5th Cir. en banc  ·  No. 25-50695  ·  Decided April 21, 2026  ·  9-8 to reverse

After the Decalogue, the Voucher

Carson, Nathan, and the Problem of Sectarian Purpose

Format: An essay for law students and law professors
Companion to: the full case brief and the first essay, What Is Left of the Wall?
Length: ~2,570 words / ~12-minute read

I · The Puzzle

Texas's Religion Clauses politics looks, at first glance, internally contradictory. On April 21, 2026, the en banc Fifth Circuit decided Nathan v. Alamo Heights Independent School District, No. 25-50695 (5th Cir.), holding that Texas may compel every public-school classroom to display the King James Decalogue. The day before, the comptroller's office was still sorting roughly 250,000 surviving applications to the state's new Texas Education Freedom Account (TEFA) voucher program — a program that, on its first-year rollout, is about to send most of its public dollars into evangelical Christian and Catholic private schools. Either move alone is constitutionally marginal after Kennedy v. Bremerton School District, 597 U.S. 507 (2022), and Carson v. Makin, 596 U.S. 767 (2022). Both moves together create a doctrinal puzzle that no Supreme Court opinion has yet addressed.

The puzzle has three facets. First: did Nathan's validation of state-sponsored religious content inside the public school dissipate political demand for vouchers? Second: can a state exploit Carson's neutrality premise by operating a voucher program whose practical universe of participating schools is almost entirely sectarian, and whose administrative decisions about which religions count as acceptable sectarian providers are themselves being made with apparent denominational bias? Third: if such a program can be challenged, on what doctrinal theory — given that the most natural theory sounds suspiciously like the "secular purpose" prong of Lemon v. Kurtzman, 403 U.S. 602 (1971), which the Supreme Court recently declared abandoned?

This essay takes up each question in turn. The substitution hypothesis is now empirically falsified, at least for Texas. The exploitation hypothesis has moved from theory to pending litigation. And the doctrinal tools for attacking an engineered-sectarian voucher market — Zelman's genuine-choice prong, Larson v. Valente's denominational-preference rule, and the state-constitutional successor litigation Professor Caroline Mala Corbin predicts will displace much of the federal inquiry — are already being deployed in Texas courts.

II · The Substitution Hypothesis, Empirically Tested

The first question is political. If classroom Ten Commandments displays have now been legalized, one might expect the movement demanding vouchers for religious schools to lose some of its animating complaint: that the public schools have been scrubbed of religion, forcing faithful parents to pay twice for religious education.

That complaint has been the rhetorical backbone of the school-choice movement since James Forman's canonical history of the voucher cause. See James Forman, Jr., The Rise and Fall of School Vouchers: A Story of Religion, Race, and Politics, 54 UCLA L. Rev. 547 (2007). If Nathan now permits the public schools to host the Decalogue and, on the majority's reasoning, arguably much more, then in principle the public-school-as-hostile-secular-space narrative should weaken.

The numbers show the opposite. Texas Senate Bill 2, which Governor Abbott signed on May 3, 2025, created the TEFA program and appropriated $1 billion to it for the FY 2026–2027 biennium. The first application window ran February 4 through March 31, 2026, and drew more than 274,000 applications — roughly three times the number of available slots. Because demand outstripped funding, the comptroller's office turned to the lottery system the statute prescribes. Nearly 30,000 applicants qualify for the first priority tier (low- and middle-income children with disabilities); approximately 79,000 more qualify for the second tier (low-income households); funding will run out before the comptroller reaches the rest. The Legislative Budget Board projects demand will push appropriations to $3.3 billion by 2028 and $4.8 billion by 2030.

Three reasons explain why Nathan has not drained the voucher coalition.

First, the coalitions were never identical. Orthodox Jewish, Catholic, and Muslim parents want genuinely sectarian instruction — Hebrew, catechism, Qur'anic recitation — that no public school offers even under Nathan's permissive reading of Kennedy. A Protestant Decalogue on a public classroom wall is the wrong content for most of these communities.

Second, the voucher movement absorbed other agendas. As Forman documents, modern voucher politics tracks dissatisfaction with desegregation and urban school administration at least as much as it tracks religious grievance. Wall displays do not address those concerns.

Third, political wins generate demand rather than absorb it. The classroom-display statute and TEFA emerged from the same political coalition in overlapping legislative sessions, and that coalition has already begun pressing for TEFA expansion — a pressure reflected in the Legislative Budget Board's $4.8 billion projection. The substitution thesis was plausible in the abstract. The March 2026 application totals refute it.

III · The Exploitation Hypothesis, Now in Litigation

Carson holds that "[a] State violates the Free Exercise Clause when it excludes religious observers from otherwise available public benefits," and that "once a State decides to subsidize private education, it cannot disqualify some private schools solely because they are religious." 596 U.S. at 778. The Court rejected the distinction between religious status and religious use that Maine built into its "nonsectarian" requirement. Id. at 786–87.

