Establishment Clause Doctrine: Supreme Court Jurisprudence

The First Amendment to the U.S. Constitution provides that ‘‘Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof.’’ This single, integrated statement limits the power of the federal government in religious life. But it contains two distinct principles—free exercise and nonestablishment—that the courts have often treated separately. The Free Exercise Clause prohibits governmental interference with the religious practice of private individuals and groups; as such it resembles other personal liberties in the Constitution such as freedom of speech. But the second component, the Establishment Clause, is unusual among constitutional limitations. Government typically may promote or embrace particular ideas as long as it leaves people free to dissent, but the Establishment Clause prohibits government from promoting religious doctrines in certain ways, and from creating certain relationships with religious institutions. These mandates may be crucial to both religious freedom and social peace. But because they restrict government more than with respect to nonreligious ideas, the Establishment Clause has been especially difficult and controversial to interpret in the circumstances of modern, pervasive government. Just what promotion of and involvement with religion are prohibited?

Background and Adoption

The ‘‘establishment[s] of religion’’ familiar to the founders had three main features, present in varying degrees in the Church of England and in the colonial establishments (Anglicanism in the South, Puritan Congregationalism in New England). One feature was governmental coercion to support the favored church: mandatory membership, mandatory financial contributions, and even prohibitions on practicing other faiths and preaching their doctrines. Prohibitions on dissenting faiths largely disappeared by the time of the founding, with the rise of the ‘‘free exercise’’ ideal; but requirements of financial support for religion continued. A second feature was government regulation of the established church, including selection of clergy and oversight of doctrine and liturgy (for example, Parliament approved the successive Books of Common Prayer). The third feature was symbolic recognition of the favored faith, through official ceremonies and titles. The monarch headed the Church of England, bishops held seats in the House of Lords, and state ceremonies incorporated the established faith.

The practice of establishment rested on both theological and political rationales: protecting the true faith, maintaining social unity, and promoting ideals that would make citizens virtuous. Likewise, the rise of disestablishment in the colonies and the early American republic rested on both theological and political arguments. Certain Christian leaders from Roger Williams through founding-era Baptists argued that the union of religion and government would interfere with the individual’s free response to God and would corrupt (in Williams’s words) the ‘‘garden’’ of the church with the ‘‘wilderness’’ of the world. Meanwhile, Enlightenment figures such as Thomas Jefferson and James Madison attacked establishments on the ground that they threatened social peace and minority rights—‘‘destroy[ing],’’ in Madison’s words, ‘‘moderation and harmony . . . amongst [society’s] several sects,’’ and ‘‘degrad[ing] from the equal rank of Citizens all those whose opinions in Religion do not bend to those of the Legislative authority.’’

The latter quotes come from Madison’s ‘‘Memorial and Remonstrance’’ against a 1785 proposal in Virginia to renew tax assessments in support of clergy. Madison masterfully combined political arguments with theological (for example, that ‘‘ecclesiastical establishments’’ have undermined ‘‘the purity and efficacy’’ of religion). Disestablishment in America, unlike its counterpart in France, was generally not hostile to religion, and indeed took much of its inspiration from Baptists and other dissenting Protestant groups. Most Americans continued to believe that religion was essential to civic virtue (an ‘‘indispensable suppor[t]’’ for ‘‘political prosperity,’’ in George Washington’s words). They simply concluded, in increasing numbers, that governmental promotion of religion would weaken, rather than strengthen, it in its social role.

By the late 1780s, these arguments had led to the elimination of taxes for clergy support in the majority of states. Virginia, the most notable case, rejected even a liberalized proposal that would have allowed a citizen to direct to which sect his payment would go. But the pattern was not uniform: clergy taxes remained in three New England states well after the adoption of the Constitution (the last system, in Massachusetts, was repealed in 1833). Like the defeated Virginia proposal, these taxes permitted dissenters to support their own clergy or opt out altogether: yet even these nonpreferential systems were typically referred to as ‘‘establishments.’’

