We live in a land of religious freedom, but our legal system in reality highly regulates religion. The legal foundation of the United States, the Constitution, contains the First Amendment’s “religious clauses,” which affirm that “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof… .” These two clauses delineate a spectrum of possible relations among the government, religious institutions, and individual religious believers (and, by implication, nonbelievers). In general, the Free Exercise Clause may require the government to exempt religions and religious believers from general laws, while the Establishment Clause may prevent the government from providing funding or special preferences to religious entities. These two poles delimit a gray area where government may acknowledge religion without being required to do so. Determining what governments properly do within these two poles remains quite challenging, controversial and even somewhat ephemeral.
The Constitution’s statements on religion may appear self-explanatory, but how we are to remain faithful to them has seemed far from obvious, required judicial interpretation, and evolved to face new challenges. Indeed, the wide reach of the First Amendment’s religion clauses emerged only quite recently, and it was only after the U.S. Supreme Court established its own expansive reach that the clauses were essentially revived from nearly one hundred fifty years of dormancy to contribute to much heated judicial, scholarly, and public debate about religion’s role in public life. As suggested by the phrase “Congress shall make no law …” the clauses actually regulate only the federal government, and they were applied only to acts of Congress until the 1940s. Through case law and judicial interpretation of historical accounts, the Supreme Court made the Free Exercise Clause applicable to the states by “incorporation,” an interpretive process by which the Court extended to the liberties protected from federal encroachment by the First Amendment similar protections from state encroachment through the Fourteenth Amendment’s Due Process Clause. The case in question, Cantwell v. Connecticut (1940), involved a Connecticut law that prohibited door-to-door solicitation, a law declared unconstitutional because it violated the Free Exercise rights of Jehovah’s Witnesses. It was seven years later that the Court similarly extended the reach of the Establishment Clause. In that case, Everson v. Board of Education (1947), the Court, while supporting a very high standard of church-state separation, held constitutional a New Jersey statute that reimbursed parents for the cost of busing their children to parochial school. From a constitutional and historical perspective, then, the mandate for states to follow the Constitution’s provisions regulating religion appears quite momentous, even amazing, when we recognize that this interpretation of the religion clauses, catalyzed by the two cases, emerged largely as a phenomenon of the second half of the twentieth century, a period that itself witnessed, largely due to legal changes, dramatic transformations in the role of religion in public life.
The cases that extended the Constitution’s reach to state and local governments have significance beyond the actual law they settled. The cases created an important breakthrough for understanding the legal regulation of religion. Given that most religion cases arise as a result of state rather than federal judicial action or legislation, the cases instrumentally fostered litigation to which the Supreme Court could respond in efforts to guide and help address religious issues confronted by all levels of government. Since incorporation, the federal courts have largely determined national, state, and local policy on the proper role of religion in public life. Such enormous federal power seems especially dramatic in view of the cogent arguments put forth asserting that the incorporation cases actually repealed the Constitution’s provision that Congress may make no laws regulating religion—that the clauses regulating religion amounted to a decision by the federal government not to address questions concerning the proper relationship between religion and the government (see Smith, 1995). This approach would mean that there would be no national law, and thus no constitutional law, theory, or principle prescribing the proper relationship between government and religion and that the regulation of religion would rest solely with the states. Given the latter possibility, it is not surprising to find that considerable judicial and political work still continues the task of interpreting the Constitution’s clauses as applicable to states and local governments (see Glendon & Yanes, 1991).
Litigation constantly challenges where and how the law can draw lines to separate the government, religion, and individuals, and the resulting jurisprudence draws and implements those lines with important opportunities to respond to changing conceptions of religion’s role in public and private life. That legal institutions can respond to religion’s role in society is evidenced by the Court’s significant doctrinal shifts in its regulation of religion and by the federal and state legislatures’ adaptation to those shifts. Given these shifts, analyzing how the law determines the role of religion in adolescents’ lives necessarily must start with an understanding of how and why the legal system regulates religious institutions, practices, and even beliefs.
The Supreme Court’s Free Exercise jurisprudence defines the limits on how much government regulation may intrude into citizens’ religious practices. The Court must define that limit, given that the Court itself has interpreted the Constitution’s Free Exercise Clause as providing religious believers with the right to exemptions from laws that infringe on their religious practices. The right to exemptions, by implication, raises the question of the extent to which the Constitution requires that citizens be protected from laws that regulate religious beliefs (important attempts have been made to distinguish beliefs from practices and the legitimate right of the state to regulate either or neither). To understand how the Court balances these concerns, one must understand how it has developed relevant jurisprudential standards and how it envisions further developments in the application of those standards.
The Court’s first major free exercise case set the standard for nearly a century. That case, Reynolds v. United States (1878), involved the constitutionality of a federal law that prohibited bigamy by male Mormon settlers, who claimed that they had a religious duty to marry more than one woman. While the Court acknowledged that the Free Exercise Clause prevented the government from prosecuting citizens for their religious beliefs, it distinguished that from the enforcement of laws governing conduct and held that Mormons could not claim a constitutional exemption. The Court interpreted the Free Exercise Clause as, in actuality, a freedom of belief clause that merely protects religious opinion: “Congress was deprived of all legislative power over mere opinion, but was left free to reach actions which were in violation of social duties or subversive of good order” (Reynolds v. United States, 1878, p. 164). As a result, the Court found that the state had a legitimate right to apply general laws that could infringe on religious adherents’ sincere practices. The Court’s rigid deference to congressional will resulted, in this case, in particularly onerous intrusions, given that Mormon doctrine required male members of the Church to practice polygamy when circumstances permitted and that the penalty for failure to do so would be “damnation in the life to come” (Reynolds v. United States, 1878, p. 161). From the outset, then, the legal system arrogated to itself considerable power—it would judge how people practice their religions and could even require individuals to act against their faith.
Although Reynolds continues to be the subject of considerable commentary and controversy, for our purposes the case provides two fundamental lessons. From its first analysis of laws that impact religion, the Court assumed that the legal system possessed the authority to regulate religious conduct in the name of advancing social welfare. That authority meant that the Court had to defer to the state’s regulatory power. Equally important, in the Court’s first analysis of these laws, it took the opportunity to limit the freedom to follow the dictates of a religious practice. In fact, even though the Court had asserted that the Free Exercise Clause may require a religious exemption from a general law that regulates conduct when legislation conflicts with religious precepts, the Court never granted such an exemption until it shifted its standard of analysis in the 1960s. Until that time, the Court’s reasoning in Reynolds helped decide a long line of cases that allowed the government to prohibit religious practices even when individuals had acted under the loftiest spiritual motives as required by their religious beliefs.
It was during the socially and politically tumultuous years of the 1960s that the Court, rather than simply relying on the government’s interests, developed important standards to protect citizens making free exercise claims. Two cases involving observance of the Saturday Sabbath illustrate the major shift. In Braunfeld v. Brown (1961), the Court upheld Sunday closing laws that burdened a Jewish businessman’s ability to compete, since he observed a Saturday Sabbath. The Court found no unconstitutional burden on the plaintiff’s religious practice, since nothing in the plaintiff’s faith required him to run a business and the legislature could not operate effectively if every law that increased costs to some religious adherents were unconstitutional.
The Court shifted its analysis in a case that followed two years later. That case, Sherbert v. Verner (1963), involved a Seventh-Day Adventist who lost her employment because she refused to work on her Sabbath day, Saturday. Her refusal to violate her religious beliefs meant that she did not qualify for state unemployment benefits after her discharge because she had “refused suitable work” (Sherbert v. Verner, 1963, p. 401). The Supreme Court departed from rulings in prior cases by applying strict scrutiny to the state law, a level of scrutiny that seeks to determine whether the challenged regulation burdened religious beliefs and whether the state had a compelling interest in enforcing a narrowly tailored regulation. As a result, the Court held that the state must justify the rule by showing not only a neutral governmental intent but also a compelling government interest. Under this rule, the Court discerned that the unemployment compensation law unconstitutionally burdened the plaintiff’s religious practice and found no sufficiently compelling government objective that would justify that burden. In addition, the Court found that permitting the plaintiff to receive an exemption from unemployment compensation laws did not interfere with the operation of the state unemployment compensation system, because the state always used a case-by-case analysis to evaluate individuals’ qualifications for benefits. Thus, the Court reasoned that the right was so compelling that it justified burdening the state, especially since the state could adapt appropriately and address the religious objector’s claims without excessively burdening itself.
