Concept
Aid to Religious Schools
Establishment
Aid to Religious Schools
Public money for religious schools raises two questions that now point in opposite directions: may a state include religious schools in a program of aid, and may it exclude them? The Establishment Clause answers the first, the Free Exercise Clause answers the second, and the second answer builds on the first. Both questions trace back to Everson v. Board of Education.
Everson announced that no tax in any amount, large or small, may be levied to support religious activities or institutions, and it nevertheless upheld a program that reimbursed the bus fares of parochial school children as part of a general scheme that also covered public school children.1 Its image for the principle was a wall between Church and State that “must be kept high and impregnable.”2 The majority treated the fares as a general government service, like police and fire protection, available to every schoolchild on equal terms. The dissenters found the benefit to religion too pronounced for the no-aid principle to survive. Tax-raised funds, they said, forced taxpayers to contribute to the propagation of opinions they disbelieve, exposed religious liberty to dependence on state money, and had already sparked political conflict.3 The second worry has Madisonian roots. Madison did not rest his opposition to establishment on its effect on dissenters alone; he argued that ecclesiastical establishments, “instead of maintaining the purity and efficacy of Religion, have had a contrary operation.”4
Justice Souter’s dissent in Zelman v. Simmons-Harris divides the cases since Everson into stages. In the first, running from 1947 to 1968, a general principle of no establishment, no aid held sway. In the second, which lasted roughly fifteen years, the Court tried to bar aid that could be diverted to the religious, as distinct from the secular, activity of a school. From 1983 the Court approved aid that was modest in amount, offered evenhandedly without regard to religious character, and channeled to a religious institution only by the genuinely free choice of private individuals. In Souter’s view a fourth stage has now arrived, in which the substantial character of the aid no longer matters and the criteria of neutrality and private choice have become examples of verbal formalism.5 That is one Justice’s reading, and it is a dissent. The majority saw an unbroken line of decisions rejecting challenges to neutral private-choice programs.6
The Court now sorts school aid by the route the money takes. A “no religious uses” approach still governs when aid flows directly to a religious institution. A private-choice approach governs when government gives aid to individuals who may use it at religious or secular institutions, and recent decisions have read even the direct-aid approach to permit more aid, reflecting the Court’s emphasis on religious-secular equality and individual choice.7 The direct-aid cases worried about the risk that aid would support religious activity, and that risk varied with the school. The Court reasoned that college students are less impressionable and less susceptible to religious indoctrination, that religion is less likely to permeate secular education where indoctrination is not a substantial purpose of the institution, and that less government surveillance, and so less entanglement, is therefore needed.8 The reason the route matters is attribution, which Zelman develops below: a school that the legislature pays is a school the government chose to fund, while a school that families pay is one the families chose.
The Free Exercise Clause now sits on the other side of the question. Beginning in 2017 with Trinity Lutheran Church v. Comer, the Court has turned its permissive Establishment Clause approach into a mandatory Free Exercise rule. It approved, indeed required, a direct money grant to a church without suggesting any Establishment Clause problem, which leaves little room for play in the joints between the two clauses.9 In Trinity Lutheran, Espinoza, and Carson v. Makin, the Court has held that denying aid to religious schools, when the same aid goes to secular private schools, violates the Free Exercise Clause.10
Private Choice Programs
In Zelman v. Simmons-Harris the Court upheld Ohio’s pilot scholarship program for Cleveland against an Establishment Clause challenge, and the decision remains the modern statement of when aid that reaches religious schools through parents is permissible. The majority found a program of “true private choice,” in which a neutral government program provides aid directly to a broad class of individuals, who in turn direct the aid to institutions of their own choosing.11
Ohio designed the program for a Cleveland school district that the state had taken over after a federal court intervened, and no one disputed that it had a secular purpose: educational help for poor children in a failing system. Purpose is a real limit. The Kentucky Ten Commandments statute failed because it had no secular legislative purpose, a result the Court reached by summary reversal over a dissent, and the Court has rejected the secular purposes a school district recited for Bible reading and the Lord’s Prayer.12 Ohio’s program cleared that gate, so the case turned on effect.
