of powers between the states and the national government.

As a practical matter, the justices’ passivity on the “states’ rights” front may not matter much. In an era of Covid relief, the Inflation Reduction Act, and sundry other boondoggles, state and local governments function principally as recipients of federal largesse. While ruinous, the practice poses no legally cognizable threat to the “states as states.” In that regard, it is not a federalism issue to feel morose about.

Not so with the Roberts Court’s excursions into federalism’s “horizontal,” state-to-state dimension. It cannot have escaped judicial notice that red and blue states form cohesive political blocs, fiercely divided on any salient issue you can think of. Increasingly, the controversies wind up in the Supreme Court. One would hope for a judicial supply of sensible, constitutionally grounded rules to dampen sectarian litigiousness and to keep the states out of one another’s hair. The Court, however, has failed to supply a single rule of that kind; and in important respects, it has made matters worse. Two landmark decisions, both handed down in the Court’s 2022–23 term, illustrate the point.

In National Pork Producers Council v. Ross, a splintered majority upheld a California statute prohibiting the sale of pork products in the state unless the animals had been raised under conditions deemed sufficiently humane by the state. And in Mallory v. Norfolk Southern, a narrow majority held that any corporation that registers to do business in a state thereby consents to jurisdiction in that state even if neither the plaintiff, nor the defendant, nor for that matter the injury had anything else to do with that jurisdiction. In the case at bar, a Virginia railroad worker with temporary residence in Ohio sued a Virginia company over injuries sustained in Virginia. Naturally, he sued in Pennsylvania, where the trial lawyers own the local courts and Norfolk Southern is registered to do business. No way, said Pennsylvania’s supreme court. Way, said the U.S. Supreme Court.

Both Pork Producers and Mallory were wholly gratuitous. They wantonly erased perfectly sensible, time-honored legal distinctions. They knowingly courted interstate strife and economic mayhem and then said, “So what?”

At issue in Pork Producers was the so-called dormant commerce clause. That doctrine, which has been with us in various permutations since 1824, holds that the Constitution’s commerce clause, of its own force and without any congressional legislation, prohibits certain state interferences with interstate commerce. The justices still agree that the doctrine forbids patently discriminatory state acts that treat outsiders worse than citizens. California’s statute, however, equally applied to the Golden State producers who account for a minuscule 0.23 percent of the state’s pork supply. The key question, then, was whether California had impermissibly regulated on an “extraterritorial” basis. That inquiry, Justice Gorsuch’s principal opinion explained, has no place under the dormant commerce clause. Any state regulation has some extraterritorial effect; and we the justices decline to engage a standardless how-much-is-too-much inquiry.

That, frankly, is deliberately obtuse. It is one thing for a state to regulate, on a nondiscriminatory basis and in the absence of a preemptive federal statute, the quality of a product (as healthful or safe); that’s just the police power. It is a very different thing to regulate the method by which things are produced elsewhere. Laws of that sort have nothing to do with the product itself or with the health, safety, or welfare of the state’s citizens. They regulate conduct in other states — or, as the Supreme Court used to say in better days, “commerce as such.” The line between regulating products and regulating production is intuitive, constitutionally grounded, supported by precedent, well recognized in international trade law, and judicially manageable. Yet not one of the several opinions in Pork Producers even adverts to it.

As for Mallory: In a trilogy of cases preceding that decision, the Supreme Court had gravitated to the position that corporations could be sued over anything they might have done only in their “home,” meaning their state of incorporation or their principal place of business. For lawsuits filed elsewhere, you’d need “specific” jurisdiction — that is, some set of facts that connects the plaintiff, the defendant, and the forum to the events at issue. Those were the rules until Mallory held that corporations are “at home” in any state where they are registered, meaning all of them.