A decade later, the Court used this “strict scrutiny” test to establish the high-water mark for religious protection under the Free Exercise Clause. In Wisconsin v. Yoder (1972), the Court sided with Amish parents who refused to comply with the state’s compulsory school attendance statute after their children had completed the eighth grade. In an effectively unanimous opinion, the Court recognized that enforcing the statute would “gravely endanger if not destroy the free exercise of respondents’ [the Amish parents’] religious beliefs” (Wisconsin v. Yoder, 1972, p. 219). Moreover, the Court fully rejected the outmoded distinction between belief and action, stating that, at least in the case of the Amish, “belief and action cannot be neatly confined in logic-tight compartments” (Wisconsin v. Yoder, 1972, p. 220). Most important, the opinion demonstrated the Sherbert approach’s potentially strong protective stance toward religious liberty. When subjected to “strict scrutiny,” Wisconsin’s interest in preparing students for participation in modern life could not stand against the success and agrarian self-sufficiency of the Amish people. Religious freedom would be secured against undeniably neutral legislation that served highly laudable goals. Quoting Sherbert, the Court in Yoder indicated that, to be “compelling,” the interest a state regulation must address would have to involve “some substantial threat to public safety, peace, or order” (Wisconsin v. Yoder, 1972, p. 230). Given that the Amish traditions posed no such compelling threat to the general public, the state had to accept and provide an exemption.
While the strict-scrutiny standard introduced in Sherbert and affirmed in Yoder promised extensive protections for religious believers burdened by state and federal laws, the cases’ significance was not as profound as it might appear. Although the robust “strict-scrutiny” test developed in Sherbert and Yoder greatly influenced lower-court decisions, the Supreme Court itself appeared markedly less hospitable to subsequent Free Exercise claims. Virtually all of its later decisions, while often invoking strict scrutiny, ruled in favor of government regulations and rejected exemptions to accommodate religious claims. In fact, religious adherents’ claims for exemptions from generally applicable laws remain for the most part unsuccessful. Thus, in Lyng v. Northwest Indian Cemetery Protective Association (1988), the Court permitted the U.S. Forest Service to build a road through government land that was located on areas sacred to Native Americans; in Goldman v. Weinberger (1986), the Court upheld a military dress code that barred the plaintiff, an ordained rabbi, from wearing a yarmulke while in uniform; in Bob Jones University v. United States (1983), the Court denied an exemption from antidiscrimination laws to a religious university that claimed that its religious beliefs required racial discrimination; and in United States v. Lee (1982), the Court required an Amish employer to pay social security taxes, despite his religious objections. The newly applied strict-scrutiny standard, then, hardly resulted in a wholesale protection of religious rights and claims.
The Court’s move away from protecting free exercise claims, despite its rigorous scrutiny of alleged infringements, became even more obvious in the context of unemployment, the very context in which Sherbert introduced the high-scrutiny standards. In this context, the Court’s approaches again reveal dramatic shifts in thinking. At the end of the 1980s, the Court closely followed Sherbert in resolving disputes involving eligibility for unemployment compensation. Most notably, in Frazee v. Illinois Department of Employment Security (1989), the Court upheld the right to compensation of a worker who objected to working on Sunday, even though he did not belong to a particular Christian sect. However, a year later, even that context no longer benefited from the strict-scrutiny approach. The Court formally abandoned the strict-scrutiny standard in Employment Division, Oregon Department of Human Resources v. Smith (1990) and upheld the state’s decision to withhold unemployment benefits from Native American plaintiffs terminated from their positions as drug rehabilitation counselors because they had made sacramental use of peyote in one of their religious ceremonies. While the Court acknowledged that, like the law in Sherbert, the Oregon law forced the plaintiffs to choose between upholding their religious beliefs and accepting state benefits, it concluded that “to make an individual’s obligation to obey such a law contingent upon the law’s coincidence with his religious beliefs … contradicts both constitutional tradition and common sense” (Employment Division, Oregon Department of Human Resources v. Smith, 1990, pp. 885–890). Smith explicitly limited Sherbert to the employment compensation context where “the State has in place a system of individualized exemptions” (Employment Division, Oregon Department of Human Resources v. Smith, 1990, p. 884). Equally important, the Court did not even require a compelling government interest before denying the free exercise claim. The Court explained that the only cases that justified free exercise exemptions from generally applicable laws were those where there was more than one constitutional right at issue (Employment Division, Oregon Department of Human Resources v. Smith, 1990, p. 881), such as a combination of free exercise claims and claims involving the right to free speech or freedom of the press. The Court noted, for example, that it would uphold exemptions for religious exercise combined with the right of parents to direct the upbringing of their children, as in Wisconsin v. Yoder (1972). Essentially, the Court held that, as long as religious exercise is free from deliberate political persecution, the Free Exercise Clause does not require the Court to provide heightened “strict scrutiny” to laws that burden the exercise of religion.
It was in the wake of the Smith decision that Congress swiftly acted to pass the Religious Freedom Restoration Act (RFRA) (1994). RFRA explicitly restored the strict-scrutiny test of Sherbert, which required narrowly tailored laws to achieve the government’s compelling interest when the law “substantially” burdens a religious belief or practice. Congress argued that it had the power to pass RFRA because, under § 5 of the Fourteenth Amendment, it was required to protect citizens’ liberty interests in practicing religion unimpaired by government interference. However, in City of Boerne v. Flores (1997), the Supreme Court found RFRA unconstitutional on the grounds that it violated the separation of powers between the legislative and the judicial branches of government. In that case, zoning authorities denied Flores, the archbishop of San Antonio, a building permit to expand a church in Boerne, Texas. The archbishop challenged the permit denial, claiming that the authorities could not restrict the church building plans, even though the church was located in a historic preservation district. The Court, applying the Smith standards, decided that the archbishop could not obtain an exemption from this neutral law of general applicability by relying on the invalid test prescribed by RFRA. According to the Court, RFRA impermissibly violated the Establishment Clause, since it gave special benefits to religious believers that nonreligious citizens and organizations could not share. The Court explained that applying the Sherbert and the RFRA compelling-interest test would create “an anomaly in the law, a constitutional right to ignore neutral laws of general applicability” (City of Boerne v. Flores, 1997, p. 513).
Although the Court took a seemingly strong stance against efforts to elevate religious rights, religious rights still enjoy considerable protection. The likelihood of protection is particularly great when the laws that impact religion are neither neutral nor general in nature, as when a law specifically targets certain religious practices. The Court’s approach in these contexts is highlighted well by Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah (1993). In that case, the Court examined several city ordinances that effectively prohibited the ritualistic sacrificial killing of animals, a practice of the Santeria religion. The Court’s majority first noted that, “[a]lthough the practice of animal sacrifice may seem abhorrent to some, ‘religious beliefs need not be acceptable, logical, consistent, or comprehensible to others in order to merit First Amendment protection’” (Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 1993, p. 531). It continued by agreeing with the Church’s assertion that animal sacrifice is an integral part of its religion and that it cannot be deemed bizarre or incredible. Given the centrality of the prohibited sacrifices to the Church’s religious beliefs, the Court proceeded to address the constitutional claim. In its analysis, the Court first returned to Smith for the proposition that “a law that is neutral and of general applicability need not be justified by a compelling governmental interest even if the law has the incidental effect of burdening a particular religious practice” (Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 1993, p. 531). The Court noted, however, the existence of an important interrelationship between “neutrality” and “general applicability” and added that failure to satisfy one suggests the failure to satisfy the other. As a result, it concluded that “[a] law failing to satisfy these requirements must be justified by a compelling government interest and must be narrowly tailored to advance that interest” (Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 1993, pp. 531–532). Because the ordinances aimed to suppress the central element of the Santeria worship service, the Court found that the ordinances were neither neutral nor of general applicability. In its defense, the city advanced two interests it had sought to target with the ordinances: protection of the public health and prevention of cruelty to animals. The Court found the ordinances to be underinclusive on both counts because they failed to prohibit nonreligious conduct that was equally threatening to the public health or equally likely to cause cruelty to animals. Having determined that the laws were neither neutral nor of general application, the laws had to undergo the most rigorous of scrutiny under the compelling interest standard. The Court easily found the ordinances to be both overbroad and underinclusive; it concluded that “[t]he absence of narrow tailoring suffices to establish the invalidity of the ordinances” (Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 1993, p. 546). The Court then noted that the city had failed to demonstrate that its governmental interests were compelling, finding that “[w]here government restricts only conduct protected by the First Amendment and fails to enact feasible measures to restrict other conduct producing substantial harm or alleged harm of the same sort, the interest given in justification of the restriction is not compelling” (Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 1993, pp. 546–547). By addressing only Santeria sacrifices and allowing similar and similarly motivated killings and risks to public health, the city had inappropriately infringed on Santeria religious believers’ free exercise rights. Hialeah, then, affirms the high protection religions receive when laws affecting them are neither neutral nor of general applicability; failing to satisfy these requirements, the law must be justified by a compelling governmental interest and must be narrowly tailored to advance that interest. Religions, then, can receive the Court’s highest level of protection.