Carson does not, however, require any state to subsidize private education. It is a nondiscrimination rule, not a funding mandate. A state may decline to fund private education altogether; and if it does fund, it cannot carve out religious schools. Carson says nothing about the state's animating purpose in deciding to fund, and nothing about the composition of the private-school market the state is paying into.

That silence — the opinion is about discriminatory effect, not animating purpose — is the hinge of the exploitation hypothesis. Carson does not decide whether a state's religious motive in enacting a voucher program could render the program unconstitutional, nor whether the composition of the private-school market the state is paying into bears on the Establishment Clause analysis. Imagine a state that designs a voucher program with the specific purpose of funneling public dollars into a narrow band of Christian academies. Its statute is facially neutral. Any accredited private school may participate. But the state sets accreditation rules, voucher amounts, and administrative discretion in ways that ensure, practically, that the participating schools will be overwhelmingly sectarian within a single tradition.

TEFA's first-year data fits that picture closely. Of the 291 TEFA-participating schools that the Texas Observer analyzed, 268 (roughly 92 percent) were religious: 176 Catholic, 91 Protestant, and one Jewish, alongside 23 secular schools. As more schools enrolled, the share stabilized near 80 percent of the 1,700-plus participating campuses. Forty percent of the participating religious schools require or prioritize coreligionist applicants; thirty percent maintain written policies excluding LGBTQ+ students. Indiana's voucher program previewed this pattern in earlier years. See Eberle-Peay, Chalk Talks — Indiana's Choice Scholarship Program, 41 J.L. & Educ. 553 (2012) (reporting that 97 percent of schools participating in the Indiana program were sectarian).

The denominational-preference concern has moved from theoretical to concrete. Acting Comptroller Kelly Hancock approved hundreds of Christian schools into TEFA and simultaneously excluded every Islamic school that applied. The stated basis for exclusion was that the accrediting agency used by the Islamic schools had hosted events organized by the Council on American-Islamic Relations — a civil-rights organization Governor Abbott has labeled a "terrorist" group, a characterization the federal government has not adopted. Four Muslim parents and three Islamic schools filed suit. A federal judge granted a temporary restraining order on March 2, 2026, and extended the application deadline. Hancock subsequently admitted five of the previously excluded Islamic schools. The next hearing in the consolidated case is April 24, 2026 — two days from the date of this essay.

The Muslim-exclusion lawsuit presents, in real time, the question this essay asks in the abstract: can a state satisfy Carson facially while administering its program in a manner that expresses a preference among religions?

IV · The Religious-Motive Problem

The natural vehicle for the exploitation claim is a religious-purpose challenge: Texas enacted, funded, and administered TEFA because it wanted to advance Christian education, and that purpose should be disqualifying. The argument runs directly into the most disputed holding of Kennedy.

Kennedy declares that "[i]n place of Lemon and the endorsement test, this Court has instructed that the Establishment Clause must be interpreted by 'reference to historical practices and understandings.'" 597 U.S. at 535 (quoting Town of Greece v. Galloway, 572 U.S. 565, 576 (2014)). The Court's disapproval of Lemon's purpose prong is explicit. Id. at 534. McCreary County v. ACLU, 545 U.S. 844 (2005), which invalidated a Ten Commandments display on purpose grounds, remains formally on the books but is, after Kennedy, doctrinally inert — a point the Nathan majority exploits when it abrogates Stone v. Graham.

So the doctrinal problem is this. A challenger to TEFA who argues that the program was enacted with a sectarian purpose is, in the eyes of post-Kennedy doctrine, pressing exactly the kind of subjective-purpose inquiry the Court said it had abandoned.

Three responses are available.

First, Kennedy does not decide what it does not decide. The opinion sweeps Lemon aside but does not specify whether any form of the secular-purpose inquiry survives as a distinct sub-test within the new historical framework. The historical-practices framework itself cannot avoid reference to governmental motive: the Founding-era establishments that the framework takes as its touchstone were defined in part by their purpose of compelled religious support. Everson v. Board of Education, 330 U.S. 1, 15–16 (1947). A sharply sectarian motive may remain relevant evidence that a contemporary program lies outside the historical tradition.

Second, the challenger can recast purpose as denominational preference. Larson v. Valente, 456 U.S. 228, 244 (1982), survives Kennedy untouched: "[t]he clearest command of the Establishment Clause is that one religious denomination cannot be officially preferred over another." A voucher program deliberately designed or administered to favor one religious tradition over another is a paradigmatic denominational-preference case. Catholic Charities Bureau v. Wisconsin LIRC, 605 U.S. ___ (2025), recently reaffirmed Larson's strict scrutiny rule. The Muslim-exclusion lawsuit in Texas is the cleanest Larson vehicle to reach a federal court in years: a state official's categorical rejection of Islamic schools while approving hundreds of Christian ones. This is the purpose inquiry wearing textualist clothing; it turns on the comparative treatment of religions rather than on the bare existence of a religious purpose.