Scholarships went to parents according to financial need, and where the money was spent depended on where parents chose to enroll their children. Any private school in the district, religious or not, could participate if it met state standards, and public schools in adjacent districts could join as well, though none did.13 The numbers drew the challenge. Of the 56 private schools in the program, 46 were religious, and 96 percent of the scholarship students who attended private schools were in religious schools.14
To measure effect, the Court sorted its precedents into two lines. One concerns programs that give aid directly to religious schools; the other concerns programs of true private choice, in which government aid reaches religious schools only through the genuine and independent choices of private individuals. In Witters, for example, the Court rejected a challenge to a vocational scholarship that a student used to study to become a pastor. Any aid that ultimately flows to religious institutions, the Court observed, does so only as a result of the genuinely independent and private choices of aid recipients, and the program was available generally without regard to the sectarian-nonsectarian or public-nonpublic nature of the institution benefited.15 Zelman placed Cleveland’s program in that line.16
The division matters because of attribution. When government pays a religious school itself, government has chosen to fund that institution, and whatever religious mission the money advances is government’s doing. The direct-aid cases already carried that logic: the Pennsylvania statute in Lemon had the further defect of paying the church-related school itself, which distinguished Everson and Allen, where aid went to students and parents.17 When government gives aid to a broad class of people and each decides where to spend it, a religious school receives the money only because a family picked that school. The Court therefore treated any incidental advancement of a religious mission, or any perceived endorsement of a religious message, as attributable to the individual recipient and not to the government, whose role ends with the disbursement of benefits.18
From those cases the Court drew its test. Where a government aid program is neutral with respect to religion and provides assistance directly to a broad class of citizens who, in turn, direct the aid to religious schools wholly as a result of their own genuine and independent private choice, the program is not readily subject to challenge under the Establishment Clause.19 The test has two working parts, neutrality and genuine choice, and Ohio’s program satisfied both.20
Neutrality asks whether the program differentiates on the basis of the religious status of beneficiaries or providers, the touchstone of neutrality under the Establishment Clause.21 Ohio’s benefits went to a wide spectrum of individuals defined only by financial need and residence in the school district, and the program let those individuals exercise genuine choice among options public and private, secular and religious.22
Neutrality also reaches the terms of the aid, because a state that sweetened the religious option would be deciding for the family. Under Lemon, aid advanced religion if it was, or even might be, used for religious purposes. Under Mueller and Witters, government advances religion only if it gives more favorable aid for religious uses or otherwise creates an incentive to choose religious institutions over comparable nonreligious ones. Neutrality in this sense means no greater aid to religion than to the alternatives, not no aid to religion.23 Ohio’s terms tilted nothing toward religion. Cleveland’s public schools already received $7,097 per pupil, $4,167 of it from the state, so a student who stayed in public school with tutoring aid directed almost twice as much state money to that school as a scholarship student directed to a private one.24 Nothing in the record showed that the State deliberately skewed incentives toward religious schools.25
The second part of the test asks whether families have a real opportunity to choose secular options. Ohio’s menu was broad: a child could stay in public school, with or without publicly funded tutoring, use a scholarship at a religious or a nonreligious private school, or enroll in a community or magnet school.26 The Establishment Clause question, the Court said, is whether Ohio is coercing parents into sending their children to religious schools, and that question is answered by evaluating all the options Ohio provides, only one of which is to take a scholarship and then choose a religious school.27 Coercion, if it exists, is a feature of the whole menu, and it cannot be found by examining one item on it.
That framing disposes of the 96 percent figure. The figure describes where scholarship families chose to go; it does not show that anyone was coerced, and the Court was unimpressed by it, and by the fact that no suburban public school had agreed to participate.28 It also describes only one of the options Ohio provides. A rule keyed to the percentage would make validity depend on geography as well. Eighty-two percent of Cleveland’s participating private schools were religious, but so were 81 percent of the private schools across Ohio, and treating the number as decisive would permit the same neutral program in Columbus, where a lower share of private schools are religious, while forbidding it in Cleveland.29
The challengers also relied on Committee for Public Education & Religious Liberty v. Nyquist, and the Court distinguished it. The New York program was a different program: it gave a package of benefits exclusively to private schools and their parents, including direct grants to the schools, tax benefits unrelated to actual tuition, and reimbursements designed to give parents an incentive to choose sectarian schools, and it barred public schools and public-school parents from participating. The Court held that Nyquist does not govern neutral educational assistance programs that offer aid directly to a broad class of individual recipients defined without regard to religion.30 A program that shares New York’s features, with eligibility tied to private-school status, public schools excluded, and incentives toward sectarian choices, falls outside that holding.
The fifth step is where the majority and the dissent part company, because the answer depends on which options are counted.
