While the wall of church-state separation reached its apex in the 1985 Ball and Aguilar cases, other Supreme Court decisions around the same time started to put some cracks in the wall. In some cases, the court upheld indirect government funding of religion – that is, situations in which the government gave aid to an individual who then chose to use the aid to support a religious activity or group. In other cases, the court found direct government funding of religious groups and institutions to be constitutional in situations in which there was some assurance that the group would not use the aid for religious purposes.

Indirect Aid for Religious Institutions

A significant case that signaled an eventual move away from strict separationism was Mueller v. Allen (1983), which the high court decided two years before the influential Ball and Aguilar cases. The court’s decision revolved around the “child-benefit theory,” the seeds of which could be found in Everson and Board of Education v. Allen. According to the child-benefit theory, the government may give aid to religious organizations if the primary purpose and effect of the aid is to benefit children.

While Justices Rehnquist and White wrote several dissenting opinions in the 1970s laying out the foundation for this theory, no majority opinion explicitly adopted it until Mueller, a case in which the court retreated from separationism, due largely to changes in the court’s membership.

In Mueller, the Supreme Court upheld, by a 5-4 vote, a Minnesota statute that allowed parents to deduct from their state income taxes any money they spent on “tuition, textbooks and transportation” for their children attending public and private elementary and secondary schools. The court concluded that the program was constitutional because it assisted religious schools only indirectly by giving tax benefits to parents who decided to send their children to such private schools.

The court expanded this line of reasoning in Witters v. Washington Department of Services for the Blind (1986), which, as previously mentioned, involved a Washington state program that helped blind people by giving them tuition assistance to attend higher education or vocational training institutions. In the case, an individual wanted to use his benefits to pursue a career in the Christian ministry. The state would not allow him to do so, claiming that using public funds in this way would violate the Establishment Clause. But all nine justices rejected the state’s argument. The court explained that because the program provided so many secular educational and vocational choices, the state was not responsible for the student’s decision to pursue a religious education. Therefore, the court concluded, the Establishment Clause did not bar the student from using the tuition benefits for religious purposes.

The court extended the Witters rationale in Zobrest v. Catalina Foothills School District (1993), a case involving a hearing-impaired student who asked the school district to provide him with a sign language interpreter at his religious high school, pursuant to a state program to aid the hearing-impaired. The school district denied the student’s request on the ground that providing the interpreter would violate the Establishment Clause because the interpreter would need to communicate any religious material taught in his classes. But a five-justice majority rejected the school district’s argument. The interpreter, the court explained, did not provide an independent source of religious content; rather, the interpreter merely conveyed a teacher’s lessons. Moreover, the court said, it was the hearing-impaired child, not the religious school, that benefited from the interpreter. Thus, the court concluded, providing the interpreter would not violate the Establishment Clause because such indirect aid would not make the state responsible for the child’s religious instruction.

Almost a decade later, in Zelman v. Simmons-Harris (2002), the court issued its most influential decision on indirect aid for religious institutions. As previously mentioned, the Zelman case involved an Ohio program that gave various types of aid to students in Cleveland’s schools, some of which were among the worst-performing schools in the nation at the time. The lawsuit challenged the constitutionality of giving vouchers to low-income parents who chose to send their children to eligible private schools. Although all accredited private schools in the city could participate in the program, more than 80 percent of the participating private schools had a religious affiliation since the government’s voucher amount was not sufficient to cover the expensive tuitions often charged by secular private schools.

By a 5-4 vote, the court ruled that Ohio’s tuition aid plan was constitutional under the principle of “true private choice,” as articulated in Mueller, Witters and Zobrest. In his majority opinion, Chief Justice Rehnquist noted that, under the court’s precedents, a government program that funds religion indirectly is constitutional if the funding goes to individuals who have true private choice in deciding whether to use the funding for religious purposes. Rehnquist then wrote that to determine whether the Cleveland parents exercised true private choice, the court had to consider all the options available to the parents. In analyzing these options, the court found it significant that the Cleveland plan supported public charter schools and public magnet schools in addition to private schools. Including these public schools in the equation, the court found, meant that secular alternatives significantly outnumbered the available seats in religious schools. Because parents enjoyed this broad range of choices, the court concluded that the government was not responsible for the religious instruction of the children whose parents chose to use public funds to send them to religious schools.

The court’s decision in Zelman opened a wide range of possibilities for public funding of religious groups and institutions, including public funding of religious education at the elementary and secondary levels (though, as discussed above, some state constitutions specifically bar public funding of religious schools, even if that funding is indirect). Additionally,the court’s decision in Zelman has been used to justify government funding of other religious services outside of education. For example, in Freedom from Religion Foundation v. McCallum (2003), the 7th U.S. Circuit Court of Appeals held that the Establishment Clause allowed public funding of a religious group’s substance abuse treatment program because its beneficiaries exercised true private choice in determining whether to use public funds for religious or secular treatment.