Thus, when the First Amendment was ratified in 1791, states remained divided over the basic principle of disestablishment (in contrast, all had some commitment to basic free exercise for dissenters). But this division did not matter, for the amendment’s purpose was to restrain the new federal government from interfering in religious matters; the Bill of Rights left untouched the states’ powers over religion. The brevity of congressional debates over the religion provision reflects that the founders were not thinking about the many questions that might arise if state governments were limited from interacting with religion. Despite this brevity, the First Amendment prohibition on laws ‘‘respecting an establishment of religion’’ has some clear meanings. As noted above, a tax to support religion almost certainly qualified as an establishment even when it was nonpreferential among denominations or faiths; this proposition is confirmed by the fact that the Senate rejected several proposals to prohibit establishment of ‘‘one religious sect or society’’ and instead ultimately approved the broader ban on ‘‘an establishment of religion.’’ Moreover, given the Framers’ agreement on the inviolability of religious conscience, they almost certainly rejected federal power to coerce anyone to practice religion.

A far more complicated historical question, however, concerns noncoercive acknowledgements of religion: official prayers, religious proclamations, and religious symbolic displays. Such practices can implicate several concerns underlying disestablishment— diluting the religious message to a politically acceptable content, alienating citizens who dissent from it—and some in the founding generation criticized the practices in the name of a stricter separation of church and state. President Jefferson refused to follow his predecessors in issuing Thanksgiving proclamations, and Madison in a series of letters and memoranda in the 1820s and 1830s attacked such proclamations as well as congressional prayers. Yet the practices were commonplace in the early republic: the First Congress initiated daily prayers, and the congressional chaplaincy and Washington began the nearly unbroken practice of presidential Thanksgiving proclamations. Just how ecumenical such expressions were required to be is also ambiguous. Although most of them referred in general terms to God or the ‘‘Supreme Being,’’ many people in the founding generation would likely have agreed with the narrower position of Joseph Story (in 1835) that the First Amendment was intended ‘‘not to countenance, much less to advance Mahometanism, or Judaism, . . . but to exclude all rivalry among Christian sects.’’

As the nineteenth century progressed, formal disestablishment mixed even more with government promotion of a generalized Protestantism, producing what historians have called a ‘‘de facto establishment.’’ A series of revivals driven by evangelicals, especially Methodists and Baptists, made Americans much more thoroughly and explicitly religious, and this widespread faith pervaded public discourse. Much of this expression occurred within state governments, but it nonetheless reflected Americans’ understanding of the disestablishment ideal. A different controversy arose in the mid-1800s when Catholics departed the new, Protestant-dominated public school system and sought equal funding for their own schools. Most states banned such support in their own constitutions—a move that has been controversial ever since—thus extending the founding-era ban on specifically religious taxes to prohibit even general tax support for education from including religiously affiliated schools.

State Establishments, Federalism, and Incorporation

As enacted, the First Amendment, like the rest of the Bill of Rights, bound only the federal government. But the Supreme Court has held, of course, that the adoption of the Fourteenth Amendment in 1868 made most of the Bill of Rights applicable to state and local government actions. The incorporation of the Establishment Clause was declared in Everson v. Board of Education (1947). Although several commentators have argued that the Fourteenth Amendment was not intended to incorporate the Bill of Rights, the Court continues to adhere emphatically to incorporation (see, for instance, Wallace v. Jaffree [1985]).

Special challenges have been raised to the incorporation of the Establishment Clause. The clause, it is asserted, was meant not just to prohibit establishment of a national church, but as much to prohibit federal interference with the state establishments still existing in 1791. This interpretation could explain the peculiar phrasing ‘‘no law respecting an establishment.’’ In Justice Thomas’s words, ‘‘the Establishment Clause is best understood as a federalism provision—it protects state establishments from federal interference but does not protect any individual right,’’ which in turn ‘‘make[s] incorporation of the Clause [to limit state governments] difficult to understand’’ (Elk Grove Unified School Dist. v. Newdow [2004] [concurring in the judgment]). Thomas would incorporate, at most, the aspects of non-Establishment that protect individuals from direct coercion. But no other justice has joined him; the Court implicitly follows Justice Brennan’s argument that a fully incorporated Establishment Clause makes sense ‘‘as a co-guarantor, with the Free Exercise Clause, of religious liberty’’ for individuals against all levels of government (Abington School Dist. v. Schempp [1963] [concurring opinion].

