The Court's decision in Lemon applied to two separate but consolidated cases. The first case involved the constitutionality of a Rhode Island statute authorizing the state to supplement the salaries of parochial teachers who taught secular subjects using secular books. The law was based on the legislature's finding that nonpublic school salaries could not attract quality teachers, thereby jeopardizing the education of nonpublic school students in these schools. The law specifically prohibited the teaching of religion by any teacher receiving such supplemental salary. The second case involved a Pennsylvania statute that similarly provided public funds to nonpublic schools by reimbursing the schools for expenses associated with teachers’ salaries and teaching materials, such as textbooks.

In striking down the statutes, the Court articulated three elements, or prongs, necessary for constitutionality, which would come to be known as the Lemon test: “first, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion;… finally, the statute must not foster an excessive government entanglement with religion” (Lemon v. Kurtzman, 403 U.S. 602, 612-13 [1971]). The first two prongs of the test were taken from Abington School District v. Schempp, 374 U.S. 203 (1963), and the third prong was taken from Walz v. Tax Page 176  |  Top of ArticleCommission, 397 U.S. 664 (1970). With respect to the third prong, the Court stated that the following factors would be considered: “the character and purposes of the institutions that are benefited, the nature of the aid that the State provides, and the resulting relationship between the government and the religious authority” (Lemon v. Kurtzman, 403 U.S. 602, 612–13 [1971]).

Applying this three-part test to the statutes at issue in Lemon, the Court accepted the legislature's stated secular purpose of trying to improve the education received by nonpublic school students, thus satisfying the first prong. However, the Court held that the statutes violated the third prong, finding that extensive and continuous state monitoring would be needed to ensure that the subsidized teachers complied with the statutory restrictions. The Court concluded that this monitoring would excessively entangle the government in the operation of the religious schools. Because the statutes were found to violate the third prong of the test, the Court did not apply or explain the effects, or second, prong.

The Lemon test arose out of the “wall of separation” metaphor articulated in Everson v. Board of Education, 330 U.S. 1 (1947), which was the Court's first significant foray into establishment clause jurisprudence. Although the Court stated in Everson that the establishment clause was intended to erect a wall of separation between church and state, it provided no model or guideline on how to build that wall. Not until Lemon were such guidelines provided. Subsequent to the separationist approach of Lemon, courts became far more reluctant to uphold governmental accommodations of religion than they were prior to Lemon.

Writing for the Court in Lemon, Chief Justice Warren Burger argued that a wall of separation was necessary to prevent a destructive kind of political divisiveness. (The only dissenter in Lemon was Justice Byron White, who concurred in the judgment on the Pennsylvania statute but dissented on the Rhode Island statute.) “Ordinarily,” the chief justice wrote, “political debate and division, however vigorous or even partisan, are normal and healthy manifestations of our democratic system of government, but political division along religious lines was one of the principal evils against which the First Amendment was intended to protect.” The separationist legacy of Lemon was perhaps best revealed in Committee for Public Education v. Nyquist, 413 U.S. 756 (1973), which held that aid provided to parents through a tax deduction was legally no different than providing direct aid to religious schools, and hence a violation of the establishment clause. The Court not only found that the tuition relief in question provided parents an incentive to send their children to religious schools, and hence to practice religion, but it also suspected that public funds would be used by the religious schools for improper sectarian purposes.

Throughout the ensuing decade and a half, the Lemon test became the staple by which to adjudge establishment clause issues. The Court used the purpose prong of the Lemon test in Stone v. Graham, 449 U.S. 39 (1980) to strike down a law mandating the posting of the Ten Commandments in public schools, and again in Wallace v. Jaffree, 472 U.S. 38 (1985) to nullify a statute calling for a moment of silence or prayer in public schools. It used the effects prong in Committee for Public Education v. Nyquist to strike down a law providing maintenance funds to nonpublic schools and tuition reimbursement for parents paying tuition at nonpublic schools, finding that such aid impermissibly subsidized parochial schools. In its applications of the third prong, the Court began to focus on whether the institution receiving public aid was pervasively sectarian and had a primary purpose of inculcating religious values.