Direct Aid for Religious Institutions

The high court also has found permissible several instances of direct government funding of a religious group or institution. The high court has eased restrictions on direct funding in two ways. First, it has held that religious institutions may receive secular aid, such as textbooks, as long as the aid is not used for religious purposes and the institution is not “pervasively sectarian,” such as a house of worship. Second, the court has expanded this rule, allowing aid to flow even to these religious institutions as long as there is some assurance that the support will not be used for religious purposes.

Ironically, on the same day that the Supreme Court decided the Lemon case, which created the three-part test later used to support separationism, the high court limited the reach of separationism in another case, Tilton v. Richardson (1971). In Tilton, the court upheld the Higher Education Facilities Act (HEFA), a federal statute that awarded construction grants to colleges and universities, including those affiliated with religious institutions. The lawsuit had charged that the Establishment Clause prohibited the government from giving these grants to four Connecticut colleges and universities affiliated with religious denominations. But a 5-4 majority upheld the grants to these religiously affiliated colleges because HEFA specifically barred all grant recipients, whether religious or secular, from using the grant-funded buildings for religious instruction, worship, or seminary or divinity school classes. Moreover, the court explained, the four religiously affiliated colleges in question were unlikely to violate this prohibition because, apart from their religious affiliations, these colleges were just like secular colleges in that “the schools were characterized by an atmosphere of academic freedom rather than religious indoctrination.”

Almost two decades later, in Bowen v. Kendrick (1988), the court applied the Tilton reasoning to the federal government’s funding of faith-based groups that provide sex education. The case involved a challenge to the 1981 Adolescent Family Life Act (AFLA), a federal program that allowed religious groups to receive grants to provide instruction on teen sexuality. The lawsuit alleged that AFLA violated the Establishment Clause because the statute did not prohibit the religious grant recipients from incorporating religious messages into their instruction. The high court, by a 5-4 vote, upheld the program, but with important limitations. Following its precedents, the court explained that the government may not directly fund pervasively sectarian organizations or specifically religious activities. Because it was unclear whether administration of the AFLA program had actually complied with these requirements, the Supreme Court sent the case back to the trial court to resolve these factual issues.

Ultimately, the U.S. government agreed to a settlement stipulating that all AFLA-funded groups must not incorporate religious references into their sex education programs and that these groups must not offer their programs in locations that provided space for religious worship or that featured religious symbols.

Nine years later, the court ruled in Agostini v. Felton (1997) that the government may directly provide aid to religious schools as long as the aid itself is secular, such as secular textbooks, and as long as the government provides safeguards to ensure that the school uses the aid for secular purposes. In Agostini, the court was asked to re-evaluate its earlier decision in Aguilar, the case that 12 years earlier had invalidated a federal program that provided remedial instruction in New York City private schools, including religious schools. The court’s basis for its ruling in Aguilar had been that the government would excessively entangle itself with religious matters (thus violating the Lemon test’s third criterion) in its efforts to ensure that its public school teachers did not provide religious instruction in religious schools.

In reconsidering the constitutionality of the federal program, Justice O’Connor, writing for the majority in Agostini, harkened back to her Aguilar dissent, in which she had argued that the majority in Aguilar had wrongly presumed that public school teachers sent to provide instruction in religious schools would incorporate religious content into their lessons. As she had written 12 years before, O’Connor explained that instead of presuming that public school teachers would violate government regulations, courts should presume instead that teachers, as public servants, would teach only the secular content they were charged to teach. With this reversed presumption, O’Connor concluded, the government need not excessively monitor its teachers, and the program therefore did not violate the Lemon test’s third criterion.

The Agostini ruling led to Mitchell v. Helms (2000), which further limited the reach of separationism. The case involved a constitutional challenge to a federal program that provided all schools, both public and private, with instructional materials and equipment, including computers and film projectors. Six justices voted in Mitchell to uphold the program, but they disagreed on the reasons for finding it was constitutional. Justice Clarence Thomas, joined by three other justices, claimed the program was constitutional because it provided secular benefits to schools without regard to whether they had a religious affiliation. Justices O’Connor and Stephen Breyer, while agreeing that the program was constitutional, did not join Thomas’ opinion because they thought it deviated too far from the Agostini ruling. Writing for herself and Breyer, O’Connor reasoned that the Agostini ruling allowed the government to provide secular aid to any institution as long as the aid did not directly support religious activities. Therefore, she explained, the government must take reasonable steps to ensure that aid recipients use the aid only for secular activities. O’Connor concluded that the federal program at issue in Mitchell satisfied this requirement, largely because of three features of the program: (1) it prohibited schools from using the aid for religious purposes; (2) it limited the aid to supplemental materials, such as computers, that were not already available in the schools; and (3) it did not transfer ownership of the materials to the schools.

In a Supreme Court decision with no majority opinion, the narrowest opinion supporting the result becomes the controlling law. In this case, O’Connor’s concurring opinion is narrower because, under her reasoning, a funding program would be constitutional only if it ensured that recipients used the aid strictly for secular activities. Thomas’ opinion is broader because it does not impose such limits on religious schools’ use of funds, as long as assistance is provided to religious and secular schools on an evenhanded basis. Thus, O’Connor’s narrower ruling now governs when the government may directly support religious institutions.