Modern Interpretation: The Lemon Test and Church–State Separation

The modern Court’s Establishment Clause interpretations reflect a fundamental tension. As American society has become more religiously diverse—with Christians joined by substantial numbers of Jews, Muslims, and Eastern-religion adherents as well as atheists and believers in generalized spirituality—any religious involvement by government appears more partial to one group of citizens and thus threatens some of the general harms of establishment: division among faiths and alienation of dissenters. At the same time, given the pervasive activity of the modern welfare state, to keep government wholly separate from religion can discriminate against religious citizens and artificially restrict religion’s role in public life.

The tension appeared in the very first modern Establishment Clause case, Everson (1947), where a local government reimbursed parents’ costs for sending their children to parochial schools on public buses. The majority opinion by Justice Hugo Black began by emphasizing a strict ‘‘wall of separation between Church and state’’ (quoting Thomas Jefferson) that prohibited ‘‘any tax in any amount . . . to support any religious activities.’’ Next, however, the opinion argued that the First Amendment prohibited excluding individuals, ‘‘because of their faith, or lack of it, from receiving the benefits of public welfare legislation,’’ and it ended up narrowly approving the reimbursements. Strict church–state separation conflicted with and gave way to the equal participation of religion in the widespread programs of the welfare state.

In Lemon v. Kurtzman (1971), however, the Court formulated a general Establishment Clause test that, as applied, elevated strong church–state separation over equal religious participation. Lemon required that a law (1) have a secular legislative purpose, (2) have a primary effect that neither advanced nor inhibited religion, and (3) not create an excessive entanglement between church and state. Under this framework, the Court in the 1970s and early 1980s struck down several programs of government aid to religious education through a ‘‘Catch 22’’: if the statute allowed aid to benefit religious teaching in church-related schools, it violated the ‘‘primary effect’’ prong, but if it imposed restrictions on religious uses, policing those restrictions created ‘‘excessive entanglement.’’ The no-advancement and no-entanglement prongs both required separation from the religious school.

The Lemon test, or nascent versions of it, also invalidated religious exercises and symbols in public schools: classroom prayers (Engel v. Vitale [1962]), classroom Bible readings (Schempp), postings of the Ten Commandments (Stone v. Graham [1980]), and classroom ‘‘moments of silence’’ (Jaffree). These decisions found it insufficient for Establishment Clause purposes that the exercises were formally voluntary and were general enough in content to encompass many different faiths. The clause, the Court said, reflected the broader ‘‘belief that a union of government and religion tends to destroy government and to degrade religion’’ (Engel). Thus, the government should not promote religion even in noncoercive or denominationally neutral ways; government must be neutral toward religion altogether.

Although official religious practices in public schools could be invalidated based on their religious ‘‘effects,’’ in several cases (Stone, Jaffree) the Court struck them down under the first Lemon prong for lack of a ‘‘secular purpose.’’ To invalidate a law solely because of an impermissible legislative intent, without regard to a forbidden effect, is an unusual principle in constitutional law, and reflects a prophylactic approach: the Court seeks to keep the government from even trying to involve itself in religious matters.