In the wake of Lemon, with the Court adopting a separatist view, the Court articulated an establishment clause jurisprudence that appeared to deny any direct government aid to all pervasively sectarian organizations. In a series of cases involving state attempts to aid religiously affiliated elementary and secondary schools— such as Hunt v. McNair, 413 U.S. 734 (1973), Roemer v. Maryland Public Works Board, 426 U.S. 736 (1976), and Wolman v. Walter, 433 U.S. 229 (1977)—the Court reasoned that public aid to pervasively sectarian organizations either inevitably aided their religious function or required an excessive entanglement with government so as to ensure that any public funds were not put to sectarian purposes.

The Court's use of the Lemon test led to persistent charges that the Court was hostile to religion. Even some members of the Court, such as Justices Anthony Kennedy, Sandra Day O’Connor, William Rehnquist, Antonin Scalia, and Byron White, have sharply criticized Lemon. One criticism is that the secular purpose prong often creates the assumption that any law motivated by a desire to promote religious freedom or to accommodate religious practice automatically constituted an establishment of religion.

Aside from this criticism, the Lemon test has produced an inconsistent legacy. In Committee for Public Education and Religious Liberty v. Regan, 444 U.S. 646 (1980) the Court recognized that the inconsistencies of Lemon would continue until it could find a different, less fact-sensitive test. For example, although the Court had held in Board of Education v. Allen, 392 U.S. 236 (1968) Page 177  |  Top of Articlethat states could lend textbooks to religious schools, in Lemon the Court ruled that states could not supplement the salaries of religious school teachers who taught the same subjects offered in public schools. Although it later allowed book loans from public to parochial schools, in Wolman v. Walter the Court prohibited states from providing to religious schools various instructional materials, such as maps and lab equipment. The Court allowed states to provide busing for students to and from religious schools, but it forbade states from paying for the busing costs of field trips for those same students.

In Meek v. Pittenger, 421 U.S. 349 (1975) the Court struck down Pennsylvania's provision of remedial instruction and guidance counseling to parochial school students, only to uphold Ohio's provision of speech and hearing services to such students in Wolman. In Regan, the Court permitted states to pay the costs incurred by religious schools to administer standardized exams, but in Levitt v. Committee for Public Education, 413 U.S. 472 (1973) it prohibited the state from helping finance the administration of state-required exams that were prepared by religious school teachers.

The Lemon test became so unpredictable that the Court took the unusual step in Agostini v. Felton, 521 U.S. 203 (1997) of overruling its earlier decision in Aguilar v. Felton, 473 U.S. 402 (1985) on the grounds that its holding in Aguilar had been undercut by later cases. In Agostini, the Court overturned the injunction sanctioned in Aguilar, which prohibited the government from funding a program providing remedial instruction to disadvantaged children in religious schools. In doing so, the Court altered its stance on several issues.

First, it abandoned the presumptions against pervasively sectarian institutions, rejecting the idea that the placement of public employees on parochial school grounds inevitably results in the impermissible effect of state-sponsored indoctrination or constitutes a symbolic union between government and religion. Second, it rejected the notion that the government will have established a religion if individual recipients divert government funds for religious activities, holding that such a diversion is not attributable to the government. The Court still adhered to Lemon as providing the applicable law, but it refashioned the Lemon test and, in essence, adopted a new test for determining if government aid had impermissibly advanced religion. According to this newly modified Lemon test, any aid passing to religious entities must have a valid secular purpose, must not result in indoctrination, must not define its recipients by reference to religion or contain incentives favoring religious organizations, and must not create an excessive entanglement of government and religion.

