chevron-down Created with Sketch Beta.
July 01, 2006

Religious Liberty in America: A Rapid-Fire Overview

by Douglas Laycock

Three great sets of issues produce persistent controversy over the meaning of religious liberty in America: the funding of religiously affiliated activities; religious speech, with the allegation or reality of government sponsorship; and the regulation of religious practice.

Funding

The law on government funding of religiously affiliated activities has changed dramatically as the Supreme Court has struggled with two conflicting principles, announced in consecutive paragraphs in Everson v. Board of Education, 330 U.S. 1 (1947). First, the no-aid principle: "No tax in any amount, large or small, can be levied to support any religious activities or institutions." Id. at 16. And second, the nondiscrimination principle: Government "cannot exclude individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Nonbelievers, Presbyterians, or the members of any faith, because of their faith, or lack of it, from receiving the benefits of public welfare legislation." Id.

In the beginning, the no-aid and nondiscrimination principles did not conflict; earmarked taxes for the support of colonial-era churches violated both principles. There is still consensus that government should not fund the religious functions of churches. But the modern cases are very different. Today, government is funding some secular services, and it offers the money on equal terms to religious and secular providers alike. In that context, the Court has to choose. Either government money will flow though to religious institutions or students in religious schools and patients in religious hospitals will forfeit instruction or services that the state would have paid for if the students and patients had chosen secular schools or hospitals.

The nondiscrimination principle prevailed until 1971, when the Court changed direction. In Lemon v. Kurtzman, 403 U.S. 602 (1971), the Court for the first time struck down a funding program, holding that states could not subsidize teachers' salaries in religious schools. This no-aid principle predominated until early 1985. But even in this period, it never completely triumphed. The Court restricted government aid to elementary and secondary schools but permitted aid to religious colleges. In K-12, states could provide books (but not maps) and bus rides to school (but not for field trips). Perhaps the most absurd, the Court prohibited government-funded remedial instruction to low-income students but permitted the same instruction in vans parked nearby. Few of the justices really believed in these distinctions, but some were unwilling to overrule earlier cases and some were still trying to preserve something of both competing principles: no aid and no discrimination.

Beginning in 1986, the Court progressively elevated the nondiscrimination principle and subordinated the no-aid principle. Since then, the Court has upheld six programs that permitted government funds to reach religious institutions. it has invalidated none. Four Lemon-era decisions have been overruled in whole or in part.

The most important of the decisions since 1986 is Zelman v. Simmons-Harris, 536 U.S. 639 (2002), which upheld the use of vouchers to pay tuition at any public or private school, including religious schools. Zelman reasons that the government funding supports students, students and their parents decide where to spend the money, and there is no state action in their choice of school. The Court also upheld long-term loans of equipment to religious schools if the equipment is distributed to all schools on the basis of enrollment. Mitchell v. Helms, 530 U.S. 793 (2000). Lemon's ban on direct cash grants to religious institutions is still good law, but in the school context, legislatures can deliver as much money as they are willing to spend in the form of vouchers.

Although Zelman largely eliminates federal constitutional barriers to voucher plans, these plans remain difficult to enact politically and are subject to state law challenges. Most state constitutions have detailed restrictions on financial aid to sectarian schools. Conflicting decisions of the state supreme courts show that these clauses can be interpreted either way.

Some sates aid secular private education but not religious education. It had been thought that extending these programs to religious schools would violate the federal Establishment Clause of the First Amendment, but after Zelman, that is clearly not true. And the Supreme Court generally says that government cannot discriminate against religion. Voucher supporters thought it followed that states violate the Free Exercise Clause when they fund private secular education but refuse to fund similarly situated religious education.