These decisions also led some observers to suggest that government might violate the Establishment Clause if it relied too heavily on religious doctrines as the basis even for legislation on facially nonreligious matters such as abortion or other issues of sexuality. See, for instance, Webster v. Reproductive Health Services (1989) (Stevens J., dissenting) (arguing that abortion law based on declaration that life begins at conception ‘‘serves no identifiable secular purpose’’). The Warren Court invalidated an Arkansas law forbidding the teaching of evolution in public schools—a remnant of the Scopes-era fundamentalist movement—on the ground that the law could not ‘‘be justified by considerations of state policy other than the religious views of some of its citizens’’ (Epperson v. Arkansas [1968]. Since then, however, the Court has generally affirmed the legitimate role of religious activism in politics, noting that ‘‘[a]dherents of individual faiths and individual churches frequently take strong positions on public issues’’ and have the right to do so ‘‘as much as secular bodies’’ (Walz v. Tax Commission [1970]). In the same vein, the fact that denials of government funding for abortions ‘‘coincide with the religious tenets of the Roman Catholic Church does not, without more, contravene the Establishment Clause’’ (Harris v. McRae [1980]).

Lemon’s Decline and Less Separationist Alternatives

The Lemon test came under substantial attack from some justices and commentators. Some found the test too indeterminate, especially in the various rulings on when a law’s ‘‘primary effect’’ was to advance religion or promote a secular goal such as education. But other objections ran deeper.

Some critics objected to the requirement that government not advance any religious proposition. They argued that the demand of neutrality was contradicted by the long historical tradition of government endorsements of religion: legislative prayers, presidential proclamations, symbolic displays, official mottoes, and statements such as ‘‘in God we trust’’ on coins and ‘‘under God’’ in the Pledge of Allegiance. As Justice Kennedy claimed, ‘‘A test for implementing the protections of the Establishment Clause that, if applied with consistency, would invalidate longstanding traditions cannot be a proper reading of the Clause’’ (County of Allegheny v. American Civil Liberties Union [1989] [concurring in part and dissenting in part]). Indeed, some critics observed, landmark founding- era documents themselves argued for religious liberty on religious grounds—‘‘Almighty God hath created the mind free,’’ began Jefferson’s bill establishing religious freedom in Virginia—and thus, paradoxically, might have been ruled invalid under the neutrality or no-advancement tests. Some critics argued that because many religious ideas concern matters of morality and justice, government cannot be neutral toward them but must embrace or reject them as it does secular ideas.

Critics of ‘‘neutrality’’ tended to train their attack on the decisions invalidating government-sponsored religious ceremonies or symbols. The Court turned to a historical approach in Marsh v. Chambers (1983), upholding the practice of legislative prayer on the basis that it had been instituted by the First Congress— three days before adoption of the Religion Clauses—and the delegates could not have ‘‘intended the Establishment Clause to forbid what they had just declared acceptable.’’ Relying solely on this history, the Court did not apply Lemon or any other analytical test. It used history not to develop a general legal principle reflecting the clause’s original meaning, but simply to validate the specific practice in question. Accordingly, it was unclear if or how Marsh would apply to any other issue, and the Court has never followed the Marsh approach again.

Then-Associate Justice Rehnquist offered an alternative principle to Lemon: the Establishment Clause was intended only ‘‘to prevent the establishment of a national religion or the governmental preference of one religious sect over another. (Jaffree [Rehnquist, J., dissenting]). He relied on quotes from the congressional debates on the First Amendment, as well as on many founding-era official statements endorsing religion in generalized form. But no one else on the modern Court ever limited the Establishment Clause solely to forbidding preferences between sects. As Professor Douglas Laycock pointed out, the first Congress’s rejection of drafts that would have prohibited only preferences for one sect further undermines the no-preferences position as a general Establishment Clause rule. Recently, however, Justice Scalia has urged the more limited position that the government must be neutral toward religion in financial aid cases, but may favor religion in noncoercive acknowledgments as long as they are nonpreferential among monotheistic faiths (McCreary County v. American Civil Liberties Union [2005] [dissenting opinion]).

