A. Lemon test
For many years, the Court exclusively used the Lemon test to analyze Establishment Clause issues - under Lemon, a challenged government program was constitutional if it satisfied a 3-prong test - (1) if it had a secular legislative purpose (that was not a sham), (2) a principal or primary effect that neither advanced nor inhibited religion and (3) did not create an excessive government entanglement with religion (page 119). The first two prongs of the Lemon test were first announced in Schempp (page 38) and the final prong in Walz (page 61). While the test has now been substantially modified in the parochial school aid context in Agostini v. Felton (see C below) and often ignored in other contexts, its exact status outside the parochial school aid context is not clear. Lower courts still commonly apply the original 3-prong version of the Lemon test in other than parochial school aid cases as one of several alternative tests they apply to analyze the constitutionality of government action challenged under the Establishment Clause. Moreover, the Supreme Court has used the Lemon test as originally formulated in recent years in cases involving school-sponsored religious activity such as football prayer. In addition, the test has never been overturned by the Supreme Court despite much criticism of the test. Rather than rejecting the test, a number of members of the Court have now expressed the view that no one test is useful in all Establishment Clause cases. Therefore, different factual contexts require the use of different tests and even different versions of the same test, as is the case with the Lemon test (see Agostini test in Section C below).B. Endorsement Test
Justice O’Connor suggested a clarification (or gloss or enhancement) of the Lemon test in her concurring opinion in Lynch v. Donnelly (pages 443-45). Under her approach, both the purpose and effect prongs of Lemon are examined through the lens of endorsement. Under this clarification, the issue is whether the government has a purpose to endorse religion and whether the effect of the challenged practice is to endorse religion so as to “send a message to nonadherents that they are outsiders, not full members of the political community” and a “message to adherents that they are insiders, favored members of the political community.” Under Justice O’Connor’s version of the test, whether the effect is to endorse religion or not must be viewed from the vantage point of “a reasonable observer who evaluates whether a challenged governmental practice conveys a message of endorsement of religion.” In her view, “the reasonable observer is knowledgeable and aware of the history and context of the community and the situation in which the religious practice occurs.” Even though Justice O’Connor offered her endorsement test as a gloss on Lemon rather than as a separate test, the endorsement test is often used by lower courts as an alternative to the Lemon test. Typically such courts alternatively analyze the case under Lemon and the endorsement alternative. The endorsement test has been criticized, much as the Lemon test has, by some members of the Supreme Court including Justice Kennedy. However, it is sometimes used in opinions of the Court either as a gloss on Lemon or as alternative to Lemon. Although Justice O’Connor first crafted the test in a case involving a religious display on government property, its use has not been limited to display cases. It is also used to evaluate the constitutionality of public school practices such as school prayer.C. Agostini Test
The Lemon test was modified in Agostini v. Felton (page 244), a case involving aid to parochial schools. Under the Agostini modification, there are only two prongs - purpose and effect. The effect of advancing or inhibiting religion is now reviewed, at least as stated in Agostini, by a focus on three factors: (1) whether the program results in governmental indoctrination of religion; (2) whether it defines its participants by reference to religion; and (3) whether it creates an excessive entanglement between government and religion. Therefore, excessive entanglement, instead of being a factor unto itself has become just one facet of the effect inquiry. While the Agostini test has been used in cases that challenge both direct and indirect government funding of parochial school education (like Mitchell v. Helms (page 259) and Zelman v. Simmons-Harris (page 281)), it has not yet been applied to other Establishment Clause cases such as school prayer. In fact Lemon was applied in 2000 in Santa Fe Independent School District v. Doe (page 387), the football prayer case, even though it was decided 3 years after Agostini. Therefore, it is not clear that the Agostini test will be used outside of the parochial school aid context.D. Strict Scrutiny