The Supreme Court rejected such a claim in Locke v. Davey, 540 U.S. 712 (2004). Under Davey, government funding of religious schools is permitted but not required, and with respect to funding, government is permitted to discriminate against religion. Despite the traditional suspicion of government discretion in constitutional law, government now can fund religious education, or not fund it, or fund it on condition that the student or the school comply with special regulations that apply only to those who accept government money. Parts of the opinion suggest that its rule is confined to programs for the training of clergy. Other parts suggest that it will apply generally to any exclusion of religious institutions from state funding programs.

For most of the twentieth century, this dispute over funding religious institutions was confined to schools. Religious hospitals and social service agencies received government funds with little controversy. That has changed with the Bush administration's Faith-Based Initiative. These proposals increased the visibility of government grants to religious charities, and they introduced new protections for the autonomy of religious charities accepting government funds.

Zelman suggests that there is no constitutional barrier to government funding of religious charities. But some social services may require direct grants to agencies instead of vouchers to the intended beneficiaries, because legislators will be reluctant to give vouchers to such vulnerable groups as neglected children, mentally ill persons, or drug addicts. And these government programs are not so well funded that government can support all providers of services; government has to choose which agencies to support. So these programs may present questions of discretionary direct grants to religious charities, with the resulting risks of favoritism and religious discrimination. And there certainly will be litigation over the right of religious charities to hire persons of their own faiths for government-funded positions.

Speech

In Engel v. Vitale, 370 U.S. 421 (1962), and School District v. Schempp, 374 U.S. 203 (1963), the Supreme Court held that public schools violate the Establishment Clause when school officials lead students in prayer or Bible reading. These cases provoked a religious backlash while simultaneously raising expectations that non-Christians no longer would be subjected to government-sponsored Christian religious observances. The result has been endless efforts to disguise government sponsorship of school prayer. In Santa Fe Independent School District v. Doe, 530 U.S. 290 (2000), the Court held that school boards cannot conduct student elections to decide whether to have prayer. Engel and Schempp involved prayer in classrooms, but more recently the Court has invalidated prayer at graduation in Lee v. Weisman, 505 U.S. 577 (1992), and at athletic events in Santa Fe. But in March v. Chambers, 463 U.S. 577 (1983), the Court refused to invalidate prayer at legislative sessions, principally because of its long tradition.

The secular side opened a second front when it began challenging government-sponsored religious displays. The Court held that public schools cannot display the Ten Commandments in classrooms in Stone v. Graham, 499 U.S. 39 (1980), and that a county cannot display a Nativity scene at a central location in its courthouse in Allegheny County v. ACLU, 492 U.S. 573 (1989). But it permitted a government-sponsored Nativity scene displayed alongside "secular" symbols of Christmas, such as Santa Claus and reindeer, in Lynch v. Donnelly, 465 U.S. 668 (1985). It similarly permitted a Christmas tree next to a menorah and a salute-to-liberty sign in Allegheny County.

Most recently, the Court decided that Texas can maintain a large granite monument displaying the Ten Commandments on the lawn of its state capitol, Van Ordern v. Perry, 545 U.S. 677 (2005), but that two Kentucky counties cannot display the Ten Commandments on courthouse walls, surrounded by patriotic documents and a statement claiming that the Ten Commandments are the foundation of the Western legal tradition. McCreary County v. ACLU, 545 U.S. 844 (2005). Only Justice Stephen Breyer supported both results. He approved the Texas display because it had been in place for forty years before it aroused controversy, which suggested, that the display contained both a religious and secular message and that the secular message predominated in public perception. He rejected the more recently installed Kentucky displays because county officials had stated clearly that their purpose in installing them was to promote Christianity.

The underlying conceptual dispute in these cases is about the scope of the government's obligation to be neutral. Must government be neutral toward religion only when it coerces someone's religious belief or behavior, or must government also be neutral in what it says about religion? The Court has found this distinction irrelevant to prayer. It has said that if there is a public event that many people attend for secular reasons, and someone offers a prayer at that event, government impermissibly is coercing persons who attend to participate in prayer. But that argument only goes so far; it is hard to find coercion in a passive display. So the Court also says government may no "endorse" a religious viewpoint; government must be neutral even in what it says about religion. This argument over endorsement, while at the heart of the dispute over government-sponsored religious speech, is largely irrelevant in other religious liberty contexts.