By crafting a modified test for evaluating programs that provide public funds to religious organizations, Agostini discarded the presumption that aid to pervasively sectarian institutions would result in the impermissible effect of state-sponsored indoctrination or a symbolic union between government and religion. This presumption, calling for the exclusion of pervasively sectarian entities from governmental aid, was further undermined by the plurality and concurring opinions in Mitchell v. Helms, 530 U.S. 793 (2000), which involved public funding of educational materials for schools in low-income areas. Private schools that received aid under the program were permitted to make only secular use of these materials. The challengers to this program argued that some of the religious schools receiving government funds were pervasively sectarian, and therefore constitutionally ineligible for such aid. Although rejecting a categorical exclusion of pervasively sectarian organizations from direct government support, the Court ruled that any public funding program must have a secular purpose and must be offered to a broad range of groups or persons without regard to their religion. In her concurring opinion in Mitchell v. Helms, Justice O’Connor depicted the establishment clause issue as one of whether the government aid results in religious indoctrination.

In Mitchell, the Court essentially rejected the notion that the establishment clause mandates distrust and pervasive monitoring of religious organizations that receive public aid. Previously, in such cases as Meek v. Pittenger and Wolman v. Walter the Court had operated on an assumption of bad faith on the part of religious organizations, presuming that they would divert public aid to religious uses, and then placing the burden on those organizations to rebut that presumption. Obviously, this presumption would necessitate pervasive monitoring of the uses to which the religious organizations put the government funding, which in turn would violate the excessive entanglement prong of Lemon. In Mitchell, however, the Court found that the government may fund religious entities, but only if the aid is directed to the entity's secular functions.

Due to the criticisms, inconsistencies and unpredictability of Lemon, the Court eventually began moving away from the Lemon test and started introducing other establishment clause tests. Although all attempts at total replacement or nullification of Lemon have been unsuccessful, various modifications and offshoots of the test have been repeatedly introduced. For instance, in County of Allegheny v. American Civil Liberties Union, 492 U.S. 573 (1989), involving the constitutionality of holiday displays on public property, the Court employed Page 178  |  Top of Articlethe endorsement test, indicating that it was a modification of the Lemon effect prong. In Lee v. Weisman, 505 U.S. 577 (1992), a case involving a rabbi-led prayer at a public high school graduation ceremony, the Court used the coercion test. Under the coercion test a religious activity is unconstitutionally coercive if the government directs it in such a way as to force objectors to participate. Finally, in Zelman v. Simmons-Harris, 536 U.S. 639 (2002), in which the constitutionality of Cleveland's school voucher program was upheld, the Court embraced the neutrality approach. Under the neutrality doctrine, courts have ruled that the government must have a neutral purpose, or act with no purposeful discrimination, with respect to its dealings with religious and secular entities.

In Zelman, the Court ruled that indirect aid, such as vouchers provided to parents, terminates any link between government and religion, and thus insulates the government from responsibility for any religious indoctrination of the beneficiaries. This meant that the government did not have to become entangled with religious organizations because it did not have to monitor the use of the funds, since those funds were not directly furnished by the government but were indirectly provided through a voucher system. Thus, it was not the government but the private choices of individual parents that determined where the funds would be spent.

The inconsistencies and modifications of the Lemon test continued with the Court's 2005 decisions in the companion Ten Commandments display cases, McCreary County v. American Civil Liberties Union, 545 U.S. 844 (2005) and Van Orden v. Perry, 545 U.S. 677 (2005). Prior to McCreary County, the purpose prong of the Lemon test simply demanded the existence of a secular purpose. But in McCreary County, which struck down a display of the Ten Commandments on a courthouse wall, Justice David Souter interpreted the purpose prong to mean that the secular purpose predominates over any purpose to advance religion. Thus, McCreary County resulted in a stricter application of Lemon’s purpose prong. However, in Van Orden v. Perry, the Court did not even apply the Lemon test in determining the constitutionality of a Ten Commandments monument on the Texas State Capitol grounds, opting instead for the historical traditions test articulated in Marsh v. Chambers, 463 U.S. 783 (1983). Under this test, courts have upheld certain practices such as a state legislature opening its sessions with a chaplain's prayer by finding that those practices were deeply embedded in the history and tradition of the community.