Since neutrality among different religions is a central concern of the Establishment Clause, it is not surprising that the Court employs strict scrutiny (the government must demonstrate that it seeks to accomplish a compelling governmental objective and that it is employing means that are necessary means to accomplish that objective) to analyze Establishment Clause cases where the statute on its face discriminates against a particular religion. However, the government rarely enacts a facially discriminatory measure. The Court has also employed strict scrutiny in cases that do not involve facial discrimination against a particular religion, but where the legislature intended to discriminate against a particular religious group and enacted a measure that is the functional equivalent of a facially discriminatory statute. An example of such a statute is found in Larson v. Valente (page 526), the case in which the Minnesota legislature passed a law that was purposefully intended to subject “the Moonies,” otherwise known as the Unification Church, to worse treatment than mainstream religions. In that case, mainstream religions received an exemption from various regulations that applied to other charities, but the Unification Church did not. The Court interpreted the statute as being an example of facial discrimination against a particular religion even though the statute did not refer to “the Moonies” by name. However, the statute did differentiate between two types of religious organizations - those that got the majority of their contributions from members and those that didn’t. Since the statute on its face did distinguish between two types of religious organizations, the Court applied strict scrutiny and found the law to be unconstitutional. Because the statute did not discriminate on its face, however, the Court also considered whether the statute was constitutional under the Lemon test and found that it failed that test as well. By contrast, in Hernandez v. Commissioner of Internal Revenue (page 533), the Court rejected the argument that the federal government was discriminating against particular religions when Congress defined "contributions or gifts" under the Internal Revenue Code. The Court, therefore, applied the Lemon test rather than the strict scrutiny test.In rare cases, the Court will uphold a practice that could not withstand scrutiny under any of the traditional Establishment Clause tests because the challenged practice is a long-standing unbroken practice that dates back to the country’s earliest days such as the legislative chaplain and prayer in Marsh v. Chambers (page 431). Another example occurred in Walz v. Tax Commissioner of City of New York (page 61) where at least part of the reason for upholding the property tax exemption was based on the fact that it was a long-standing practice. Since the widespread existence of public schools in this country only began in the middle of the 19th century, the Court does not use this exception to uphold practices in the public school context. This historical exception to the Establishment Clause has only been relied on in a few cases.
F. Specific Contexts
Having looked at the various tests, I’m now going to review the various contexts in which Establishment Clause challenges occur.(1) Government Funding of Parochial Education
Purpose - In cases involving government funding of parochial education the government is usually able to assert a credible, secular purpose such as furthering educational opportunities or improving the quality of education.
Effect Prong
(a) programs which benefit a broad class of beneficiaries including religious and nonreligious beneficiaries such as the tax exemptions in Walz and tax deductions in Mueller v. Allen (page 176) or a program to benefit all schoolchildren like the bus transportation in Everson v. Board of Education (page 1) are more acceptable in their effects than programs that benefit a narrow class that includes parochial schools or religious institutions, but not all schools or schoolchildren. In recent cases, this concern with the breadth of the beneficiary class has been incorporated into a concern with neutrality or evenhandedness in the criteria employed to distribute the aid. In this inquiry, the Court is more concerned with the scope of the eligible class than the composition of the group that actually is able to utilize the benefit, as it made clear in Mueller. A related question arose in Zelman v. Simmons-Harris (page 281), where the members of the Court debated the nature of the choices available to participating students with some members of the Court (the dissent) focusing only on the schools accepting vouchers and others (the majority) focusing on all educational choices including community and magnet schools.
