Restoring the Establishment Clause to the States

One of the quieter arguments in modern church-state law is also one of the most radical: that the Establishment Clause was never meant to bind the states at all, and that returning it to its original posture would, paradoxically, widen rather than narrow religious freedom. The thesis travels under the banner of “restoring” the clause to the states, and its proponents pair it with a claim about tolerance — that a single national rule of church-state separation, imposed from Washington, has flattened the older constitutional design in which different communities answered the religion question differently. Whatever one makes of that claim, it rests on a serious reading of text and history, and it has drawn votes on the Supreme Court. This is a survey of the argument, the doctrine it would unsettle, and the open questions it leaves behind.

What “restoring the clause to the states” actually means

The First Amendment opens, “Congress shall make no law respecting an establishment of religion.” Read against its history, the disincorporation thesis treats those words as a limit on the federal government specifically, and as a guarantee that Congress would leave each state’s own arrangement undisturbed. At the founding, several states retained religious establishments: Massachusetts funded Congregationalist “public teachers of piety, religion and morality” through its 1780 constitution, and did not formally disestablish until 1833. Connecticut, New Hampshire, South Carolina, and Maryland carried establishments of varying kinds into the early republic. On this reading, the clause did double duty — it forbade a national church and it shielded the surviving state churches from federal interference.

If that is the original sense, then a clause aimed at protecting state choices is an awkward candidate for incorporation against the states through the Fourteenth Amendment. To “incorporate” it, the argument runs, is to turn a federalism rule on its head: a provision designed to keep Washington out of state establishment decisions becomes a tool for Washington’s courts to police those very decisions. Restoring the clause to the states, then, means undoing that move and returning the establishment question to state constitutions and state electorates.

How the modern rule was built

The contemporary rule runs the other way. The Free Exercise Clause was applied to the states in Cantwell v. Connecticut, 310 U.S. 296 (1940), through the Due Process Clause of the Fourteenth Amendment. The Establishment Clause followed seven years later in Everson v. Board of Education, 330 U.S. 1 (1947), where the Court — while upholding New Jersey’s reimbursement of bus fares for parochial-school pupils — declared the clause binding on the states and read it to require, in Justice Black’s much-quoted phrase, “a wall of separation between church and state.”

From Everson forward the clause became one of the most heavily litigated provisions in the Constitution. The Court built a test in Lemon v. Kurtzman, 403 U.S. 602 (1971), asking whether a challenged practice had a secular purpose, whether its primary effect advanced or inhibited religion, and whether it produced excessive government entanglement with religion. The Lemon framework, and the “endorsement” gloss layered onto it, governed school-prayer, public-display, and funding disputes for decades. Disincorporation would not merely trim that body of law at the edges; it would remove its constitutional footing as against the states entirely.

The federalism reading on the Court

The argument is not confined to the academy. Justice Thomas has pressed it repeatedly, most prominently in his concurrence in Elk Grove Unified School District v. Newdow, 542 U.S. 1 (2004), where he wrote that the Establishment Clause “is best understood as a federalism provision” that “resists incorporation.” His point is structural. The Free Exercise Clause protects an individual right and so translates naturally into a liberty the Fourteenth Amendment can carry against the states. An anti-establishment rule, by contrast, looks on this reading like a restraint on the level of government — a jurisdictional allocation rather than a personal freedom — and incorporating it would bar the states from doing the very thing the clause was meant to leave to them.

Scholars across the spectrum have found the historical case at least respectable. Akhil Amar, working from the Privileges or Immunities Clause rather than Due Process, has argued that the original Establishment Clause carried a federalism dimension and that any application to the states must be a refashioned one, not a mechanical transfer of the 1791 rule. That convergence between an originalist justice and a scholar often associated with liberal constitutionalism is part of why the thesis has staying power, even as a majority of the Court has never adopted it.

Why “tolerance” is the contested word

The framing that restoring the clause would restore tolerance is where description shades into contention. Proponents argue that a uniform, judicially enforced separationism has been less neutral than advertised — that it pushed religious expression out of public life under the banner of even-handedness, and that returning the question to the states would let communities accommodate religious practice without federal courts second-guessing every legislative prayer or holiday display. They can point to a doctrinal drift already underway: Town of Greece v. Galloway, 572 U.S. 565 (2014), upheld sectarian legislative invocations by looking to historical practice rather than the Lemon test, and Trinity Lutheran Church v. Comer, 582 U.S. 449 (2017), held that excluding a church from a neutral public grant solely because of its religious character burdened free exercise.

Critics respond that the clause protects individuals, not states — that an establishment harms the dissenter, the minority believer, and the nonbeliever regardless of which level of government imposes it, and that disincorporation would license precisely the coercion the modern doctrine guards against. They also note that the Reconstruction generation that wrote the Fourteenth Amendment may have understood the religion clauses differently than the founders did, which complicates any clean appeal to 1791 meaning. On this view, “tolerance” is exactly what a robust, incorporated Establishment Clause secures, and “restoring” it to the states would shrink the rights of religious minorities in places where the local majority is unified.

What courts would have to rebuild

Disincorporation would not answer church-state questions so much as relocate them. State constitutions would do the heavy lifting, and many contain their own, often stricter, establishment and no-aid provisions — the so-called Blaine Amendments — so the practical effect would vary sharply by jurisdiction. Public-school curricula, government displays, and legislative prayer would be governed by fifty different texts and fifty different courts rather than one federal standard. Federal questions would not vanish: the Free Exercise Clause, the Equal Protection Clause, and federal antidiscrimination statutes would still constrain state action, and a state that favored one sect could face challenges sounding in those provisions even without an incorporated Establishment Clause.

That is why the thesis remains, for now, a minority position argued mostly in concurrences rather than holdings. The Court has shown more appetite for changing the test than for changing the scope of the clause — emphasizing history and tradition over the old tripartite inquiry while leaving incorporation untouched. Readers tracking how the religion clauses interact with privacy, employment, and equal-protection doctrine may find the through-lines in the publication’s ongoing commentary and its docket coverage at the case tracker; the structural questions here echo the federalism debates that surface elsewhere, including in work on employee privacy rights where individual entitlements and institutional power are likewise in tension.

Where the question is heading

The likeliest near-term path is not a dramatic disincorporation but a continued narrowing of what the incorporated clause forbids — a Court that keeps the clause applicable to the states while reading it through historical practice, so that more accommodation survives review. That would deliver much of what the federalism camp wants without the doctrinal upheaval of overruling Everson. But the deeper argument has not gone away, and its persistence is a reminder that incorporation, treated as settled for most of the rights in the Bill of Rights, is genuinely contested for this one. The text invites the question, the history does not resolve it cleanly, and the votes to keep asking it are on the bench.

The publication offers commentary and analysis on developing areas of constitutional law; it is not legal advice.