Behold what the decisions mean in real life. Per Pork Producers, each state is free to prohibit the in-state sale of goods or services produced under purportedly offensive conditions — in a state that prohibits (or permits, as the case may be) abortion; in a state with (or without) right-to-work or minimum-wage laws; in a state with (or without) “sanctuary” jurisdictions. Per Mallory, to repeat, any corporation can be sued over practically anything, in any state that decides so to extend its jurisdiction. For folks who produce stuff that people may want to buy, the message is the same: Click your heels; go home to Kansas and stay there. Or else dance to the plaintiff bar’s or Gavin Newsom’s fiddle.

Things may not turn out as bad as all that. States and the constituencies that stand to gain from an anything-goes environment may not all want to push the outer limits all the time. The question, though, is whether a legal regime should tolerate state-to-state imperialism, aggression, and exploitation, let alone incentivize it.

From a certain vantage, the answer might actually be yes. There are only two ways, the thinking goes, of bringing corporate behemoths to heel: subject them to centralized control (including, at the limit, outright nationalization); or else effectively compel compliance with the most restrictive state regime in which a firm’s products might end up. Since the former is impossible in the U.S., it’s got to be the latter. This was the explicit program of the architects of the New Deal Constitution, foremost Justice Louis Brandeis. A “federalism” of embargoes and universal state jurisdiction, they realized, was the only socialism that could be had under our notoriously fragmented institutional system. This was what Brandeis meant when he celebrated states as “laboratories of democracy.” It was the intended effect of Brandeis’s legendary opinion in Erie Railroad (1938), which held that in “diversity” cases that pit one state’s citizens against another’s, federal courts are bound by the common law of the plaintiff’s home state: Incentivize states to exploit one another’s citizens and to home-cook out-of-state defendants, and disable the federal courts from doing anything about it.

Pork Producers and Mallory are the apotheosis of that regime, and they are but the most glaring exhibits of an entire jurisprudence that pushes in that direction. May a state impose tax-collection obligations on entities whose only connection to the taxing jurisdiction is a website that is accessible to buyers in that state? Sure (Direct Marketing Association v. Brohl, 2014). May a state send its agents hither and yon to enforce its unilateral mandates? No prob (Brohl, Pork Producers). And so on, to consumer class actions and products liability under exotic state-law doctrines and, more recently, climate-change “public nuisance” suits against energy producers under state law, in state courts.

This corpus juris is not a product of undercover socialists on the federal bench. For the most part, it is an originalist product — more precisely, the product and indeed the project of a particular brand of originalism. Its principal proponents are Justice Gorsuch, author of the lead opinions in Pork Producers and Mallory, and Justice Clarence Thomas, who joined those opinions in full and has forcefully articulated very similar arguments in other cases.

The obvious point that ostensibly originalist interpretive commitments in this context threaten real-world havoc cuts no ice. (From a certain dogmatic perspective, absurd results just show how principled originalists are.) Perhaps, though, there is still room for objections of a loftier kind.

Whatever else one may want to pack into “the rule of law,” one of its essential attributes is a framework to govern what belongs to whom. Without reliable distinctions between mine and thine — as well as rules of exchange, prohibitions against procuring by force or fraud what belongs to others, and a commitment to enforcing those boundaries — the world becomes one vast common pool: a “state of nature,” as theorists of old called it.

The thought that these precepts ground a rule-of-law order not just among individuals but also, and arguably more, among the political beasts we call “states” or “nations” was vivid to the Founders. Contentions among the former colonies — over boundaries, commerce, other vital questions — and the Continental Congress’s inability to do much about them under the Articles of Confederation propelled the enterprise to form a “more perfect Union.”

In deliberating over rules of engagement for the more firmly united states, the Founders did not draw on a blank slate. They drew on what was then called the “law of nations,” as it had developed among civilized nations and by then been imported by Lord Mansfield into English common law (which of course also governed in the colonies). The Constitution reflects this understanding throughout. “Commerce among the several states” meant transactions that counted as such under the law of nations. The “full faith and credit” that the Constitution obliges the states to give to one another’s law meant such credit as one would give under the law of nations. This meant, among other things, full recognition of property titles acquired or contracts made under foreign law, with one crucial exception: slavery. That institution was “so odious, that nothing [could] be suffered to support it, but positive law,” meaning the positive law of the forum state. This language from Lord Mansfield’s opinion in Somerset’s Case (1772) was known to everyone at the time. This explains why the Constitution’s fugitive-slave clause permits owners to demand the delivery of slaves who escaped to free states — and nothing beyond it. You bring your slave into a free state and he manages to file for habeas: “The black must be discharged.” That was the dispositive ruling of Somerset’s Case, and antebellum courts adjudicated accordingly until Dred Scott. None of this is remotely comprehensible without the law-of-nations backdrop. All of it forms a context that all but the most doctrinaire originalist would accept as a perfectly legitimate form of legal argument.