(b) An impermissible effect is usually avoided if the program provides its benefits directly to parents and students and they are transferred to parochial schools only as the result of independent private choices by the direct beneficiaries and not by the government. These indirect aid cases include Mueller, Witters v. Washington Department of Services for the Blind (page 215), Zobrest v. Catalina Foothills School District (page 237) and Zelman and such an arrangement usually results in the program being upheld. This is even true in cases where the government money, once it passes through private hands, is used to pay all of the costs of education at a pervasively sectarian educational institution as in Witters.(c) Benefits that involve the use of public school employees providing varying services at the site of a parochial school are constitutional after Agostini - these include counseling, testing, psychological services, health services, speech and hearing therapy, diagnostic and therapeutic services, remedial education, special education, and programs for gifted children. Thus far all of the programs upheld by the Court have fallen into two categories: they have supplemented and not supplanted the education provided by the parochial school or they provide auxiliary services rather than direct provision of education. The constitutional status of a program that supplanted the education the parochial school would otherwise be required to provide has not yet come before the Supreme Court.
(d) After Agostini, it is acceptable to directly aid the educational function of religious schools so long as the government does not aid the religious aspects of that education. The earlier notion that it was impossible to separate the secular and sectarian aspects of the education provided by a pervasively sectarian elementary or secondary school has been abandoned. In addition, Agostini abandoned the concern that government employees would be influenced by working in the environment of a parochial school and, therefore, needed to be monitored. The Court has not revisited the issue of whether parochial school teachers need to be monitored if the state is paying their salary, the precise issue raised in Lemon. It did, however, conclude in Mitchell v. Helms (page 259) that parochial school teachers can be presumed to correctly use equipment provided by the state and only use such equipment for secular teaching (plurality opinion by Justice Thomas) or, at least presumed to correctly use such equipment if there are safeguards in place (concurring opinion by Justice O’Connor) to monitor use (see section (e) below).(e) indirect economic benefits or the loan or provision of equipment that is nonreligious in character (Mitchell v. Helms) are more acceptable than direct money payments to parochial schools, particularly where those payments can be used to fund the religious aspects of parochial education. The Supreme Court’s fear that the government will be directly funding the religious indoctrination that is part of parochial school education surfaces in cases in which state money is given to a parochial school to pay for work done by parochial school teachers that lends itself to inculcation of religion - composing tests in Levitt (page 143) or teaching after school classes in Grand Rapids School District v. Ball (page 185). This concern has not been repudiated in the more recent cases. The Court does not have this same concern where the services provided do not lend themselves to indoctrination such as grading state-prepared tests. In cases where the aid could be diverted for religious use, like the computers in Mitchell v. Helms, a bare majority of the Court, including Justice O’Connor, rejecting the view of the plurality opinion, wants to assure itself that there is sufficient monitoring of use so that diversion is unlikely. Such monitoring does not create excessive entanglement.
(f) While the Court has not abandoned completely its concern with the creation of a symbolic union of church and state as seen through the eyes of impressionable children or the appearance of government endorsement of parochial education, it no longer is as likely to find the existence of such an impermissible union. For example, the Court does not assume that such an impermissible appearance is created by the presence of government employees or equipment purchased by the government in a parochial school. In addition, while the Court is still concerned with the appearance of endorsement in the public school setting, as can be seen in the recent school prayer cases where this factor played a role in invalidating the challenged practices, this impermissible effect is downplayed in the public forum cases including Mergens (page 319) and Good News Club (page 397). In Mergens, the Court was not concerned that meetings of a student Bible club would give the appearance that the school endorsed the club’s activities. In Good News Club, the Court rejected arguments that elementary school children would conclude that the government endorsed the activities of the Good News Club by allowing it to use school facilities immediately after the end of the school day. The Court rejected this view in part because parents would need to give permission for their children to participate in the club and thus the impressionability of the children was not at issue since they were not the relevant audience. In the public forum cases, rather than prevent religious speech, the Court suggests that the schools use devices such as disclaimers to make clear it is not endorsing the private speech at issue.(g) Grants of benefits to institutions of higher education have always been treated as less susceptible to Establishment Clause invalidation under the effects prong for a number of reasons: college students are less impressionable, many religiously affiliated colleges and universities are not pervasively sectarian, and private colleges and universities are dominated by secular institutions and are not overwhelmingly religiously affiliated unlike their elementary and secondary school counterparts.