Indeed, the more persistent alternative to Lemon is the no-coercion test: government may promote or acknowledge religion (perhaps only in generalized form) but may not coerce anyone to profess or participate in religion. The test fits much of the historical record, for most official acknowledgments of religion appear in noncoercive settings: nonbinding proclamations, or displays on government property. But proponents of the neutrality approach answer that limiting the Establishment Clause’s reach to coercive practices would make it redundant with the Free Exercise Clause, which (arguably) prohibits coercion to practice religion as well as not practice religion. In addition, they argue, noncoercive acknowledgments still threaten the harms at which the Establishment Clause was aimed—social division, alienation of dissenting citizens, corruption of religious integrity—especially in a society as pluralistic in religious views as America.

After several years of raising doubts about Lemon, the Court relied on the non-coercion test in Lee v. Weisman (1992), striking down officially sponsored prayers at public school graduation ceremonies. But Weisman specifically declined to overrule Lemon, and it referred to noncoercion as only the undisputed ‘‘minimum’’ guarantee of the Establishment Clause. Moreover, proponents of the test divided over its meaning, with Justice Kennedy’s majority opinion finding ‘‘subtle’’ peer pressure on dissenting students to participate in the prayer, while Justice Scalia’s dissent insisted that only coercion ‘‘by force of law and threat of penalty’’ was unconstitutional. Thus, the no-coercion test remained doubtful in both its status and its scope as of early 2006.

The other chief criticism of the Lemon test was that despite its invocation of ‘‘neutrality,’’ it actually produced hostility and discrimination against religious ideas and religiously motivated citizens. This objection primarily targeted the decisions on financial aid, which disqualified religious schools from the taxfinanced support available to public and secular private schools. Those rulings treated public education was ‘‘neutral’’ because it explicitly advanced neither religious nor antireligious views. But critics argued that education limited to secular viewpoints competes with education based on religious viewpoints, and therefore providing equal aid to religious schools coincides more with Establishment Clause values of neutrality and individual choice in religious matters. These arguments, unlike the criticisms of the school prayer decisions, eventually made substantial headway on the Court.

Recent Case Law: Equal Access for Religion, But No Government Endorsement

The Rehnquist Court (1986 to 2005) retreated from strong church–state separationism in its Establishment Clause rulings, but only in part.

The major shift came in financial aid cases, where the Court more and more allowed religious institutions to participate on equal terms with nonreligious institutions. The ‘‘equal access’’ principle developed with respect to the less controversial issue of whether religious groups could meet voluntarily on public school grounds on the same terms as other student groups. Beginning with Widmar v. Vincent (1981), involving evangelical students at a public university, the Court held that permitting religious meetings along with other meetings did not violate the Establishment Clause and indeed, in several cases, was required by the Free Speech Clause principle of viewpoint neutrality. See also, for example, Board of Education v. Mergens (1990) (high school groups); and Good News Club v. Milford Central School (2001) (elementary school students).

Soon after Widmar, financial aid cases were reconceptualized as involving equal access for religion to government funding. Mueller v. Allen (1983) upheld a state tax deduction for tuition and other educational expenses of families, even though the vast majority of deductions went for religious schooling and the tuition benefit was not limited to the secular aspects of such schooling. Under a strict application of the Lemon approach, these features would have doomed the program. But the Court, while still using Lemon’s phrases, emphasized different features: The deduction’s terms were neutral and thus created no incentive for parents to choose religious schools over public or secular private schools. The emphasis on neutral terms and individual choice continued in Witters v. Dept. of Services (1986), which allowed a blind student to use generally available state rehabilitation funds to study for the ministry at an evangelical Bible college. Under the program, the aid flowed to religious teaching ‘‘only as a result of the genuinely independent and private choices of aid recipients,’’ much like a state employee ‘‘donat[ing] all or part of [his] paycheck to a religious institution.’’

The ‘‘private choice’’ approach culminated in Zelman v. Simmons-Harris (2002), which upheld the inclusion of religious schools in a program of vouchers for students in Cleveland’s failing public schools. Like the tax deductions and rehabilitation funds, the voucher’s terms were neutral and left the family the choice of the school at which to use it. The Court also found that ‘‘genuine secular alternatives’’ existed—charter and magnet public schools, secular private schools— so that children would not be pushed into religious education to escape poor public schools. Zelman set forth a permissive blueprint for aid, and by treating public schools as simply one ‘‘secular alternative’’ to religious education, it rejected Lemon’s premise that they were a neutral baseline.