Although the Supreme Court tightly restricts government-sponsored religious speech, it vigorously protects religious speech by private citizens. Religious speakers have full rights of free speech on government property as long as they speak without government sponsorship. They must be treated the same as other speakers, with no special access to facilities or to government-assembled audiences. This right to religious free speech extends even to elementary schools. Good News Club v. Milford Central School, 533 U.S. 98 (2001). The Court has never held, in any context, that religious speech by private actors is subject to restriction because of its religious content.

The Court's rules on religious speech have been remarkably stable for half a century. Yet from 1994 to 2005, these rules had the support of only two justices, Anthony Kennedy and Sandra Day O'Connor. Because the others were divided, Kennedy and O'Connor generally had six votes to prohibit government sponsorship of religious speech and at least five votes to invalidate government discrimination against private religious speech. But now O'Connor is gone. If President George W. Bush has accomplished what he hoped with his first two Court appointments, Kennedy is the new swing vote on these issues. And Kennedy had one important disagreement with O'Connor: he distinguished religious displays, which passersby may ignore, from religious exercises, which often trap a captive audience.

Many of the same political forces that support government funding of religious schools also support government-sponsored prayers. Why has the Court changed its mind on funding but not on prayers? The explanation again lies with Kennedy and O'Connor, who see these two sets of cases as very different. In the funding case, each family gets a voucher, and each family can choose a religious or secular school. In the private religious speech cases, each speaker can decide what to say, and each person around him can decide whether to listen. But prayer at a government function requires a collective decision. Either there will be prayer for everyone, or there will be prayer for no one. If there is prayer, there will be only one, and it will be in a form more consistent with some religious traditions than with others. No one gets to make an individual choice. By permitting vouchers and protecting religious free speech but restricting government-sponsored prayers, Kennedy and O'Connor protected the right of individual choice.

Practice

Regulation of religious practice is the most fundamental and least understood of the three sets of issues. Only in these cases can persons be threatened with civil or criminal penalties for practicing their religion.

From 1963 to 1990, the Supreme Court's position was that when a regulation burdens religious practice, government must either exempt the religious practice from the regulation or show that applying the regulation to the religious practice is necessary to serve a compelling government interest. In Sherbert v. Verner, 374 U.S. 398 (1963), the leading case, the Court held that a state could not refuse unemployment compensation to a Sabbatarian who lost her job because she was unavailable for work on Saturdays.

Despite this rule, the Court did not actually exempt many religious practices from regulation. Prison and military regulations restricting religious practices in those settings were given much deference And the Court found compelling interests in enforcing the draft, collecting taxes, and prohibiting racial discrimination in education. These findings of compelling interest were entirely plausible. In each of these cases, there were reasons of secular self-interest to adopt, or falsely claim, the religious belief that led to the exemption. But some commentators think that these interests were not compelling and that the Court was never serious about exempting religious practice from nonessential regulation.

In Employment Division v. Smith, 494 U.S. 872 (1990), involving a Native American employee fired and subsequently denied unemployment benefits because of his use of peyote in a religious ritual, the Court changed the rules. Smith introduced an additional requirement for litigants seeking religious exemptions: Is the law that burdens religious exercise “neutral” and “generally applicable”? If so, the burden on religion requires no justification. If not, the burden on religion is subject to the compelling interest test.

The Supreme Court has decided only one subsequent case under Smith. Church of the Lukumi Babalu Aye v. City of Hialeah, 508 U.S. 520 (1993), involved an Afro-Caribbean religion that sacrifices small animals to its gods. Hialeah argued that it had enacted a generally applicable ban on sacrifice. The church argued that the ordinances were a ban on killing animals for religious reasons, carefully drafted not to prohibit any killings of animals for secular reasons. The Court unanimously agreed with the church, holding that the ordinances were neither neutral nor generally applicable and that they served no compelling government interest.