(h) Impermissible effects are sometimes not found by the Court because of the separation of a facial challenge to a program and an as applied challenge. In a facial challenge, a court will not examine facts that relate to particular applications of the law. In its earlier cases, the Court was willing to examine facts related to the general operation of the challenged law as part of a facial challenge to the law. More recently, the Court has shown a tendency to deem such operational or application facts as only relevant to an as-applied challenge. This "divide and conquer" strategy explains the Court’s unwillingness, for example, to look at the behavior of religious grant recipients in Bowen v. Kendrick (page 221) when the statute on its face provided for the delivery of secular services. This recent trend is seen in parochial school funding cases, but has not been consistently applied in Establishment Clause cases outside the funding context.Prior to Agostini, entanglement often created a Catch-22 paradox for the government. If the government failed to supervise government employees who provided services at parochial schools, it risked an impermissible effect; if it constantly supervised those employees, it created excessive entanglement. With the elimination in Agostini of the presumption that public school employees would act impermissibly when teaching in a parochial school, the need for constant supervision was eliminated.
The Court has recognized a number of different forms of entanglement over the years. These include administrative cooperation, political divisiveness and pervasive monitoring. Only the third of these, pervasive monitoring, can, standing alone, create excessive entanglement after Agostini. Excessive entanglement does not exist where there is only administrative cooperation such as filling out forms, providing financial information, occasional inspections, etc. Additionally, the Court has abandoned, or at least downplayed, the importance of political divisiveness which occurs when a law has the potential to produce political debate and division along religious lines. This may occur where the law requires continuing annual appropriations from the legislature and where the appropriated money is directly provided to eligible schools. Such a concern, even if it exists, is limited to elementary and secondary schools and not colleges and universities since colleges and universities generally do not have a local constituency since their student bodies often come from a widely dispersed geographic area. While the concern over a need for pervasive monitoring is still valid post-Agostini, such monitoring is no longer required in numbers of situations, particularly where public school teachers teach in parochial schools. The one recent case where the Court talked about political divisiveness and identified it as a problem was in Doe, the football prayer case, where the school was to oversee an election to determine if there would be an invocation prior to football games (see section (2) below).Another form of unconstitutional entanglement occurs where government lends its authority to a religious organization as in the Larkin v. Grendel’s Den (page 426) case or where it draws governmental boundaries along religious lines as in Kiryas Joel (page 558). This form of entanglement is still a concern to the Court and has been recognized by Justice Thomas as a form of religious establishment.
Entanglement arguments sometimes have a comparative dimension whereby the argument is presented that the entanglement created by the challenged statute is less than the entanglement that would exist in its absence - as in the case of tax exemptions for buildings used as a place of worship.(2) Intrusion of Religion into the Public Schools
Purpose ProngThe purpose prong is particularly important in cases in which religion is introduced into the public schools, as in the school prayer cases, and often is the basis for the Court’s invalidation of the challenged practice (Stone v. Graham (Ten commandments - page 310), Wallace v. Jaffree (moment of science - page 326)). In these cases, the Court looks to evidence of purpose found, for example, in statements of the law’s sponsors and the history of the enactment of the law. Even if the government asserts a secular purpose, the Court may be willing to find that the asserted purpose is a sham as in Edwards v. Aguillard, the creation science case (page 343), and Sante Fe Independent School District v. Doe, the football prayer case (page 387). The Court states that it will defer to a plausible secular purpose, but not if the secular purpose is a sham. In addition, if the law is justified by more than one purpose, the purpose prong only requires that one plausible purpose for the law must be secular. For example, the daily recitation of the Pledge of Allegiance in a public school classroom can satisfy the purpose prong because the school district’s purpose in reciting the Pledge may be to encourage patriotism rather than religion, despite the inclusion of the “under God” language.