The federal courts still follow the basic precepts in the international domain. For example, the Supreme Court applies a firm “presumption against extraterritoriality” to congressional enactments: Unless Congress very clearly meant a statute to apply to conduct outside the United States, it won’t. Likewise, in international cases, the Court has been very hostile to “f-cubed” actions of the Mallory variety (foreign plaintiff, foreign defendant, foreign events). In analogous domestic contexts, in contrast, just about anything goes — state statutes that govern practically nothing but conduct wholly foreign to the jurisdiction (Pork Producers); universal state-court jurisdiction on the flimsiest pretense of “consent” (Mallory). There are, of course, important considerations that distinguish the international from the domestic arrangement. For starters, states are not quite as sovereign under the Constitution as are nations. None of those distinctions, however, warrants the Court’s willful disregard not only of the spirit and context of the Constitution but of its plain text. We now turn, as we must, to that text.

The Constitution teems with explicit “horizontal” federalism clauses that reflect the law-of-nations background. Many are contained in Article IV; others, such as prohibitions against state compacts and import-export taxes, in Article I, Section 10. Jointly and severally, the clauses embrace, extend, and domesticate the law of nations. They envision a union of equal territorial states, organized on principles of nonaggression (e.g., no standing armies), nondiscrimination (no import-export taxes, no laws impairing the obligation of contract), and mutual recognition (full faith and credit to sister states’ public acts, records, and proceedings). In addition, the 14th Amendment’s due-process clause protects citizens against unilateral aggression by a sister state’s courts: The tribunal that holds you liable must have legal power over you.

Many of the clauses may seem redundant — redundant, that is, of federal legislative powers. Even under the most literalist reading of the commerce power, for example, Congress could surely enjoin states from imposing taxes or regulations on “commerce among the several states.” Similarly, the full-faith-and-credit clause specifically authorizes Congress to enact implementing legislation. But leaving matters to Congress is leaving them to chance. With respect to the vital questions of interstate relations, the Founders did not want to run that risk. That is why virtually all the clauses are directly enforceable by private litigants as of right. To that considerable extent, the federal courts are federalism’s principal protectors. That is why Article III confers upon them diversity jurisdiction over controversies between different states and their citizens.

All of this is, or ought to be, common ground. Yet none of it resonates in the Supreme Court’s decisions and opinions. Once the context goes missing, so do the textual clauses.

Take Pork Producers. Justice Gorsuch’s forceful averment that the dormant commerce clause imposes no impediment — none — to extraterritorial state regulation left Justice Kavanaugh uneasy. Maybe the dormant commerce clause does not bar this sort of thing, he opined in a partial dissent. But maybe the import-export clause (Article I, Section 10) does. Or maybe the privileges-and-immunities clause of Article IV, which bars states from treating out-of-state citizens worse than their own. Or maybe the full-faith-and-credit clause of Article IV. Those provisions and perhaps others, Justice Kavanaugh wrote, are “implicated” by state laws such as California’s, even if none of them might provide relief at the end of the day.