Third, the challenger can reach for Shurtleff v. Boston, 596 U.S. 243 (2022), and specifically Justice Gorsuch's concurrence, which identified six historical hallmarks of unconstitutional establishment — including compelled financial support for a favored religion. A voucher program that channels public revenues into a narrow sectarian band through an engineered market is, on one reading, a modern version of the compelled-support problem that animated Madison's Memorial and Remonstrance against Religious Assessments (1785). This is the move Judge Oldham's partial concurrence in Nathan gestured at — a Shurtleff-grounded framework can host a compelled-support claim even after Lemon's demise.

V · What Survives

Four doctrinal hooks remain available to challenge an exploitation-style voucher program in the post-Kennedy, post-Nathan world.

The first is the Zelman genuine-choice prong. In Zelman v. Simmons-Harris, 536 U.S. 639 (2002), the Court upheld Cleveland's voucher program because the aid reached religious institutions "wholly as a result of [parents'] own genuine and independent private choice," id. at 652, and the "constitutionality of a neutral educational aid program simply does not turn on whether and why, in a particular area, at a particular time, most private schools are run by religious organizations, or most recipients choose to use the aid at a religious school," id. at 658. Zelman specifically approved the 96 percent religious-beneficiary figure from Mueller v. Allen. Id.

But Zelman's test looks at the full educational market — public schools, magnet schools, community schools, secular private schools, and religious private schools, id. at 655–56 — not at the voucher program in isolation. If plaintiffs can show that Texas has engineered the private-school market (through regulation, accreditation, or selective approval) so that the sectarian outcome is predetermined rather than chosen, TEFA may fail the genuine-choice prong even though it passes facial Carson review. This is the argument Lupu and Tuttle foreshadowed in Zelman's Future: Vouchers, Sectarian Providers, and the Next Round of Constitutional Battles, 78 Notre Dame L. Rev. 917 (2003), and that Thomas Berg developed in Vouchers and Religious Schools: The New Constitutional Questions, 72 U. Cin. L. Rev. 151 (2003).

The second is Larson v. Valente. The Muslim-exclusion lawsuit supplies a test case. Either the state loses — in which case Larson is reaffirmed as a live constraint on administrative discretion in voucher programs — or the state wins, in which case the First Amendment permits a modern denominational preference that would have astonished Madison. Either outcome will reshape the exploitation-hypothesis debate.

The third is the Shurtleff compelled-support line. Gorsuch's concurrence identifies compelled public financial support for religious instruction as a paradigm historical establishment, and the Nathan majority has just adopted the six-hallmark framework. A plaintiff challenging TEFA on Shurtleff grounds can argue that the majority's own analytic vocabulary compels a different result in the voucher context, because the voucher case involves the transfer of tax dollars to sectarian instruction in a way that a classroom poster does not.

The fourth — and, as Corbin has argued, the most likely actually to succeed — is the state-constitutional route. See Caroline Mala Corbin, New Judicial Federalism and the Establishment Clause: Classroom Ten Commandments as a Case Study in State Constitutional Protection, 104 B.U. L. Rev. (forthcoming 2026). Texas Article I, § 6 provides that "no human authority ought, in any case whatever, to control or interfere with the rights of conscience in matters of religion." Many state constitutions contain similar provisions that long predate incorporation and that the Supreme Court's Carson line does not touch. If the federal Establishment Clause has been reduced to a rough denominational-neutrality rule under Kennedy, the serious work of policing compelled religious funding will migrate to state courts interpreting state constitutions.

VI · Coda: The Candor Note

The neat version of the story runs: Nathan legalizes religious symbolism in public schools, Texas is not satisfied by symbols, TEFA is already oversubscribed and already channeling most of its dollars to Christian schools, and the doctrinal tools for resisting the resulting near-establishment are rusty but not broken. Law professors will like this story. It has an arc. The cultural victory in Nathan becomes the institutional scaffolding of the successor program, and defenders must improvise with Zelman, Larson, and state constitutions while Lemon lies in state.

The story is also plausible. But candor requires acknowledging that governing doctrine has not yet resolved the central question. The Supreme Court has never held that a facially neutral voucher program fails because the state enacted it for a sectarian purpose or administered it with denominational bias. Kennedy has made that question harder to litigate at the federal level; Carson has made it harder to dismiss at the threshold; and Nathan has now made it more concrete by showing that the underlying political mood is not a secret. Students graduating law school in 2026 will watch this doctrinal space being built in real time — starting with the April 24 hearing in Houston on the Muslim-exclusion lawsuit, and continuing for a decade of litigation after. They will have to decide, as judges, advocates, and scholars, whether the First Amendment contains enough architecture to prevent the cover story Carson makes available from becoming the reality. Nothing in the Court's current Religion Clauses jurisprudence compels an answer either way.

That is the best reason to teach this problem now.

Sources

Cases

Statute and program

Scholarship

Contemporaneous reporting (April 2026)

An essay for law students and law professors, April 2026.

← Read the first essay, "What Is Left of the Wall?" Browse the scholarly bibliography →