EffectThe effect prong may be violated as well when religious practices such as prayer are introduced into the public schools. In such cases, the Court also may rely on Justice O’Connor’s endorsement test and find that the government intends to convey a message of endorsement of religion and that the objective, knowledgeable observer will perceive such a message. In the prayer cases, when relying on whether the government will be perceived to endorse a religious message, a key distinction is between government-sponsored prayer and private prayer. Government sponsorship will be found when there are indicia of state involvement as in the football prayer case even if the prayer is recited by a student speaker. In addition to an impermissible effect of endorsement of religion, an additional aspect of effect in the school prayer cases is focused on by Justice Kennedy in Lee v. Weissman (page 356). In that case, he focuses on the psychological coercion experienced by those that attend the graduation ceremony to participate in the prayer or at least to stand in respectful silence. Some members of the current Court, particularly Justices Scalia and Thomas, but not yet a majority, view coercion more narrowly to only exist when there is legal compulsion to attend or specific penalties attached to non-attendance. Moreover, for those members of the Court coercion is not just an aspect of the effect prong, but is instead a required element of an Establishment Clause challenge. In their view, government religious exercises do not violate the Establishment Clause in the absence of coercion. However, Justice Scalia would require such noncoercive religious exercises, such as graduation prayer, to be nondenominational.
In Newdow (page 408), some members of the Court who reached the merits of the case distinguished the effect of the recitation of the Pledge of Allegiance from the recitation of a prayer. For Chief Justice Rehnquist, the Pledge was a patriotic exercise rather than a religious one and therefore the Establishment Clause was not implicated at all. For Justice O’Connor, the Pledge was acceptable as ceremonial deism that had a secular purpose (either to reflect the country’s history or to solemnize an occasion) and did not create the appearance of endorsement of a particular religion or even religion in general. See Section (4) below for a further discussion of ceremonial deism.Entanglement
Entanglement issues surface on occasion in the cases in which a religious practice occurs in a public school setting. One example of this is in Lee v. Weissman where the school administration gave guidelines to the clergy member invited to recite a prayer which guidelines were designed to spell out the characteristics of a suitable ceremonial prayer to mark the occasion. Doe, the football prayer case, was also notable because it was one of the rare recent cases where the Court was concerned with the political divisiveness aspect of excessive entanglement. The Court feared that such divisiveness would result from subjecting the prayer decision to a vote of the student body.
(3) Intersection of the Free Speech and Establishment ClausesIn a group of cases, we saw the intersection of the Free Speech and Establishment Clauses. In these cases, the government has created a public forum (a place designated for expression by private speakers such as the school facilities in Widmar, Mergens, and Lamb's Chapel) and excluded religious speakers from that forum. The religious speaker, such as the student who is a member of a campus religious group, sues the government under the First Amendment’s free speech guarantee. The government then raises the Establishment Clause as a defense. The government justifies the exclusion of the religious group based on the need to abide by the Establishment Clause. Since these cases involve viewpoint-based exclusions from a public forum, the standard of review the Court uses to examine the challenger’s free speech claim is strict scrutiny. In each case falling within this category, the government lost because the Court concluded that the Establishment Clause did not provide a compelling reason for the government’s action. This was because the Establishment Clause would not have been violated if the government included private religious speakers in the forum and, therefore, it lacked a compelling government interest. The Court’s view was that the government would not be perceived as endorsing the private religious message when it merely included religious speakers in a forum available to a wide diversity of speakers (Widmar, Mergens, Lamb’s Chapel, Rosenberger, Good News Club). It is possible in some circumstances that a disclaimer may be necessary to avoid the appearance of endorsement (Rosenberger). If it is, the government must use that method and others available to it to disassociate itself from the religious speech rather than prohibit the speech entirely. While some members of the Court have suggested that the First Amendment free speech clause always protects private religious speech in a public forum, others have suggested that there may be limited situations where the inclusion of the religious speech creates an unavoidable appearance of endorsement or coercion to participate so that an Establishment Clause violation can be made out.