The Court’s enormous power has meant that it routinely struggles to envision appropriate standards upon which to judge and guide governmental efforts that could limit individuals’ practice of their religion. In fact, the modern Court has modified the separationist and protective stance it took in the 1960s and 1970s, and its most recent doctrinal move, coupled with highly split decisions (a high number of dissenting and concurring opinions urging different approaches to resolving issues but nevertheless achieving similar outcomes), indicates that the future path of the law may be unclear. Nevertheless, recent cases reveal that the Court does not take a strong stand on free exercise rights when they are pitted against governmental interference. The Court now articulates a fairly low governmental standard for infringement of free exercise rights. Laws that incidentally burden religious practice are permissible if they are neutral laws of general applicability. Despite this apparently permissive approach to regulating religion, the Court still imposes more demanding obligations on the state to justify its laws and to narrowly tailor its intrusion on religious practices if the laws also impact another important right or if they are not meant to be neutral in their application. In these circumstances, the Court will look favorably upon free exercise claims when governments can accommodate those claims without excessively burdening themselves. Despite these considerations, the right to practice one’s faith remains not as strong as many would believe.
While the Free Exercise Clause determines the minimum accommodation that the government must provide for religious believers, the Establishment Clause sets a maximum amount of assistance that the government may offer. The express language of the Establishment Clause presumably prohibits a governmental body from advocating, proselytizing, or taxing citizens to promote a religion or an article of religious faith. This extreme view of the establishment prohibition becomes highly problematic in that practically any proposition, even the most uncontroversial, can be alleged to implicate religion if a church or sect has adopted it (or its opposite) as an article of faith. To prevent unintended and paralyzing results, the Court has subjected the provision to numerous refinements and explications that mirror the dramatic shifts in free exercise jurisprudence.
The Supreme Court took nearly 150 years before it availed itself of the opportunity to consider directly the nature and reach of the Establishment Clause. As we saw earlier, that case, Everson v. Board of Education of Ewing (1947), involved a law that reimbursed parents for their children’s bus transportation to private, religious, and public schools. In its analysis, the Court invoked Thomas Jefferson’s “wall of separation” between church and state as it expressed concern about government assistance to children who attended religious schools. Even as it emphasized that the wall “must be kept high and impregnable” (Everson v. Board of Education of Ewing, 1947, p. 18), the Court decided to allow the extension of these state subsidies to religious schools. The Court characterized the subsidies as standard state benefits that only indirectly touched religious schools. The cautious Court warned that citizens should not be denied these regular benefits because of an “overzealous” enforcement of the Establishment Clause (Everson v. Board of Education of Ewing, 1947, p. 18). Rather, the Court’s ruling focused on the fact that the policy assisted in the transportation of children who attended any accredited school and did not directly fund or support religious schools. On the basis of these facts, the Court held that the policy was neutral toward religion and did not violate the Establishment Clause. As a result, the most far-reaching impact of the Everson decision was the Court’s endorsement of government neutrality toward religion.
The Court’s decision to endorse government neutrality was significant, but it did not signal the crumbling of Jefferson’s wall. Most notably, the Court re-emphasized the strength and great height of the wall one year after Everson, in Illinois ex rel. McCollum v. Board of Education (1948). In this case, the Court had to evaluate a public school program that made use of privately paid religious instructors who conducted religious instruction in the Protestant, Catholic, and Jewish religions. The Court’s chief concerns were that the school board was supporting religion by letting facilities that were built with tax funds be used for religious instruction and that, although only those students whose parents gave permission attended the religious classes, the state used its “compulsory public school machinery” to furnish the pool from which these students came (Illinois ex rel. McCollum v. Board of Education, 1948, p. 212). In considering whether the Establishment Clause permitted such instruction, the Court rested its analysis on the notion that the clause was founded on “the great American principle of eternal separation” (Illinois ex rel. McCollum v. Board of Education, 1948, p. 231). On the basis of its concerns and its view of the Establishment Clause, the Court ruled the policy unconstitutional and concluded that allowing clergy to use public school facilities for religious instruction during school hours moved too close to government support of religion.
Although the Court envisioned the separation between church and state as one of the vital safeguards of our constitutional system, the Court soon softened its separationist stance and moved toward a jurisprudence of accommodation. In yet another case involving the place of religion in public schools, Zorach v. Clauson (1952), the Court approved a New York program that allowed public school students whose parents had requested their early release to attend religious classes. Although the Court referenced Everson in announcing its departure from a strict separationist view, it went much further and accepted a state policy that revealed an accommodationist stance toward religion. In lowering the wall, the Court first noted that only the presence of a genuine tension between religious expression and government indoctrination requires that the separation “be complete and unequivocal” (Zorach v. Clauson, 1952, p. 312). In apparent response to the public criticism ignited by McCollum, the Court opined that complete separation of church and state would lead to alienation between the two systems, in turn breeding hostility (Zorach v. Clauson, 1952, p. 312). The Court found that the policy of accommodating student dismissals for religious education did not engender genuine tension between the free expression of religion and the bar against the government’s establishment of religion because the classes were held outside the public school building and without administrative involvement. The Court was well aware that the release time necessarily involved state action that included the force of the state’s laws on compulsory attendance and truancy. But the Court deemed the policy a reasonable accommodation that sought to lessen the level of government involvement. By playing down the potential use of public funds to enforce the truancy laws for those who might have violated released time attendance requirements, the Court’s analysis allowed for a significant relaxation of a hard-line separationist orientation. Recognizing the need to allow for some state involvement opened the door to the less extreme accommodationist approach to Establishment Clause issues.
Early Establishment Clause jurisprudence, then, presented two approaches, one compelling strict separation between religion and the state and the other supporting accommodation of religious expression and practice. The Court left unclear, however, where permissible accommodation might end and separation might begin. The balance of accommodation and separation fluctuated for nearly three decades, until 1971, when the Court articulated what was then thought to be a clearer standard of analysis in Lemon v. Kurtzman (1971). It was in Lemon that the Court synthesized multiple lines of Establishment Clause cases and established a complex standard that would become the source of the most significant jurisprudence involving the Establishment Clause.
In Lemon, the Court considered a program that allocated state funds to reimburse the expenses and supplement the salaries of private school teachers in Pennsylvania and in Rhode Island. The Court approached the issue by devising a three-part test in which it forbade any governmental action that (1) has no secular purpose, (2) has a primary effect of advancing religion, or (3) cultivates excessive entanglement between church and state. Though holding that the purpose of the statutes—to improve secular education—was sufficient to survive the first part of the test, the Court struck down both programs by applying the third part. The Court concluded that the programs created excessive entanglement because they presented the potential for administrative entanglement between state and religious institutions, which could in turn lead to political divisions along religious lines. This test, conventionally known as the “Lemon test,” subsequently was considered the standard for Establishment Clause analyses that seek to erect a high wall between church and state—but it too was scrutinized, refined, and eventually rejected by some members of the Supreme Court.
