The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Structure, Text, and History in PennEast Pipeline v. New Jersey

Justice Barrett bests the Chief on sovereign immunity and eminent domain.


The topic of sovereign immunity has long been a controversial subject in constitutional law. Chisholm v. Georgia. The 11th Amendment. Hans v. LouisianaSeminole TribeAlden v. Maine. And so on. These issues sharply divided the Rehnquist Court. But I think the unanimity in Allen v. Cooper reflected a generational shift. Now, all of the Justices seems to be on the same page about issues already decided. But what about sovereign immunity issues of first impression? PennEast Pipeline v. New Jersey presents such a case.

Chief Justice Roberts wrote the majority opinion, which was joined by Justices Breyer, Alito, Sotomayor, and Kavanaugh. Justice Gorsuch wrote one dissent, which was joined by Justice Thomas. Justice Barrett wrote a second dissent, which was joined by Justices Thomas, Kagan, and Gorsuch. This divide was not ideological. I described the two camps as the Court's non-curious and curious caucuses.

In this case, FERC delegated to PennEast the power to condemn land for a pipeline. PennEast exercised eminent domain on land owned by the New Jersey state government. In order to pursue that remedy, PennEast (a Delaware corporation) sued New Jersey. This suit squarely falls within the prohibition of the 11th Amendment. Thus, the case turns on whether New Jersey's sovereign immunity is waived.

The majority and dissenting opinions take very, very different approaches to the case. Justice Barrett writes that the majority's "argument has no textual, structural, or historical support." This triad–text, structure, and history–is significant. I recently criticized Justice Barrett's Fulton concurrence. She wrote that history did not justify overruling Smith, but text and structure might. But text and structure didn't actually mean the text of the First Amendment or the structure between the federal and state governments. Rather, she was referring to balancing the Free Exercise Clause with the remainder of the the Court's modern jurisprudence. This statement was disappointing for an originalist. In PennEast, Barrett explores all three legs: text, structure, and history.

Let's walk through the case.

First, Barrett explains that the Constitution does not grant Congress the power to exercise eminent domain. The Bill of Rights imposes limitations on federal power. It does not grant governmental authority.

First, the Constitution enumerates no stand-alone "eminent-domain power."2 The Court recognizes—as does our precedent—that the Federal Government may exercise the right of eminent domain only "so far as is necessary to the enjoyment of the powers conferred upon it by the Constitution." Kohl v. United States, 91 U. S. 367, 372 (1876); see McCulloch v. Maryland, 4 Wheat. 316, 421 (1819). 

2The Takings Clause of the Fifth Amendment is a limitation on Government power, not a grant of it. It provides: "[N]or shall private property be taken for public use, without just compensation." It thus presumes that the power exists by virtue of other constitutional provisions. 

Later, Barrett favorably cites Will Baude, who has cast serious doubt on the federal government's powers to seize land outside of federal enclaves:

Moreover, no one disputes that for 75 years after the founding, it was unsettled whether the Federal Government could even exercise eminent domain over private land within a State. See Baude, Rethinking the Federal Eminent Domain Power, 122 Yale L. J. 1738, 1741, 1761–1777 (2013).

By contrast, Chief Justice Roberts writes that the Fifth Amendment recognized a federal "power" to seize land in the states.

When the Constitution and Bill of Rights were ratified, they did not include the words "eminent domain." The Takings Clause of the Fifth Amendment ("nor shall private property be taken for public use, without just compensation") nevertheless recognized the existence of such a power. 

Chief Justice Roberts does not acknowledge this groundbreaking work from one of his most influential law clerks. (I suspect Josh Hawley tops that list). Roberts generally holds a dim view of legal scholars. (I suppose the feeling is mutual). But we aren't talking about "the influence of Immanuel Kant on evidentiary approaches in 18th Century Bulgaria." Baude relies on Roberts's NFIB opinion, which is based on Chief Justice Marshall's McCulloch decision! This slight had to be deliberate.

Second, Justice Barrett reaffirms Baude's scholarship. When the federal government exercises eminent domain, it must use one of its other enumerated powers–primarily the Necessary and Proper Clause.