Justice O’Connor joined the Court’s opinion and wrote separately to place it within the existing framework. She treated the decision as a refinement of the Lemon test that does not betray Everson: the Establishment Clause also requires that state aid flowing to religious organizations through the hands of beneficiaries do so only at the direction of those beneficiaries.31 She saw little question that the Cleveland program is neutral as between religious and nonreligious schools. Justice Souter, by contrast, would gauge neutrality by effects, asking whether a scheme covering a category of aid that may go to religious as well as secular schools favors a religious direction.32 Her own inquiry asks two questions: whether the program administers aid neutrally, without differentiating on the religious status of beneficiaries or providers, and, more importantly, whether beneficiaries have a genuine choice among religious and nonreligious organizations. A no to either condemns the program.33
Justice Souter’s dissent attacks both features. Everson, he argued, is still on the books, and a Court cannot leave it there and approve the Ohio vouchers; only by ignoring the meaning of neutrality and private choice can the majority pretend to rest on them.34 On neutrality, he argued that the majority’s “neutrality” is verbal and nothing more. It looked at every provision for educational opportunity, a method that would find neutrality in a voucher scheme for a district with no secular private schools at all, and it overlooked that tutoring aid for children who stayed in public school was capped at $324 per child while tuition vouchers paid up to $2,250.35 On choice, he noted that only 129 voucher students attended nonreligious private schools, whose total enrollment was just 510, and that the $2,500 tuition cap discouraged nonreligious schools with higher tuition, while the average tuition at participating Catholic schools was $1,592, well under the cap.36 Suburban public schools had no financial incentive to join the program, and none did, so for the overwhelming number of voucher children the only alternative to the public schools was religious. “A Hobson’s choice is not a choice.”37 Finally, he argued, the aid was substantial: it paid practically the full tuition for thousands of students, so the scholarships purchase everything that tuition purchases, instruction in math or indoctrination in faith.38
The majority answered in a footnote that Cleveland’s public schools already received $7,097 per pupil, so a student who stayed in public school with tutoring aid directed almost twice as much state funding to that school as a scholarship student directed to a private one. It added that Souter did not seriously claim that the program differentiates based on the religious status of beneficiaries or providers.39 The deeper disagreement concerns which options count. Souter’s criterion was genuinely free choice on the part of the individuals who choose, and he looked at the alternatives actually open to voucher families.40 The majority counted every option Ohio provides, and that choice of denominator decided the case.41
Justice Breyer, who joined Souter’s opinion, wrote separately to stress that publicly financed voucher programs risk religiously based social conflict, a risk he thought parental choice cannot significantly alleviate.42
Exclusion of Religious Schools
Zelman established that a state may include religious schools in a program of true private choice. Carson v. Makin asks the mirror question: may a state that has built such a program leave religious schools out? Chief Justice Roberts wrote for a six-Justice majority holding that Maine’s requirement that schools be nonsectarian to receive otherwise generally available tuition assistance violates the Free Exercise Clause.44
Under contemporary Establishment Clause doctrine, government generally may include religious beneficiaries in formally nondiscriminatory aid programs, most clearly when the aid is indirect and increasingly when it is direct. The question in Carson is what happens when a state excludes them anyway, choosing a more separationist course than the Establishment Clause requires.45 The answer is formally a Free Exercise holding, yet it rests on an Establishment Clause premise: the link between government funds and religious training is broken by the independent and private choice of recipients.46 That premise removes the Establishment Clause as a justification for exclusion. The state is left defending stricter separation than the Constitution demands, and the Court treated that preference as too weak to override the free exercise of the schools and families it burdens.
Maine’s program shows the setting. The state guarantees free public education, but many rural districts have no public high school, so the program pays tuition, up to a statutory cap, at the public or approved private school a family chooses.47 A participating private school must be accredited or approved and, since 1981, must also be nonsectarian. The state’s Department of Education reads “nonsectarian” to exclude a school that promotes a faith and presents its material through the lens of that faith. Two families sought to use the benefit at accredited religious schools and were refused. The lower courts upheld the requirement, and the First Circuit, distinguishing Espinoza, relied on two grounds that organize the discussion below: that Maine’s restriction is use-based rather than status-based, and that the benefit is the rough equivalent of a secular public-school education.