It was only two years after the pronouncement of the Lemon standard that the Court reached its high-water mark in defining how it would limit the government’s role in actively supporting religion. The Court applied its most stringent application of the Establishment Clause in Committee for Public Education and Religious Liberty v. Nyquist (1973). Nyquist involved a tuition reimbursement program in New York that authorized various direct payments to private schools and parents of private-school students, including payments to schools for the purpose of maintaining and repairing equipment and facilities to protect students’ health, welfare and safety; reimbursements to low-income parents to help them defray the cost of their children’s tuition at private schools; and tax relief to parents who did not qualify for the reimbursements. The Court first found that, because the payments were directly made to private schools (most of which were affiliated with the Roman Catholic Church) without restrictions on usage, the payments had the primary effect of advancing religion. The Court used similar reasoning to invalidate the reimbursement scheme, noting that New York could not guarantee that the funds would be used for secular and neutral purposes. A key element of the Court’s analysis was the justices’ assertion that, while the payment scheme was an important factor in determining the constitutionality of the scheme, it was not the only factor. Instead, the Court focused on the overall effect of the program, noting that the tuition grants were being offered as incentives to encourage parents to send their children to religious schools by facilitating their ability to afford such schools. By adopting that broader approach, the Court held that the Constitution had been violated, regardless of whether state aid was used for religious purposes or whether parents were free to spend the money as they wished. The Court also invalidated the third part of the program, noting that the program in effect created the same incentive as the reimbursements for sending a child to a sectarian school. Thus, Nyquist supported the view that the Court decisively had foreclosed any state-funded voucher program from surviving constitutional scrutiny if the program provided incentives for parents to enroll their children in schools that furthered a religious mission.
In cases that followed Nyquist, the Court showed that it had left itself room to accommodate state involvement in religious matters. Indeed, the Court’s strict enforcement of the Establishment Clause as pronounced in Nyquist diminished noticeably only a decade later. The Court embarked on this landmark jurisprudential shift in Mueller v. Allen (1983). The Minnesota law at issue in Mueller granted tax breaks to parents for the costs of tuition, textbooks, and transportation, regardless of whether their children were enrolled in public, private, or sectarian schools. The Court distinguished the Minnesota tax relief scheme from the one rejected in Nyquist by noting that the Minnesota scheme provided the tax benefit to all parents, regardless of whether their children were in private schools, thus making public funds available to schools on the basis of the private choices of parents. The Court readily admitted that the program provided the same financial benefit to sectarian schools as would direct aid, but the Court asserted that, because the parents were making private choices as to where to apply the aid, the state was not establishing or approving of religion. Thus, and although essentially allowing for the result it had previously forbade, the Court implicitly declined to overrule Nyquist as it concluded that the Minnesota scheme did not have the effect of advancing religion.
Mueller stands for much more than a shift in the Court’s position. It is the leading source for what the Court now considers when deciding whether a state action constitutes impermissible establishment. The majority in Mueller actually made two critical points. First, it called the Lemon test into doubt for the first time, stating that, while Establishment Clause questions have been guided by the three-part test first announced in Lemon, a close look at Lemon and its resulting jurisprudence reveals that it merely was offered as a “helpful signpost in dealing with Establishment Clause challenges” (Mueller v. Allen, 1983, p. 394). Second, the Court refused to engage in an empirical analysis of the challenged program, which would have shown that parents were using the vast majority of tax benefits to send their children to parochial, not secular, schools. Thus, the Court appeared to abandon the proscription against grants or tax benefits that created incentives for parents to send their children to sectarian schools by making it financially easier for parents to enroll their children in such schools. The Court achieved this result by ignoring evidence that demonstrated a strong incentive on the part of parents in Minnesota to do just that. In the wake of Mueller, the Court would begin to apply the Lemon test as a mere guiding tool.
The significant shift Mueller portended became reality in the Supreme Court’s next major Establishment Clause case, Witters v. Washington Department of Services for the Blind (1986). In Witters, the Court upheld a Washington state program that permitted the payment of an educational stipend for a blind student who was training for the clergy at a religious school. The Court found the program constitutional since the program paid vocational educational grants to enable blind students to attend public, private secular, and private sectarian schools. The Court embraced the incentive standard it essentially had abandoned ealier, determining that the program did not provide a financial incentive for students to choose sectarian educational institutions and that it did not bestow additional benefits on students who chose to attend sectarian institutions. Further, the Court stressed that the same ingredient present in Mueller was also present in Witters, that the aid was disbursed to sectarian institutions only as a result of “genuine and private choices of aid recipients” (Witters v. Washington Department of Services for the Blind, 1986, p. 487). Regarding the Lemon test’s concern with benefiting particular religions, the Court emphasized that the aid would be used not to promote religion but instead to aid disabled students. This iteration of the Lemon test was far less rigorous than that applied in Nyquist, or even in Lemon itself.
The Court further eroded the Lemon test in Zobrest v. Catalina Foothills School District (1993). In Zobrest, the justices upheld the use of state funds to supply sign-language interpreters to students in sectarian schools. The Court essentially ignored the Lemon test and relied more heavily on Witters and Mueller. The Court revived the neutrality standard, noting that because the program neutrally provided a benefit to “a broad class of citizens without reference to religion,” it was not readily subject to an Establishment Clause challenge (Zobrest v. Catalina Foothills School District, 1993, p. 9). The Court also reaffirmed the constitutionality of aid that states disburse to sectarian schools through “genuinely independent and private choices” (Zobrest v. Catalina Foothills School District, 1993, p. 9). The justices further held that, since the program included both public and private schools, including sectarian schools, it provided no financial incentive for parents to send their children to sectarian schools. Despite these findings, the Court still formally clung to the Lemon test, refusing to overrule it outright.
The erosion of the Lemon test continued with the most recent case involving state aid to private sectarian schools, Agostini v. Felton (1997). In Agostini, the Court upheld the provision at government expense, under the Individuals with Disabilities Education Act (IDEA) (1994, 2000), of a sign language interpreter for a deaf student at a parochial school. The Court approved of the program’s disbursement of funding to disadvantaged children via the compensation of public teachers who taught disadvantaged children in sectarian schools. In Agostini, the Court declared that an injunction against the program, originally upheld by the Supreme Court, had in fact been upheld by an analysis that relied on no longer valid law. The case that was no longer valid law, Aguilar v. Felton (1985), had in turn been based on a strict interpretation and use of the Lemon test. In Aguilar, the Court had applied Lemon and had found that the Title I program created excessive government entanglement because public employees would have to be closely monitored by the state in order to guarantee that they did not teach or approve religion. Overturning Aguilar is the closest the Court has come to directly overruling the Lemon test.
In Agostini, the Court focused its analysis, relying on Witters and Nyquist, on the permissibility of different types of direct governmental aid. The Court compared the challenged program to the use of a government employee’s paycheck, a situation in which the employee could sign over part or all of the paycheck to a religious cause or group but “only as a result of the genuinely independent and private choices of individuals” (Agostini v. Felton, 1997, p. 226). In overruling Aguilar, the Court asserted that Establishment Clause jurisprudence had undergone a change, citing intervening cases that had found no constitutional bar to government-provided aid to handicapped students who attended parochial schools (Witters v. Washington State Department of Services for the Blind, 1986; Zobrest v. Catalina Foothills School District, 1993). The Court found that the aid in question was not in the form of direct grants; that an empirical analysis of the numbers or percentages of religious school students receiving neutral aid was not relevant; and that the program did not create any financial incentive for parents to send their children to sectarian schools, because funds were disbursed to sectarian and secular school pupils on a neutral basis. The Court further noted that these cases had changed the law by creating a presumption of regularity in the actions of public employees operating in religious settings. This precedent permitted a presumption in Agostini that the teachers who administered Title I’s services in the parochial setting would not improperly advance religion even if the situation could not be monitored carefully. The entanglement-prong violation of Lemon in Aguilar thereby disappeared in a presumption.