Any taking of property provided for by Congress is thus an exercise of another constitutional power—in the case of the Natural Gas Act, the Commerce Clause—augmented by the Necessary and Proper Clause. So when Congress allows a private party to take property in service of a federally authorized project, it is choosing a means by which to carry an enumerated power into effect.

Justice Barrett properly frames the inquiry:

Here is the right way: Title 15 U. S. C.§717f(h) is an exercise of Congress' power to regulate interstate commerce. Congress cannot authorize private suits against a non consenting State pursuant to its Commerce Clause power. Seminole Tribe, 517 U. S., at 72–73. Nor does the Commerce Clause itself abrogate state sovereign immunity. Cf. Allen, 589 U. S., at ___–___ (slip op., at 8–9). Therefore, Congress cannot enable a private party like Penn-East to institute a condemnation action against a nonconsenting State like New Jersey. 

The Chief does not even suggest that Congress could rely on its Article I powers to waive sovereign immunity here. Such a ruling would run afoul of Seminole Tribe.

Third, instead, the Chief contends that New Jersey in fact surrendered immunity from the federal government's eminent domain power during the constitutional convention. (And to be precise, New Jersey ratified the Constitution in December 1787–that period would be the relevant date).

But there is zero history to show that Congress ever delegated to private parties the power to seize lands from the states. Justice Barrett writes:

But the question before us is not whether Congress can authorize a private party to exercise the right of eminent domain against another private party, which is the proposition this history supports. Nor is it whether Congress can authorize a private entity to take state property through means other than a condemnation suit. The question is whether Congress can authorize a private party to bring a condemnation suit against a State. And on that score, the Court comes up dry. . . .

The Court cannot muster even a single decision involving a private condemnation suit against a State, let alone any decision holding that the States lack immunity from such suits. It relies exclusively on suits brought by States, suits brought by the United States, suits brought by private parties against other private parties, and suits brought by Indian tribes against private parties—none of which implicate state sovereign immunity. 

And the absence of a historical practice suggests that this power did not exist.

Given the length of time that these questions lingered, it strains credulity to say that history unequivocally establishes that States surrendered their immunity to private condemnation suits in the plan of the Convention 

Moreover, when looking for waivers of sovereign immunity, you need very clear evidence. The Chiefs "dead end" does not support that argument.

Fourth, how does the Chief respond to these historical arguments? He turns to precedent. For example, United States v. Texas (1892) (Harlan, J.) held that states consented to suits by the federal government. And that decision did not rely on historical examples in which the federal government sued the states. Barrett replies:

But in that decision, the supremacy of the Federal Government in our constitutional structure, along with textual cues, were sufficient to resolve the question. Id., at 644–646. Here, there is no basis for drawing an analogous structural inference, much less any remotely relevant text. 

Eh, I thought this was the weakest part of the dissent. Like in Fulton, Justice Barrett uses the words "structure" and "text" as weasel words. Stick with history. It's okay to say a precedent was insufficiently reasoned. Or, say this entire "convention waiver" framework is nonsense. (It is.) But don't hide behind unexplained hedges like "constitutional structure" and "textual cues."

Roberts pounced on this argument:

The dissent argues that the Court in Texas relied not only on "constitutional structure," but also on "textual cues." Post, at 6. But the only relevant constitutional text in Texas was a grant of federal jurisdiction, and that cannot explainStates' implicit consent in the plan of the Convention to suits by the Federal Government.

He's right. You can't beat the Chief on precedent, so don't try. And my goodness, in the context of the 11th Amendment, you cannot rely on "text." The 11th Amendment, per Hans, doesn't mean what it says. Roberts continues:

If it could, then the extension of the judicial power to controversies "between a State and Citizens of another State," Art. III, §2, cl. 1, would suggest that Chisholm v. Georgia correctly held that nonconsenting States could be subject to private suit.

Earlier, Roberts said Chisholm was an "error." I don't know if Barrett thinks Chisholm was wrong as an original matter. She seems to suggest that she does not think it is an "error." Justice Harlan in Hans said Chisholm was correct when it was decided.

Fifth, the rest of Barrett's dissent responds to various pragmatic concerns raised by the majority.

While the Court cloaks its analysis in the "plan of the Convention," it seems to be animated by pragmatic concerns. 