The Free Exercise Clause protects against indirect coercion and penalties on religious exercise, not just outright prohibitions, and recent precedent establishes that a state may not exclude otherwise eligible recipients from public benefits solely because of their religious character.48 In Trinity Lutheran, Missouri could not deny playground resurfacing grants to an otherwise eligible applicant simply because it was a church-operated daycare. In Espinoza v. Montana Department of Revenue, a no-aid provision barring aid to schools controlled by a church violated the Free Exercise Clause.49 The mechanism is the choice such a condition forces. The church in Trinity Lutheran remained free to operate as a church, but only at the cost of automatic and absolute exclusion from a program for which it was otherwise fully qualified, and a state that conditions a benefit that way has punished the free exercise of religion.50 A law that targets religious conduct for distinctive treatment survives strict scrutiny only in rare cases.51 A state’s wish to separate church and state more thoroughly than the Establishment Clause requires does not qualify as a compelling interest. The resulting rule leaves the state free to decline to subsidize private education, but once it chooses to subsidize, it cannot disqualify some private schools solely because they are religious.52
Maine tried to take its program outside that rule with the two characterizations the First Circuit had accepted. The first treated the benefit as the rough equivalent of a public education, which Maine may require to be secular. The Court answered from the program’s own design. Participating private schools differ from public schools in many material ways: they need not accept all students, they often charge tuition well above the state maximum, they need not employ state-certified teachers, they may be single-sex, and they are exempt from many curricular requirements. The one resemblance the program requires is secularism, so calling the benefit a secular equivalent of public education restates the exclusion as the definition of the benefit.53 The argument also proves too much. If a state could define a benefit to cover only secular recipients and then defend the exclusion by pointing to that definition, Trinity Lutheran and Espinoza could be avoided by drafting alone.
The second characterization said that Maine’s restriction turns on what a school would do with the money, not on what the school is. Trinity Lutheran and Espinoza did involve status-based discrimination, but neither suggested that use-based discrimination is less offensive to the Free Exercise Clause.54 Footnote 3 of Trinity Lutheran, which did not command a majority, had reserved the question: Chief Justice Roberts wrote, “We do not address religious uses of funding or other forms of discrimination.”55 Carson answers that question for a tuition program. Educating students in a faith is central to the mission of a private religious school, so a rule keyed to religious use reaches the activity the school exists to perform. A state that could disqualify schools for their anticipated religious use could also evade the status rule by recasting every exclusion as a use restriction. And a use line requires officials to examine whether and how a school pursues its religious mission, an inquiry that raises the entanglement and denominational-favoritism concerns taught earlier in this chapter.56
Locke v. Davey was Maine’s best precedent. Washington excluded students pursuing a devotional theology degree from a scholarship program, and the Court upheld the exclusion against a free exercise challenge. It accepted that the state could, consistent with the Federal Constitution, have let those students pursue such a degree; the question was whether Washington, under its own constitution, could decline to fund it without violating the Free Exercise Clause.57 The Court explained that there is room for play in the joints between the Religion Clauses, so that some state actions are permitted by the Establishment Clause but not required by the Free Exercise Clause. Carson reads Locke narrowly. Locke relied on a historic and substantial state interest in not funding the training of clergy, and Carson found no comparable tradition against aiding private religious schools generally.58 The programs also differed in operation. Washington’s let students attend pervasively religious schools and even take devotional theology courses, and what it withheld was funding for a degree in devotional theology, whereas Maine disqualified the schools themselves.59 Locke has not been overruled, but later decisions have significantly limited it.60
Two dissents defend the exclusion, and both do so by redefining the benefit. Justice Breyer stressed that the State has a substantial antiestablishment interest in ensuring that the public education it guarantees is religiously neutral, and that Maine’s tuition program exists to deliver that education to children in districts without secondary schools.61 Justice Sotomayor argued that the benefit is the public education guaranteed by Maine’s constitution and that the Establishment Clause has long required public education to be secular and neutral. It is therefore irrational, she said, to hold that the Free Exercise Clause bars Maine from directing tuition assistance toward the one kind of education the State may provide consistent with the Establishment Clause: a religiously neutral education.62 The majority’s reply is the one given above: participating private schools resemble public schools only in being secular, so describing the benefit as a secular equivalent of public education restates the exclusion rather than justifying it.
Read together, Zelman and Carson mean that a neutral private-choice program may include religious schools under the Establishment Clause,63 and that a state that has such a program may not exclude them under the Free Exercise Clause.64 Carson concerned a program in which parents pick the school, which is the setting in which the Establishment Clause reasons for exclusion are weakest. On an exam, take the Establishment Clause question first and the Free Exercise question second, and treat a status-versus-use argument as insufficient unless the exclusion is the narrow ministry-training exclusion that Locke upheld.