Mitchell v. Helms (2000) involved a federal program that was a companion to the remedial instruction program reinstated in Agostini; the program challenged in Mitchell, long known as Chapter 2 of the Education Consolidation and Improvement Act of 1981, distributed federal money to state and local educational agencies, which used the money to purchase educational equipment for loan to private elementary and secondary schools. The statute limited the use of this “equipment,” which includes computers, software, library books, VCRs, films, tapes, and other audiovisual material, to programs that are “secular, neutral, and non-ideological” (Mitchell v. Helms, 2000, p. 2537). The program included restrictions on the diversion of the loaned equipment for religious use, although these restrictions were not carefully enforced by public authorities. The litigation in Mitchell involved a challenge to the program as applied in Jefferson Parish, Louisiana, in which Catholic schools were among the beneficiaries. In a highly contentious opinion, a plurality approached the issue of aid to sectarian schools by establishing two criteria: (1) the aid program must be neutral in aiding sectarian schools and other schools, both private and public, and (2) the government itself must not be engaged in religious indoctrination. By these criteria, the Chapter 2 program easily passed muster. The aid went to a broad array of schools. Moreover, because the aid formula turned on per capita allocation, the plurality found the program neutral in yet another way—the school received government financial aid in proportion to the number of students it attracted, and the program therefore created no incentives to choose sectarian education. Finally, the plurality of the Court rejected the argument that the possibility that the aid would be diverted and used for religious instruction created an Establishment Clause obstacle; any such diversion, though it might violate the governing statute or regulations, would not be attributable to the government and therefore would not violate the Establishment Clause. In its reasoning, the Court squarely rejected the notion that aid to “pervasively sectarian” schools should be treated under rules different from those that govern aid to other schools (Mitchell v. Helms, 2000, pp. 2550–2552). Labeling such a distinction offensive, troubling, and anti-Catholic in its “pedigree,” the plurality concluded that a doctrine “requir[ing] the exclusion of pervasively sectarian schools from otherwise permissible aid programs … [is] born of bigotry [and] should be buried now” (Mitchell v. Helms, 2000, p. 2552). The concurring opinion, expressing far more caution than the plurality, still pushed significantly beyond the prior law. Agostini, the 1997 decision that reinstated the remedial instruction program, had rested in large part on the finding that the instructors were public employees. Under Chapter 2, that feature was absent; sectarian school administrators and teachers were entrusted with the computers, software, books, and media materials under the general instruction that they be used only for secular and nonideological purposes. That entrustment would have been a fatal policy move in the constitutional jurisprudence of the mid-1970s. But the concurring justices read Agostini for the “proposition that … presumptions [that sectarian school teachers always and everywhere engage in] religious indoctrination are normally inappropriate when evaluating neutral school-aid programs under the Establishment Clause” (Mitchell v. Helms, 2000, p. 2567). Henceforth, according to their opinion, “[t]o establish a First Amendment violation, plaintiffs must prove that the aid in question actually is, or has been, used for religious purposes” (Mitchell v. Helms, 2000, p. 2567). Moreover, without joining in the plurality’s explicit condemnation of the concept of pervasive sectarianism as the product of anti-Catholic bigotry, the concurring justices joined in essentially undoing the presumption attached to that concept. On the record in Mitchell, they concluded that the safeguards against religious uses of publicly financed materials were sufficient and that evidence that the safeguards were being violated could be safely ignored given that it was so sparse and not fully exhibited (Mitchell v. Helms, 2000, pp. 2569–2571). Mitchell unquestionably broke new ground on crucial issues of money separationism. The case provided the Court with the opportunity to overrule prior law, repudiate crucial principles for blocking aid, and reverse the presumption that sectarian schools cannot be trusted to use public aid for nonsectarian purposes.
Mitchell may reveal the expected and necessary outcomes of evolving jurisprudence on entanglement issues, but Establishment Clause jurisprudence necessarily involves other important concerns as well. Indeed, to the extent that entanglement considerations become less important in determining whether a governmental action impermissibly establishes religion, questions about any challenged statute’s secular purpose increase in significance. The Supreme Court has revealed the potential significance of this shift in its response to relentless attempts to display religious texts and to prohibit the teaching of evolution in public schools. In Stone v. Graham (1980), for example, the Court rejected the avowedly secular purpose of Kentucky legislation that required the posting of the Ten Commandments in the back of every public school classroom. In an effort to ensure the legislation’s constitutionality by furthering a valid secular legislative purpose, the statute had mandated that the bottom of the display state, in small print, that the “secular application of the Ten Commandments is clearly seen in its adoption as the fundamental legal code of Western Civilization and the Common Law of the United States” (Stone v. Graham, 1980, p. 41). The Court, however, read the statute’s purpose as “plainly religious in nature” because the Ten Commandments are “undeniably a sacred text in the Jewish and Christian faiths” (Stone v. Graham, 1980, p. 41). Stone does not stand alone in its rejection of stealthy efforts to impose religious views and materials on others. The Court also imputed illegitimate purposes to a legislative act on the basis of its predictable results with regard to efforts to prohibit the teaching of evolution (Wallace v. Jaffree, 1985) and to governmental requirement that schools teach evolution only in tandem with creationism (Edwards v. Aguillard, 1987). In both cases, the Court found no likely legislative purpose other than to promote religion. As with some previous efforts, the legislative statements challenged in both Wallace and Edwards claimed that the statutes at issue promoted greater comprehensiveness or promoted academic freedom; the Court nevertheless rejected these secular assertions as mere pretense and easily found language in the legislative history and in the statutes themselves to support the conclusion that the laws sought to encourage religion in the schools. In Edwards, for example, Louisiana had enunciated a secular purpose, protection of academic freedom, but the Court reasoned that the statute would effect the opposite. By not encouraging the teaching of all scientific theories about the origins of humankind, the statute, which required a parity between evolution and creationism (i.e., teaching both or teaching neither) had the “distinctly different purpose of discrediting evolution by ‘counterbalancing its teaching at every turn with the teaching of creationism’” (Edwards v. Aguillard, 1987, p. 589). Clearly, the Court has not given up its role as arbiter of legislative intent.
A look at cases that challenge legislative purposes, however, makes it clear that the teaching of evolution in the public schools is the major exception to the rule that challenges to state regulations that affect religious establishment can easily win their claims. Even in Stone, for example, the Court’s analysis indicated that some religious displays might pass constitutional muster if the state could mitigate the religious nature of the display; for example, the Constitution permits the integration of the Bible into the school curriculum for the study of history, civilization, ethics, and comparative religion. In fact, the Court has mitigated its approach to government display of religious symbols, and later decisions suggest that the Court is not as likely to question the legitimacy of a stated government reason and to overturn a statute on these grounds alone (see Santa Fe Independent School District v. Doe, 2000). As a result, this aspect of constitutional protection typically has not been the most difficult hurdle in Establishment Clause cases. Courts typically allow a generous, deferential reading of the intent of legislative acts.
As issues involving the legislative purposes of statutes lose (or at least change in) significance, so do yet another set of concerns identified in the Lemon test: the extent to which the governmental action has a “principal or primary effect … that neither advances nor inhibits religion” (Lemon v. Kurtzman, 1971, pp. 612–613). Most notably, where the Court has inferred a legislative motive on the basis of predictable effects, the practical analysis and the eventual resolution of this concern closely mimic those involved in the second prong of the Lemon test. That these two considerations similarly can duplicate the same decisional factors has prompted some members of the Court to bundle them into an “endorsement” test. In fact, and as we soon will see, in recent cases the Court has by implication abandoned the Lemon test and has in its stead adopted the endorsement test to determine the validity of enactments under the Establishment Clause.
The endorsement test invalidates both an intended governmental endorsement of religion and an unintended but reasonably perceived endorsement. The Court uses a subjective standard to assess unconstitutional intentions to endorse. These determinations require the justices to discern and to judge whether the actual intent of the governmental body was to advance one religion over another or religion over nonreligion. An objective test determines the perceived endorsement; it seeks to adopt the point of view of a putative “reasonable observer,” an approach similar to the legal fiction of the “reasonable man” particularly common in the area of tort adjudication. The determination turns on the reasonable impression given to a hypothetical reasonable member of the community of possible observers that the Court deems to be the relevant perceivers. This test finds an Establishment Clause violation when government endorses religion, which occurs whenever a reasonable observer would conclude that official activity sends a message to nonadherents that they are outsiders, not full members of the political community, and an accompanying message to adherents that they are insiders, favored members of the political community (Wallace v. Jaffree, 1985, pp. 67–79).
Lynch v. Donnelly (1984) best illustrates the way members of the Court apply the endorsement test, which, in effect, combines the first two prongs of the ostensibly abandoned Lemon test. Lynch involved a challenge to a city’s effort to erect a Christmas creche on public property. In Lynch, the Court concluded that the relevant observers of the creche were the adult holiday shoppers (significantly, not their children). The reasonable shoppers were determined by the Court to be so inured to commercial settings that the creche (the governmental expression at issue) ought not to strike them as an endorsement of Christianity, regardless of whether testimony in the case revealed that it reasonably did strike an actual observer, certainly the plaintiff, in that fashion (if the impression of the plaintiff had not been reasonable, one expects that the case would have been summarily dismissed). As explained in Lynch,
Government celebration of the holiday, which is extremely common, generally is not understood to endorse the religious content of the holiday, just as government celebration of Thanksgiving is not so understood. The creche is a traditional symbol of the holiday that is very commonly displayed along with purely secular symbols. (Lynch v. Donnelly, 1984, p. 692)
The analysis discounts (a) the objective effect that a governmental action in fact advanced or inhibited religion and (b) the subjective perception that it did. These limitations are not explicit in the language of Lemon or Abington School District v. Schemmp (1963), the cases the Court cited as precedent for the endorsement test. As formulated and applied, the endorsement test appears to embrace a narrower prohibition than prior law, although the Court’s discounting of incidental objective effects accords with the spirit and outcomes of earlier decisional law and with legal commentary that tended to reject mere effect-in-fact alone to justify invalidation of a statute.