But Zelman has not entirely displaced the separationist, Lemon-oriented approach. In Locke v. Davey (2004), the Court held that although a state program of college scholarships could include students studying theology in preparation for the ministry, the Free Exercise Clause did not require their inclusion. The majority gave the state discretion because it found ‘‘strong Establishment Clause interests’’ in denying public funding for clergy training, interests grounded in history as far back as Madison’s ‘‘Memorial and Remonstrance’’ and the rejection of taxes to support teachers of Christianity. Although Davey’s application beyond clergy training is uncertain, the decision limits the argument that religion must be allowed equal participation in private-choice programs.

Moreover, when an aid program falls outside the rubric of ‘‘private choice,’’ the Court has continued a Lemon-like approach. In Mitchell v. Helms (2000), a splintered Court upheld a federal law providing computers and other instructional equipment and materials to public and private, including religious, schools. Justices O’Connor and Breyer, the decisive votes, upheld the program only because it included safeguards to prevent ‘‘diversion’’ of materials to religious uses. They argued that direct aid to religious schools fell closer to the core prohibitions of the Establishment Clause than did aid to individuals who could use it at religious schools. This distinction—if it survives changes in the Court’s makeup—significantly affects funding not only of education but of social services, where aid often flows through direct government contracts with providers.

In contrast to its increasing approval of aid to private religious activities, the Rehnquist Court reaffirmed the decisions prohibiting government’s own religious activity. The Court refused to limit the Establishment Clause to cases of coercion against individuals, instead gravitating toward a test championed by Justice O’Connor: government should not act with the purpose or effect of endorsing (or disapproving of) religion or any religious view. Endorsement of a religious view, she argued, ‘‘sends a message to nonadherents that they are outsiders, not full members of the political community,’’ and a message to adherents that they are favored ‘‘insiders’’ (Lynch v. Donnelly [1984] [concurring]). The no-endorsement test gathered a majority in County of Allegheny v. ACLU (1989), which invalidated a county courthouse’s stand-alone display of a nativity scene (distinguishing the Lynch case, which had upheld a municipal nativity display accompanied by numerous secular Christmas symbols).

The endorsement test paralleled Lemon in looking beyond coercion to more subtle effects such as government messages concerning the status of citizens. Accordingly, critics found the approach too subjective and complained that government disagrees on issues with dissenters every day without thereby consigning them to general second-class status. The ‘‘outsider’’ argument thus requires another step showing why disagreements on religious issues are of special concern. That step might be that religious debates are irrelevant to government policy, so the public divisions they create are unnecessary; that religious divisions are especially heated and irreconcilable; or that religious identity is more central to persons than are particular political views, and thus the alienation produced by government favoritism is deeper. But all of these propositions are debatable, some of them extremely so.

The Court recently reaffirmed its restrictions on even noncoercive government symbolism, by invalidating a display of the Ten Commandments in a county courthouse (McCreary County). Although the display also included historic secular documents of law and government, the majority interpreted it in the light of immediate previous displays that had been solely religious in nature, as well as numerous religious statements from county officials. The decision thus reaffirmed Lemon’s ‘‘secular purpose’’ prong, indeed arguably stiffening it by requiring that the secular motive not just exist but be ‘‘primary.’’

Even approaches based on separationism and neutrality, however, can be expected to permit some religious statements and displays by government— although it is not always clear how to explain this logically. Religious elements may appear in a broader display that overall does not endorse religion: a cre`che with secular Christmas symbols as was upheld in Lynch v. Donnelly, or a Ten Commandments plaque as one of an assortment of monuments on public grounds representing various historic events and ideas as was upheld in Van Orden v. Perry (2005). Moreover, individual justices who vote to invalidate some government symbols or statements have justified others—‘‘In God we trust’’ on coins, ‘‘under God’’ in the Pledge of Allegiance—as examples of ‘‘Ceremonial Deism,’’ a category of short, highly generalized religious references, often with a longstanding history, that impose on no one’s liberty and may indeed have developed largely ceremonial rather than specifically religious meanings. See, for instance, Newdow (O’Connor, J., concurring).