Much as one appreciates Kavanaugh’s concerns, they are lipstick on a pig. The import-export clause prohibits certain taxes but not regulations such as California’s. The privileges-and-immunities clause isn’t implicated at all: California’s law hits local and out-of-state producers alike. In any event the clause protects only citizens, not corporations. So the local pig farmer might get a reprieve; the outfits that turn the poor sows into chops, sausages, footballs, and sex gear would not. As for the full-faith-and-credit clause: Under current law, the clause still compels states to respect and enforce judgments rendered by sister-state courts. With respect to “public acts” (i.e., legislation), in contrast, unchallenged precedents hold that full faith and credit means such credit as the forum state may want to give — which, with dreary predictability, means none at all. ’Tis conceivable, albeit barely, that the clause might bar State A’s criminalization of conduct that is privileged or affirmatively mandated in State Z. (Justice Kavanaugh’s opinion contemplates the scenario.) Even then, though, producers in States B to Y would still be on the hook.

For complementary illumination of this wasteland, consider Justice Alito’s partial concurrence in Mallory. “A State’s assertion of jurisdiction over lawsuits with no real connection to the State may violate fundamental principles that are protected by one or more constitutional provisions or by the very structure of the federal system that the Constitution created,” Justice Alito commendably observed. “At this point in the development of our constitutional case law,” he continued, “the most appropriate home for these principles is the so-called dormant Commerce Clause.” That would be the same dormant commerce clause whose extraterritoriality prong — the only piece that matters in the Mallory context — had been buried in Pork Producers a mere seven weeks earlier.

The federalism-without-rules-or-law travesty crystallizes a deeper pathology: the unholy matrimony between the New Deal Constitution and modern-day originalism in its positivist, clause-bound configurations.

On federalism as on everything else, the New Deal project was to contract the rule of law — in any meaningful, substantive sense, or even in the simple sense of lawfulness and regularity — so as to make ample room for “democracy,” meaning the dominant New Deal coalition in Congress and Progressive states. That, to repeat, was the point of Erie Railroad and an entire jurisprudence built around it: judicial restraint; state laboratories; nothing for federal courts but to do the will of Congress; grim legal positivism; unremitting hostility to anything that smacks of federal common law.

All that burble was just the New Dealers’ polite way of saying, “We win.” Early modern-day originalists, fatefully and alas, played it as a card against the “judicial activism” of the Warren Court. With that move, though, they surrendered any serious thought about the rule of law, the Constitution, and federalism. Some originalist scholars have commendably renounced that inheritance and made room for a more substantive contextual understanding. There remains, however, a pronounced strain of narrow, clause-bound positivism: If you cannot rest your case on something that’s written down, you lose.

Take one last look at Pork Producers and Mallory: Justice Kavanaugh and Justice Alito plainly grasp that there is something wrong with the Court’s understanding, or lack thereof, of the Constitution’s federalism architecture. But even they promptly gravitate to individual clauses — all of which have long been wrenched out of context and so rendered toothless. The more clause-bound originalism becomes, the more it will tend to read clauses out of the Constitution, at least in the federalism theater of operations.

It will be exceedingly difficult to revive a more constitutional federalism through piecemeal strategic litigation. Over the past three or so decades, constitutional litigants usually have made progress by hanging their agenda on the reinterpretation of a discrete clause — the establishment clause; the free-exercise clause; the equal-protection clause; within limits, the takings clause. Federalism, and “horizontal” federalism in particular, just isn’t like that. Litigants start any given case at least four votes down (Justices Gorsuch, Thomas, Jackson, and Sotomayor), and even potentially and intentionally helpful opinions — Justice Kavanaugh’s in Pork Producers, Justice Alito’s in Mallory — chase the lawyers away from the structure and spirit of the Constitution and into some rabbit hole. That, in turn, reinforces the worst instincts of the commercial lawyers who handle Pork Producers, Mallory, and similar cases. They do not think strategically, nor even from case to case. Instead, they try to cobble together five votes for this client in this case. The transparent maneuvering invariably craters in front of justices who insist, more than many of their predecessors, on jurisprudentially and constitutionally grounded rules that will hold up in the general run of cases. The federalism difficulty is not that no rule of law can be had or discerned; it’s that the principled rule you might want to articulate will also lose you the votes you need. You skate, you lose; you engage, you lose.

This will eventually be fixed, because it must be fixed. I just cannot envision the trajectory.