Efforts to evaluate the impact of recent jurisprudence must proceed quite carefully, but the emerging trend seems well established and its themes reverberate from prior cases. The need for caution arises from the fact that the most recent developments in Establishment Clause jurisprudence generally concerned financial aid to religious schools, and such schools may present a unique context under the Establishment Clause. The recent jurisprudence therefore need not apply in other cases that challenge governmental actions as inconsistent with the Establishment Clause. The developments, though, do appear to reflect tendencies in other jurisprudential contexts. Most notably, the trend toward allowing state funds to be used, albeit indirectly, by religious organizations parallels the use of religious organizations in the provision of social services. That jurisprudence essentially reaches back to 1899, when the Supreme Court first considered the constitutionality of government construction funding for a hospital affiliated with the Catholic Church. In Bradfield v. Roberts (1899), the Court found that the hospital itself was not a religious institution, which rendered the Establishment Clause immaterial to the case. Given that almost a century has passed since the Court ruled in this case, government funding of religious entities in and of itself evidently does not violate the Establishment Clause as long as the funds can flow concurrently to both secular and religious providers. In fact, most religiously sponsored nonprofit social service organizations receive funding from the government. In its most recent related case, which involves the provision of social services in an educational setting, the Supreme Court upheld federal grants to religious organizations to provide abstinence education and counseling services to school-age teens and to support research related to adolescent behavior. In that case, Bowen v. Kendrick (1988), the Court upheld the Adolescent Family Life Act (1981, 1988). This federal law provides grants to public and nonprofit organizations for services and research in the area of premarital adolescent sexual relations and pregnancy. Among the contested activities funded by the act were counseling and educational services that could be delivered only if they did not promote abortions. The Court reasoned that, although some grantees had institutional ties to religious organizations, on its face, the law did not advance or excessively entangle secular and religious organizations and that the act served a secular purpose. The Court remanded the case to ascertain whether any specific grantees were “pervasively sectarian,” which might have had the impermissible effect of advancing religion (Bowen v. Kendrick, 1988, p. 621). The funding statute mentioned both religious and nonreligious organizations as potential grant recipients, and the Court found that the law did not direct funding preferentially to religious organizations and thus satisfied the jurisprudential standards used to interpret the Establishment Clause.
Establishment Clause jurisprudence, then, contains several important points. First, the Court permits government actions that entangle religious interests. Second, cases that have been decided tend to be limited to specific contexts, especially to the public schools, so that jurisprudential developments do not necessarily transfer from one context to another. Third, jurisprudence generally permits the neutral provision of public funds, and it especially recognizes the need not to discriminate against religious organizations that offer secular services. The Court allows government benefits to flow to religious institutions that provide secular services; even when the services are not entirely secular, the benefits may still reach sectarian institutions. Fourth, jurisprudence focuses on individual beneficiaries, the Court is more likely to permit disbursement of funds that reach religious institutions when the provisions can be framed as allowing individuals to use public funds for private ends, including supporting services offered by sectarian institutions. This follows from the strong presumption that the provision of governmental benefits to both secular and sectarian institutions is permissible as long as citizens may make their own private choices. This also follows from the strong presumption that religious convictions must be private in their provenance and that religious activity must be the product of the private judgments, convictions, and devotions of individuals and groups. Last, the Court generally lacks interest in empirical investigations to determine the extent of governmental establishment. The Court seems to presume regularity in the action of publicly funded employees in religious settings and emphasizes that law must be decided first and foremost on constitutional, rather than political or statistical, grounds.
Given that the historical approach to resolving church-state issues has rested on a separationist approach, it is not surprising that legislative mandates, especially federal mandates, pervasively have excluded religious concerns. When religious concerns have been addressed, the legal system has proceeded in a stealthy—but not necessarily pernicious— manner. Religious issues always have been present in the legal process, for example, through the provision of social services supported by government funds (Hamilton, 2001) and the advocacy of religious groups in legislative and judicial branches (Sullivan, 2001). Although religious institutions continue to embrace these efforts, the new approach to religion and religious issues is increasingly variable and visible. Rather than maintaining a wall between religion and the government, the legal system, in recent federal and state legislative mandates, increasingly has become faith-friendly.
The federal government’s latest and most dramatic welfare reform overhaul measure, the Personal Responsibility and Work Opportunities Reconciliation Act of 1996 (2001), heralds much more than a shift in the government’s distribution of welfare funds. The Act contains Section 104, the so-called Charitable Choice provision, which exemplifies recent efforts to reshape the role of religiosity and religious institutions in public life. The Act encourages states to utilize “faith-based organizations in serving the poor and needy,” requires that religious organizations be permitted to receive contracts, vouchers, and other government funding on the same basis as any other nongovernmental provider, and “protects the religious integrity and character of faith-based organizations that are willing to accept government funds.” As enacted, Charitable Choice covers each of the major federal antipoverty and social welfare programs.
The apparent success and popularity of faith-friendly measures has spurred congressional efforts to expand their scope to other social service programs and other federal policy domains. In 1999, for example, both the Senate and the House considered numerous Charitable Choice provisions that nearly succeeded in becoming law. Although none of the measures passed that year, they subsequently were reintroduced, passed, and sent to conference to work out differences between the Senate and the House versions. The Charitable Choice Expansion Act of 1999 (1999) sought to provide an umbrella provision that requires consideration of faith-based providers and prohibits discrimination against religious organizations in all programs in which the local, state, or federal branches of government fund private organizations to deliver social services. In addition, The Fathers Count Act of 1999 (1999), offered by the House, sought to permit Charitable Choice in programs to promote responsible fatherhood. In the Violent and Repeat Juvenile Offender Accountability and Rehabilitation Act of 1999 (1999), the Senate sought to permit Charitable Choice in social services related to juvenile justice. In the Youth Drug and Mental Health Services Act (1999), the Senate also sought to expand Charitable Choice in substance abuse and mental health treatment, prohibiting discrimination against religiously based substance abuse programs and allowing such organizations to accept funding without “impairing the religious character of the organizations or the religious freedom of the individuals.” Similarly, in the Consequences for Juvenile Offenders Act of 1999 (1999), the House sought to permit Charitable Choice in juvenile justice services. Most of these provisions have now successfully passed the Senate, although some are being held in conference to resolve disputes between the House and the Senate over matters unrelated to Charitable Choice (see McClain, 2001; Kole, 2001; Freedman, 2001). Faith-friendly efforts, then, clearly have become de rigueur.
The new provisions signal a fundamental shift in the relationship between government and religious groups. Although nonprofit, sectarian institutions always have played a dominant role in social service provision, especially to the poor, the new statutes propose a fundamental change in how they are required to function. Traditionally, sectarian institutions that received governmental support essentially were required to act like secular institutions (see White, 1999; Bassett, 2001). The new statutes, in contrast, allow religious groups to receive government monies for social programs without requiring them to censor their religious identity. Faith-based groups that provide state-funded services must refrain only from compelling participants to get involved in religious practices. These mandates leave many issues unresolved, most notably the extent to which individual providers may practice their religions when providing services, such as praying before serving a meal to the homeless, refusing to provide reproductive health counseling to adolescents, and refusing to terminate the life of patients despite advance directives indicating otherwise (see White, 1999; Bassett, 2001). In addition, it remains to be seen whether faith-based services could circumvent significant protections against employment discrimination, the strongest of which come through Title VII of the Civil Rights Act of 1964 (1999). Specifically, the Act prohibits certain employers from discriminating in terms of hiring or firing on the basis of “race, color, religion, sex, or national origin” (Civil Rights Act of 1964, 1999, § 2000e-2). Title VII guarantees, however, are not universal across employment sectors; employment discrimination protections do not apply to any “religious corporation, association, educational institution, or society with respect to the employment of individuals of a particular religion to perform work connected with carrying on by such corporation, association, educational institution, or society of its activities” (Civil Rights Act of 1964, 1999, § 2000e-(1)a). The religious exemption clearly allows only discrimination on the basis of religion, but the resulting impact may be quite broad, as exemplified by a slew of controversial cases permitting religious organizations to discriminate against employees based on their sexual orientation (see Hall v. Baptist Memorial Health Care Corporation, 2000) and sexual behavior deemed immoral (such as allowing the dismissal of a single, pregnant parochial school teacher on the grounds that her behavior violated the religious and moral precepts against premarital sex, see Cline v. Catholic Diocese of Toledo, 2000). We have yet to understand the full impact of this potentially broad exemption that allows faith-based services to be administered by groups that seek to uphold and practice views that discriminate against others. Although many issues may remain to be settled, legislation already has settled and clearly embraces use of religious groups to supplement (and supplant) governmental efforts to distribute and transform existing social service delivery.
