Lemon Test (Update)

views updated


in lemon v. kurtzman (1971), the Supreme Court announced a three-part establishment clause test, under which a challenged government action, to be valid, must satisfy each of the following criteria: (1) it must have a secular purpose, (2) its "primary effect" must neither advance nor inhibit religion, and (3) it must not create an "excessive government entanglement with religion." The Court applied this test regularly for about fifteen years. Since the mid-1980s, however, the test increasingly has been criticized as excessively hostile to religion, historically misguided, and incoherent. Although the three Lemon factors remain relevant to judicial decisions, their strength has been modified substantially, and the future of even this modified Lemon test remains uncertain.

Only rarely has the Court invalidated a statute under the first part of the Lemon test. In Stone v. Graham (1979), the Court held that a statute requiring posting of the Ten Commandments in public classrooms was unconstitutional for want of a secular purpose. Similarly, in wallace v. jaffree (1984) and Edwards v. Aguillard (1987), the Court invalidated statutes that required, respectively, a moment of silence in public classrooms and the teaching of creationism in public schools. In other cases, the Court readily has found a secular purpose and proceeded to Lemon 's effects and entanglement inquiries.

The Court has developed and modified its criteria for unconstitutional effects and entanglement primarily in cases reviewing government aid to religious institutions. Through the 1970s and 1980s, the premise of the Court's effects analysis was that religious schools are "pervasively sectarian" institutions in which religious and secular education are inextricably intertwined; accordingly, any "direct and substantial aid" to religious schools' "educational function" unconstitutionally advances religion. On this approach, modified substantially in recent years, the Court was most concerned that government-paid teachers or guidance counselors, sent into religious schools to perform even secular tasks, might conform to the sectarian environment and engage in religious indoctrination. The Court invariably invalidated such programs. Government must be certain that its employees do not engage in religious indoctrination, the Court said, and the sectarian environment made such certainty unattainable. Furthermore, even when government tried to avoid these unconstitutional effects—by monitoring its aid programs to ensure that they neither provided nor subsidized religious indoctrination—the Court held that these monitoring efforts unconstitutionally entangled government and religion.

The high-water mark of this approach came in the 1985 companion cases, School District of Grand Rapids v. Balland aguilar v. felton. Even by that time, however, sharp criticism had begun to develop, both off the Court and among the Justices. The year before, in her lynch v. donnelly (1984) concurrence, Justice sandra day o'connor had proposed what she called a "clarification" of the Lemon test, under which courts should inquire whether government has acted with the purpose or effect of conveying its endorsement or disapproval of religion. Alongside this endorsement test, O'Connor presented a weakened version of the Lemon entanglement inquiry. O'Connor's approach was less aggressively separationist than the conventional Lemon analysis, but more demanding than the other theories that began to proliferate on the Court.

One of those other theories was the "nonpreferentialist" approach that then-Justice william h. rehnquist, announced in his 1985 Wallace v. Jaffree dissent. According to Rehnquist, the establishment clause's meaning is defined by its drafters' intention to bar only governmental preference for one sect over another. Four years later, Justice anthony m. kennedy proposed yet another alternative test, under which the clause prohibits government from coercing participation in religious activity or delegating government power to religious groups. While Kennedy obtained a bare majority for his opinion in lee v. weisman (1992), holding that a clergy-led school prayer at a public-school graduation was unconstitutionally coercive, he was unable to persuade his majority that coercion is the appropriate test.

Many commentators in the late 1980s and early 1990s believed that the Lemon test was either dead or in its death throes. But the Court was unable to establish majority support for any of the alternative tests. Still, by 1994 it became clear that five Justices agreed on a more limited modification of the Lemon framework. In three separate opinions filed in board of education of kiryas joel village school district v. grumet (1994), Chief Justice Rehnquist, and Justices O'Connor, antonin scalia, Kennedy, and clarence thomas, suggested overruling the Ball and Aguilarprecedents at the next opportunity.