Roberts's historical analysis was largely window dressing. (The references to his homestate of Indiana were delightful). Without question, the majority opinion makes things easier for federal delegatees to seize property. If the goal is to make government work more efficiently, then sovereign immunity should be deemed waived. But Justice Barrett recognizes that the separation of powers are not designed to make government work more effectively. Justice Scalia often explained that friction is a feature, and not a bug in our system of divided government.

Barrett seems to agree. She writes:

State sovereign immunity indisputably makes it harder for Congress to accomplish its goals, as we have recognized many times before. . . . The same is true here: Sovereign immunity limits how Congress can obtain state property for pipelines. This inhibition of Congress is not, however, a reason to set sovereign immunity aside. It is instead a deliberately chosen feature of the constitutional design.

Well said.

I also commend Justice Gorsuch's dissent, which was joined by Justice Thomas. If I read Justice Gorsuch correctly, he suggests that cases like Seminole Tribe, Allen v. Cooper, and Alden do not implicate 11th Amendment immunity. Rather, they rely on "structural immunity." The 11th Amendment, or at least its text, has no bearing on this doctrine:

The first—"structural immunity"—derives from the structure of the Constitution. See Franchise Tax Bd. of Cal.  v. Hyatt, 587 U. S. ___, ___ (2019) (slip op., at 16). Because structural immunity is a constitutional entitlement of a sovereign State, it applies in both federal tribunals, Seminole Tribe of Fla. v. Florida, 517 U. S. 44, 51–52 (1996), and in state tribunals, Alden, 527 U. S., at 712. And it applies regardless of whether the plaintiff is a citizen of the same State, Allen v. Cooper, 589 U. S. ___, ___ (2020) (slip op., at2), a citizen of a different State, or a non-citizen—like a foreign nation, Principality of Monaco v. Mississippi, 292 U. S. 313, 330 (1934), or an Indian tribe, Blatchford v. Native Village of Noatak, 501 U. S. 775, 781 (1991). 

By contrast, 11th Amendment immunity derives from the text of the 11th Amendment. Here, Gorsuch cites an important article by Will Baude and Steve Sachs. This article reshaped how I think about the 11th Amendment. You should read it.

The second—what is properly termed "Eleventh Amendment immunity"—derives from the text of the Eleventh Amendment. In light of its swift adoption in response to Chisholm v. Georgia, 2 Dall. 419 (1793), this Court has read the Eleventh Amendment as pointing to the structural principle just discussed. See Allen, 589 U. S., at ___ (slip op., at 4). But the Eleventh Amendment can do two things at once. See Federal Maritime Comm'n v. South Carolina Ports Authority, 535 U. S. 743, 753 (2002). In addition to pointing us back to the States' structural immunity, it also provides an ironclad rule for a particular category of diversity suits . . . .

This text "means what it says. It eliminates federal judicial power over one set of cases: suits filed against states, in law or equity, by diverse plaintiffs." Baude & Sachs, The Misunderstood Eleventh Amendment, 169 U. Pa. L. Rev. 609, 612 (2021).

PennEast involves a suit by a Delaware corporation against New Jersey. Thus, the dispute actually fall within the text of the 11th Amendment. Gorsuch writes:

This case appears to present "the rare scenario" that comes within the Eleventh Amendment's text. …  Because PennEast sued New Jersey in federal court, this suit implicates "the Judicial power of the United States." … This condemnation suit, by any stretch, is "a[ ] suit in law or equity." … PennEast "commenced" this suit "against" New Jersey. It named the State in its complaint as a defendant as required by the Civil Rules. Fed. Rule Civ. Proc. 71.1(c)(1). And it asked the court for an injunction permitting it to take "immediate possession" of New Jersey's soil. Because the parties agree that PennEast is a citizen of Delaware, this suit is brought "by [a] Citizen[ ] of another State."

If Gorsuch is right, this case should be thrown out of court, without regard to Hans v. Louisiana. We are squarely within the text of the 11th Amendment.

If that's all true, then a federal court "shall not" entertain this suit. The Eleventh Amendment's text, no less than the Constitution's structure, may bar it. 

If the federal government wants to seize this land, it must do so directly.

What an absolutely fascinating case. So far, the most significant constitutional law decision of the term. I hope federal courts casebooks include it.