Although initially meant to increase access to traditional social services for those in poverty, the shifting orientation strikingly extends to other government-supported services. Most notably, the change helps to account for an increase in the number of students opting to attend private, sectarian schools at taxpayers’ expense. It also provides a rationale for school reform that offers parents government resources, in the form of tax incentives and vouchers, to send their children to schools otherwise funded by religious institutions (Todd, 2000). These are extraordinary developments. Rather than simply allowing religious institutions to exist and to supplement government services, the new mandates actively support parts of their religious missions. The new mandates even allow religious institutions to supplant government-provided services, since legislatures and the public perceive some services provided by private, religious institutions as more effective than those provided through strictly secular, governmental efforts. Thus, for example, early critics of voucher programs argued (and lost the argument) that vouchers may compel students either to attend sectarian schools or to forfeit their chance for a better education, since vouchers would leave public schools underfunded and increasingly deficient (Weishaar, 1994). How and whether religious groups will be able to expand their social service delivery by taking advantage of federal and state monies remains to be seen. It appears indisputable, however, that the general trend toward privatizing government services increasingly views religious institutions as viable alternatives to be treated like other private or non-for-profit service organizations.
A succinct review of complex jurisprudence necessarily glosses over important controversies. The danger of playing down controversies seems especially worth emphasizing in the contexts of laws that regulate religion. Members of the Supreme Court themselves and commentators often report feeling frustrated by what many of them view as inherently contradictory principles on the subject, a frustration that has led some to accept the impossibility of devising a coherent theory of the government’s regulation of religion and to pronounce efforts to formulate such theories “foreordained failures” (Smith, 1995, p. 45). In our analysis, though, we avoid much of the controversy and replace it with a consideration of the significant implications that arise from existing trends.
The first critical implication of the current regulation of religion involves the focus on protection from state coercion and state promotion of religious beliefs and practices. For example, a key justification for the movement from separation to religious accommodation in school cases is the legitimacy of providing government funds in a way that allows individuals to make private choices (in contrast to following government mandates) that result in the promotion of religious views and/or benefits to religious institutions. Also included under the “private actor” rationale is judicial receptivity toward incidental government aid that flows to religious schools because of individual parents’ private decisions about where and how to educate their children. The Supreme Court has distinguished aid that takes the form of an impermissible “direct subsidy” from aid that represents a permissible “transfer,” much as a state employee may donate part of his or her salary to a church. The revitalized child benefit doctrine, coupled with the view that aid can legitimately flow to religious entities as a result of personal choices, ensures greater government assistance to religious schools and increased governmental support for religious expression. It also means that current laws must focus on eliminating coercion from governmental actions. This focus will become increasingly significant, as we see in the next chapter, because it will allow the Supreme Court to support the constitutionality of private decisions to allow expressions of religious ideology in public schools on the rationale that such expression does not come from the government. The Court now interprets many instances of religious activity in public settings as “private” and therefore permissible expressions.
The second key implication involves the failure to focus on active support from the state in a nondiscriminatory manner. The Supreme Court has recognized that the government cannot condition benefits on the relinquishment of constitutional rights (Rust v. Sullivan, 1991), but it has not gone so far as to hold that the government must subsidize the exercise of such rights. If it did, then parents could assert not only a right to select private education for their children but also an entitlement to the same services for those children that the government provides for public school students. Requiring evenhanded treatment of sectarian and secular enterprises would seem to be the next logical step. The nondiscrimination and private-actor rationales support the claim that, in allocating benefits, the government must treat religious organizations the same as it treats secular entities. Although the claim may seem rather unreasonable or even absurd, the Supreme Court itself offered the possibility. The Court’s 1995 Rosenberger decision, for example, seemed to support such nondiscrimination. As the Court noted in its plurality: “We have held that the guarantee of neutrality is respected, not offended, when government, following neutral criteria and evenhanded policies, extends benefits to recipients whose ideologies and viewpoints, including religious ones, are broad and diverse” (Rosenberger v. Rector & Visitors of the University of Virginia, 1995, p. 839). Expanding the nondiscrimination principle to protect all religious claims against differential treatment, perhaps paralleling protections afforded on the basis of gender or race (see Mawdsley, 1998), does not seem to constitute much of a doctrinal leap, especially in light of the past two decades’ jurisprudential developments. Establishment Clause jurisprudence ultimately may adopt a more relaxed interpretation of restrictions on government action, but the result may vary considerably from what many theorists have sought or imagined. “Neutrality” most likely will remain an important value of the religion clauses, but unqualified adherence to this goal would allow—indeed, arguably, demand—state advancement and support of religion far beyond the traditionally more narrow jurisprudence delimiting the bounds of church-state separation.
The third key point involves the extent to which shifts and expansions matter. The recent expansion of permissible governmental accommodations of religion under the Establishment Clause portends significant implications for church-state relations in schools where the accommodation concept originated. Accommodations made to respect Free Exercise rights in public education (e.g., exemptions from specific school observances, activities, and assignments) may affect mainly the individuals treated differently and do not alter secular school programs or activities. The stakes appear considerably higher with accommodations permitted under the Establishment Clause. For example, accommodations in terms of student-initiated devotionals in a public school touch the entire student body. Moreover, accommodations that encourage parents to select parochial schools for their children (e.g., through voucher plans) potentially affect all students. Likewise, many may be tempted to dismiss as insignificant the controversy over student-initiated devotional activities in public education and incidental aid to parochial schools. A brief student-led prayer in a public school graduation ceremony may not highly influence students’ religious beliefs, and the provision of remedial services in parochial schools may not present a serious threat to the advancement of secular schools. Yet, accommodating religious groups by allowing such sectarian observances in public schools or aid to parochial schools eventually may contribute to an increasingly assertive religious establishment in education. Most proselytizing religious activities may come to seem permissible (or at least something to be accommodated) in public education if the Supreme Court ultimately interprets the Establishment Clause as not barring devotionals as long as students make the decision to conduct them and as not barring religious meetings as long as representatives of religious sects work as private individuals, rather than in their official capacities, when functioning within the public school. Similarly, most types of government aid to religious schools may gain protection under a new umbrella if the federal judiciary continues to uphold government aid that flows to religious schools because of parents’ decisions to select parochial education for their children, on the basis that the funds merely follow the child. As we already have seen and will see more concretely in the next chapter, these scenarios are not as outrageous as they may seem and actually approximate quite closely current legal trends and realities.
In addition to being significant, the changing legislative and jurisprudential trajectories may be difficult to undo, given their remarkable momentum. Indeed, one can hope that the expansion of participation by religious institutions in public programs, permitted by the change in standards in government funding cases, will produce salutary consequences. In the broadest sense, these unforeseen benefits might go beyond efficient delivery of government benefits to include a broadly enriching peace and cooperation among diverse social traditions, faith-based and otherwise. However, perfect religious equality, pluralism, and toleration seem inconsistent with the drive to know “ultimate truth” that lies at the heart of many faiths, and whether such a peace can be attained is debatable, even if it is not currently being debated. Nevertheless, such a hope remains a part of constitutional faith, expressed through the gloss on the religion clauses and the overarching and consistent concern with civil peace as a primary end of the constitutional experiment with religious freedom. If anything, the momentum at least suggests a need for an equally aggressive effort to guide policies toward positive outcomes.