To approve religious references only by stripping them of their religious meaning surely calls to mind the founding-era concern that official establishments undermine the purity and vigor of religion. It is also doubtful that some of these references, such as ‘‘under God’’ in the Pledge, have lost religious import. Justice Breyer was likely more honest when he explained his decisive vote to uphold the Decalogue display in Van Orden: striking down all official religious symbols, especially longstanding ones, would be more divisive than are the symbols themselves. But Breyer provided no clear line for whether a display was permitted or forbidden; the jurisprudence in this area remains uncertain.

The recent cases reveal a shift in answers to the question posed at the outset of this entry: beyond protecting individuals from being coerced to engage in religion, what limits does the Establishment Clause impose on government? In the nineteenth and early twentieth centuries, the majority generally understood disestablishment to prevent tax-financed support for religious schools, but to permit governmental expression endorsing a generalized Christianity. At the dawn of the twenty-first century, the Supreme Court case law has moved in the opposite direction: greater latitude for religious schools and social services to receive aid on equal terms with nonreligious institutions, but significant limits on official endorsement even of familiar symbols like the Ten Commandments. The shift has several causes. The welfare state’s widespread spending on social programs makes the denial of aid to religious providers appear less as nonentanglement and more as discrimination. And while increasing religious pluralism in America makes any explicit religious statement by government look more and more partial, financial aid can be defined in secular terms and in theory can be available to a wide range of religious groups. The pervasive modern state has reduced the viability of strict separationism as an ideal, and the competing ideals of equality, neutrality, and individual choice are more hospitable to aid programs than to explicit government endorsement of any religious view.

The Establishment Clause and Religious Institutional Autonomy

Finally, one potential implication of the Establishment Clause remains relatively unexplored by the Court: the use of the clause as a shield to protect religious institutions from government interference, at least in core or inherently religious matters such as doctrine, liturgy, and the selection of clergy. Establishments typically involved government oversight of such religious matters—albeit in the name of preserving the favored faith—and a prime purpose of disestablishment was to separate the structures of church and state to ensure the autonomy of religious life. The Establishment Clause may play an increased role in protecting religious institutional autonomy as Free Exercise Clause protections have declined with the Court’s holding (Employment Division v. Smith [1990]) that religiously motivated conduct generally has no constitutional claim to exemption from ‘‘neutral law[s] of general applicability.’’

A few Court decisions have sounded the theme of Establishment Clause immunity from regulation. Most explicitly, the Court invalidated financial aid to religious schools on the ground that the regulation accompanying aid produces ‘‘excessive entanglement’’ (Lemon) and ‘‘the spectre of government secularization of a creed’’ (Aguilar v. Felton [1985]). Ironically, such rulings used the school’s autonomy as a ground to strike down aid that the school wished to receive. ‘‘Pure’’ protections of religious institutional autonomy, by contrast, have invoked the Religion Clauses in general, a rationale consistent with, but not explicitly mentioning, the Establishment Clause. Under a century-long series of decisions involving property disputes between church factions, civil courts deciding such cases must sometimes defer to the resolution of the matter by the highest authority within the church (Watson v. Jones [1872], Serbian E. Orthodox Diocese v. Milivojevich [1976]. Although states may often apply ‘‘neutral principles’’ of property or trust law to such disputes, they must defer to the internal church authority whenever the controversy requires interpretation of religious concepts or involves ‘‘religious authority or dogma’’ (Jones v. Wolf [1979], Milivojevich, Smith). Applying these principles, lower courts have barred suits by clergy against churches for employment discrimination, suits against clergy and churches for ‘‘malpractice’’ in religious counseling, and—in some cases—suits against churches for negligent hiring or supervision of clergy who commit torts. Whether these decisions extend to nonclergy positions in religious institutions is uncertain at best, although the Supreme Court in NLRB v. Catholic Bishop (1979) held that application of collective bargaining laws to teachers in parochial schools would raise serious constitutional questions concerning church–state entanglement. Institutional autonomy from regulation remains among the least explored implications of the Establishment Clause.