Lukumi gives substance to Smith’s requirements of neutrality and general applicability, but the meaning of those requirements remains sharply disputed. Government lawyers claim that nearly every law is neutral and generally applicable and that the only exceptions are laws designed deliberately to single out a religious practice. This argument has some support in Lukumi’s facts and in the Smith and Lukumi opinions. Lawyers for religious claimants say that to be generally applicable, a law must apply to all examples of the regulated conduct, with no, or very few, exceptions. This argument has some support in Smith’s facts, more support in the Smith and Lukumi opinions, and much support in the way those opinions distinguish Sherbert and other earlier cases that have not been overruled.

If the government lawyers are right, Smith provides very little protection for religious liberty. If the religious organizations are right, Smith provides substantial protection for religious liberty, but that protection is less inclusive, more complicated, and harder to invoke than the protection of Sherbert.

Smith provoked widespread disagreement among other branches and levels of government. In 1993, Congress enacted the Religious Freedom Restoration Act (RFRA) in an attempt to restore the Sherbert rule as a matter of statutory right. In City of Boerne v. Flores, 521 U.S. 507 (1997), the Court held RFRA, as applied to the states, beyond Congress’s power to enact under Section 5 of the Fourteenth Amendment. But RFRA remains in effect as applied to the federal government. In Gonzales v. O Centro Espirita Beneficiente Uniao do Vegetal, 126 S. Ct. 1211 (2006), the Court gave RFRA full and vigorous scope. In that case, involving religious use of a mild hallucinogen prohibited by the federal Controlled Substances Act, the Court unanimously held that the government had failed to prove its claim of compelling interest, and it unanimously rejected the government’s claim that it need only point to congressional fact-finding in the course of enacting the Controlled Substances Act. The government’s interpretation would have nullified RFRA’s allocation of the burden of proof. The Court’s holding makes RFRA an important protection for religious liberty.

Since 1992, thirteen states have adopted state RFRAs, and at least twelve states—arguably as many as seventeen—have interpreted their state constitutions in ways more consistent with Sherbert than with Smith. So, in one way or another, a majority of states have rejected Smith. But there has been remarkably little state court litigation under these provisions.

Most recently, in 2000, Congress also enacted the Religious Land Use and Institutionalized Persons Act (RLUIPA). RLUIPA protects churches against local zoning laws that often make it difficult to buy or rent a place of worship. It applies only when the burden on religion would affect interstate commerce or when the zoning law is administered in an individualized rather than a generally applicable way. These restrictions are designed to ensure that RLUIPA fits within specific congressional powers, thus avoiding a charge of excessive scope that led to RFRA’s invalidation. RLUIPA also protects the free exercise rights of prisoners in state prisons that accept federal funds. State officials have bitterly resisted RLUIPA. In Cutter v. Wilkinson, 544 U.S. 709 (2005), the Supreme Court unanimously rejected a claim that the prison provisions violate the Establishment Clause. States also are arguing in the lower courts, mostly unsuccessfully, that both the prison and the land use provisions exceed the scope of powers delegated to Congress.

Continuing Controversy

Deeply inconsistent constitutional visions make this area of law especially susceptible to the effect of new Supreme Court appointments. The cases on government religious displays are in jeopardy. Another Republican appointment could roll back some of the school prayer cases. Another Democratic appointment could overrule Zelman and invalidate vouchers again. Smith is more likely to be eroded than overruled; the disagreement there does not yet track party lines. Especially on the Establishment Clause issues, too many justices are interested in promoting or restricting religion. Not enough are interested in protecting liberty for believers and nonbelievers alike.

As published in Human Rights, Summer 2006, Vol. 33, No. 3, p.3-8, 24.

Douglas Laycock

Douglas Laycock is the Yale Kamisar Collegiate Professor of Law at the University of Michigan Law School in Ann Arbor.