A last point worth highlighting involves the manner in which the Court determines whether a religion qualifies as a religion for the purposes of receiving protection. The cases we have discussed tend to skirt the issue; the Court seemingly has opted for an inclusive, expansive interpretation of what constitutes religion. The proposition has considerable merit. The Court, for example, does not require membership in an organized religious group (United States v. Seeger, 1965), or adherence to tenets, beliefs, or teachings supported by organized religious bodies (Frazee v. Illinois Department of Employment Security, 1989) in defining who is religious. Indeed, in Welsh v. United States (1970, p. 340), a case that put an end to a series of cases involving conscientious objectors to federal conscription, the Court held that “if an individual deeply and sincerely holds beliefs that are purely ethical or moral in source and content but that nevertheless impose upon him a duty of conscience … those beliefs certainly occupy in the life of that individual ‘a place parallel to that filled by … God’ in traditionally religious persons.” It is difficult to imagine a broader approach to determining what constitutes religious beliefs for the purpose of attaching constitutional significance. The need for an inclusive approach is understandable when we consider the fullness of the constitutional clauses that regulate religion; the prospect that churches and states might launch legal actions to determine what has religious meaning touches the very core of the constitutional guarantee against the establishment of religion. As a result, the Court aptly has noted that the judiciary must avoid examining the tenets of religious bodies because these inquiries impermissibly entangle the government in religious affairs (Employment Division, Oregon Department of Human Resources v. Smith, 1990). Given this need, the Supreme Court generally uses the term “religion” unreflectively, as though it were completely self-defining (see Pealver, 1997).
Although it leaves room for considerable discretion and properly admits that injustices might follow efforts to evaluate the “religiousness” of beliefs and practices, the Court’s approach to defining religion for the purposes of protection from state interference is cause for concern. Most notably, the Court allows infringement of religious beliefs when they violate broader social norms. For example, the Court allows infringement on religious rights when they contradict established ideals—monogamous marriage (as when the Court upheld laws prohibiting polygamy, Reynolds v. United States, 1878) and established laws providing for child protection (as when the Court upheld laws against child labor, Prince v. Massachusetts, 1944). The Court does protect religious beliefs when they reflect dominant social values (as when it allowed the Amish to remove their adolescents from public schools on the ground that the Amish religious community offers proper civic values, Wisconsin v. Yoder, 1972). The three case examples may not seem especially troublesome in that they reveal a need to abide by dominant social values, but it does seem peculiar that religious protections seem to offer less protection to minority religions (see Wybraniec & Finke, 2001). To be sure, the cases may have been rightly decided and may reflect the deep roots of our jurisprudential values. But, precisely which values are protected increasingly becomes contentious. Many leading scholars now note that the current Court has become more accommodationist toward the “major” or “mainstream” religions, namely the dominant Judeo-Christian faiths, and substantially less protective of “smaller” or “nonconforming” religious groups (Choper, 2000). The shift in Establishment Clause jurisprudence, the rise and fall of strict church-state separation, has seemed to mirror the rise and fall of anti-Catholicism, since many religious groups now share the Catholic Church’s political and social agenda (e.g., regarding religious schools and family life) (see Berg, 2001). Such protection seems inevitably linked to the need to operate within the confines of existing jurisprudential values that are rooted in and influenced by the dominant religious faiths (e.g., those that protect beliefs and truths that emerge from unfettered, individual choice rather than from state or church mandates; see, e.g., Craycraft, 1999). Yet, even if the courts could look beyond their own religious beliefs and dictates, they nevertheless must rule within established legal boundaries. In fact, shifting jurisprudence may account for the observed trend toward less strict separation between church and state, which may be seen as the product of greater judicial restraint on the part of the present Supreme Court with respect to individual rights—an increased willingness by the prevailing majority to defer to the elected branches of government and a greater reluctance to reject government action in favor of personal liberty. Regardless of the reasons for the trend, the inconsistencies become increasingly problematic in light of emerging legislative directions. As we have seen, modern legislation and jurisprudence increasingly allows the state to provide support for faith-based institutions; how the legal system will determine whether less dominant religious groups can receive similar support remains to be determined. In the end, it seems clear that tradition has rendered the Court reluctant to examine religious tenets and institutions, except when they are not those that are socially dominant—an exception that leads commentators to conclude that the legal system remains much more responsive to some religions than to others (see Palmer & Hardman, 1999).
How society regulates religion is the cornerstone of our effort to understand how society approaches the role of religion in adolescent life. This is especially significant given the recent increase in legislation that directly involves religious institutions and in Supreme Court cases that involve religious issues. Understanding the opportunities and boundaries, as we have seen, requires a firm understanding of constitutional mandates. Regrettably, the mandates are far from clear. On its face, the Constitution’s Free Exercise Clause prevents the government from banning religion, while the Establishment Clause prevents the government from requiring anyone to practice a particular religion. Given the pervasive reach of governmental regulations in modern life, efforts to adhere to these principles result in complex jurisprudential doctrine. To further complicate matters, the intentions of the drafters of the U.S. Constitution are not clear and do not point unequivocally to any indisputable single way to address the First Amendment’s guarantees. The divergent initial positions of the Founding Fathers are reflected in the divergent views of the current Supreme Court justices and of Americans on religion’s place in civil society.
These divergent views often lead many to describe the Supreme Court’s religion case law as confused, inconsistent, and incoherent. Jurisprudence admittedly remains incoherent if one searches for a reasoned elaboration of principles grounded in constitutional text or tradition, or if one expects a sustained, systematic effort to discern and develop the underlying values of the religion provisions in a reasonably consistent, narrow manner. The Supreme Court renders split decisions in almost all important issues relating to freedom of religion and separation of church and state. The Court often draws nuanced lines on the basis of the particular facts offered in each case. This approach means that the current development of the law provides more general themes than categorical principles and fails to provide the doctrinal stability expected from the Supreme Court. This area of law, however, is not as incomprehensible as many claim. Much can be gained by a close examination of what the Court has found worth addressing and worth ignoring.
A close look reveals significant doctrinal shifts that, in and of themselves, reveal important jurisprudential stances. First, the shifts in jurisprudence reveal the fundamental point that interpretations of the clauses can and do in fact change. Second, the shifts reveal that attempting to define the constitutional parameters for the conflicts among citizens, religious entities, and the government does not yield any clear, absolute rules. The Constitution casts the religious clauses in absolute terms; the two clauses, if expanded to a logical extreme, clash with each other. Third, the divergent initial positions of the Founding Fathers are reflected in the divergent positions of judges and justices today, leading to a lack of a single, dominant voice from the courts. Fourth, creating a coherent jurisprudence from the clauses remains exceedingly challenging, given the remarkable extent to which public life implicates religious beliefs and organizations. Fifth, the Court necessarily must balance numerous interests and efforts to protect religion. Last, despite consistent litigation asking the Court to balance the competing interests and to offer guidance, it has yet to address many important issues. In what it has addressed, however, its rulings reveal much substance.
The Supreme Court has increasingly allowed governmental interference even in the face of Free Exercise challenges. The Court has developed a remarkably low governmental standard for evaluating the permissibility of infringements upon Free Exercise rights. Generally, laws that incidentally burden religious practice are permissible if they are neutral laws of general applicability. Despite this apparently permissive approach to regulating religion, however, when laws are not meant to be neutral or of general applicability and if they concomitantly burden other important rights, the Court will impose more demanding obligations on the state to justify its laws and will require that they be narrowly tailored in their intrusion on religious practices. As we have seen, the Court especially looks favorably upon Free Exercise claims when governments can accommodate those claims without excessively burdening themselves. Despite these considerations, it seems that laws do not reach this high level of scrutiny and that religious rights do not remain as inviolate as many would believe.
Establishment jurisprudence reveals an equally impressive amount of settled doctrine and some notable voids. The Supreme Court has generally accepted that government actions can entangle religious interests. Leading cases generally permit the neutral provision of public funds to religious institutions that provide secular services; when the services are not entirely secular, the benefits may still reach sectarian institutions. In its analyses of public funding of services provided by religious institutions, the Court has focused on individual beneficiaries and is more likely to permit disbursement when it can frame the provisions as allowing individuals to decide whether to use the public funds for private, sectarian ends. In addition, the Court is generally uninterested in empirical investigations to determine the extent of establishment. The Court seemingly presumes regularity in the actions of publicly funded employees in religious settings and emphasizes that law must be decided first and foremost on constitutional, rather than political or statistical, grounds. It remains to be determined how this lack of concern for empirical fact will manifest itself in other areas of regulations that deal with religious matters.
Complexities abound as we try to understand and regulate religion’s role in public and private life. When we look at how the legal system regulates adolescents’ religiosity and religious environments, the complexities become even more striking. The general rules that guide constitutional jurisprudence still hold, but they take a different shape because they must consider (but often fail to do so) adolescents’ abilities and vulnerabilities. We now turn to those complexities.