THOMAS C. BERG

References and Further Reading

• Conkle, Daniel O., Toward a General Theory of the Establishment Clause, Northwestern University Law Review 82 (1988): 1113–1194.
• Esbeck, Carl H., Dissent and Disestablishment: The Church– State Settlement in the Early Republic, Brigham Young University Law Review 2004 (2004): 4:1385–592.
• Laycock, Douglas, ‘Nonpreferential’ Aid to Religion: A False Claim About Original Intent, William and Mary Law Review 27 (1986): 875–921.
• Lupu, Ira C., Government Messages and Government Money: Santa Fe, Mitchell v. Helms, and the Arc of the Establishment Clause, William and Mary Law Review 42 (2001): 771–822.
• McConnell, Michael W., Coercion: The Lost Element of Establishment, William and Mary Law Review 27 (1986): 933–941.
• McConnell, Michael W., John H. Garvey, and Thomas C. Berg. Religion and the Constitution. Boulder, CO: Aspen Publishers, 2002.
• Smith, Steven D., The Rise and Fall of Religious Freedom in Constitutional Discourse, University of Pennsylvania Law Review 140 (1991): 149–239.
• Sullivan, Kathleen M., Religion and Liberal Democracy, University of Chicago Law Review 59 (1992): 195–223.

Cases and Statutes Cited

• Abington School Dist. v. Schempp, 374 U.S. 203 (1963)
• Aguilar v. Felton, 473 U.S. 402 (1985)
• Board of Education v. Mergens, 496 U.S. 226 (1990)
• County of Allegheny v. American Civil Liberties Union, 492 U.S. 573 (1989)
• Elk Grove Unified School District v. Newdow, 542 U.S. 1 (2004)
• Employment Division v. Smith, 494 U.S. 872 (1990)
• Engel v. Vitale, 370 U.S. 421 (1962)
• Epperson v. Arkansas, 393 U.S. 97 (1968)
• Everson v. Board of Education, 330 U.S. 1 (1947)
• Good News Club v. Milford Central School, 533 U.S. 98 (2001)
• Harris v. McRae, 448 U.S. 297 (1980)
• Jones v. Wolf, 443 U.S. 595 (1979)
• Lee v. Weisman, 505 U.S. 577 (1992)
• Lemon v. Kurtzman, 403 U.S. 602 (1971)
• Locke v. Davey, 540 U.S. 712 (2004)
• Lynch v. Donnelly, 465 U.S. 668 (1984)
• Marsh v. Chambers, 463 U.S. 783 (1983)
• McCreary County v. American Civil Liberties Union, 125 S. Ct. 2722 (2005)
• Mitchell v. Helms, 530 U.S. 793 (2000)
• Mueller v. Allen, 463 U.S. 388 (1983)
• NLRB v. Catholic Bishop, 440 U.S. 490 (1979)
• Serbian E. Orthodox Diocese v. Milivojevich, 426 U.S. 696 (1976)
• Stone v. Graham, 449 U.S. 39 (1980)
• Van Orden v. Perry, 125 S.Ct. 2854 (2005)
• Wallace v. Jaffree, 472 U.S. 38 (1985)
• Walz v. Tax Commission, 397 U.S. 664 (1970)
• Watson v. Jones, 13 Wall. 679, 722–724 (1872)
• Webster v. Reproductive Health Services, 492 U.S. 490 (1989)
• Widmar v. Vincent, 454 U.S. 263 (1981)
• Witters v. Dept. of Services, 474 U.S. 481 (1986)
• Zelman v. Simmons-Harris, 536 U.S. 639 (2002)