Those five Justices seized that opportunity in agostini v. felton (1997). Agostini 's procedural context was unusual: the City of New York, still enjoined from providing the program of remedial instruction invalidated in Aguilar, moved for relief from the Aguilar judgment. This context had an important effect: under the Court's earlier interpretation of the relevant procedural rule, the Court could not change the law in its Agostini decision. Instead, the Court could only recognize post-Aguilar changes already in place. Agostini 's five-Justice majority, with O'Connor writing, found those changes in witters v. washington department of services for the blind (1986), and Zobrest v. Catalina Foothills School District (1993). Both decisions had held that if government aid is available generally and neutrally, without regard to the nature of the institution at which it is expended, then the program is not readily subject to establishment clause challenge. Both decisions, further, held that because the challenged aid reached religious schools only through recipients' private choices, the state had not unconstitutionally subsidized religious indoctrination.

inAgostini 's revised establishment clause test, Lemon 's purpose inquiry remains unchanged, and the clause still bars the effect of advancing or inhibiting religion. But entanglement now is but one of three factors in the effects test, not a separate inquiry. The other two criteria of unconstitutional effect are whether the government action results in government indoctrination—either by government employees' own actions, or by state subsidization of religious education—and whether the action operates by reference to religion.

Some aspects of this revision are clear. Agostini decisively repudiates the presumption that publicly employed teachers and guidance counselors are likely to engage in religious indoctrination when charged with purely secular tasks. Agostini is clear, also, that government monitoring for indoctrination is not by itself unconstitutional entanglement. But other aspects are unclear, particularly when one considers the issue of school vouchers, or school choice, that seems destined for Court review. On one hand, a program making vouchers available to all needy students, regardless of whether they attend sectarian or nonsectarian schools, would have the generality and neutrality that now point toward constitutionality. Further, whether vouchers actually benefit religious schools would depend on recipients' private choices—a second factor Agostini identified as favoring constitutionality. On the other hand, however, the Court could invalidate such a program as an unconstitutional subsidy to religious indoctrination. A comprehensive voucher program would have the two consequences Agostini considers relevant to unconstitutional subsidy: government money would flow into religious schools' coffers, and the program likely would relieve religious schools of costs they otherwise would have borne. A program making vouchers generally available for specifically religious education, not just for secular instruction, would go well beyond the programs so far approved.

No doubt one reason for Agostini 's ambiguity is continuing disagreement as to the proper establishment clause test. Another reason concerns the case's procedural context, which prevented the Court from reformulating—or admitting it was reformulating—establishment clause doctrine. The Court would have done better to reconsider Aguilar and Ball in a context that would have allowed Lemon 's systematic reexamination. The Court's claim not to have innovated in Agostini is unpersuasive, and the apparent agreement on a constitutional test could dissolve in the next case, in which the Court would be more free to revise its doctrinal position.

That next case will be before the Court in its 1999–2000 term. The issue in Mitchell v. Helms (certiorari granted on June 14, 1999) is whether the government may provide secular instructional materials, such as computers, to sectarian as well as non-sectarian schools. The lower court applied Lemon precedents from the 1970s to invalidate the challenged aid. Those precedents are at least in tension with Agostini 's rendition of the Lemon test, and the Court may well overrule them. Because the Court will be free in Mitchell from the procedural complications that affected Agostini, it could use the case as a vehicle for altering the constitutional test more systematically—perhaps adopting one of the alternative tests that various Justices have urged in separate opinions, or perhaps adopting a compromise version that presses the neutrality theme less ambiguously than did Agostini. But the Court need not do so to uphold the aid challenged in Mitchell, and nothing indicates that the Court is less divided as to the ideal establishment clause inquiry than it has been in the past. The long-term durability of Agostini 's modified Lemon test likely will remain uncertain until the Court considers the voucher issue.

Hugh Baxter


Baxter, Hugh 1998 Managing Legal Change: The Transformation of Establishment Clause Law. UCLA Law Review 46: 343–457.

Laycock, Douglas 1997 The Underlying Unity of Separation and Neutrality. Emory Law Review 46:43–74.

Levinson, Sanford 1997 Religious Language and the Public Square. Harvard Law Review 105:2061–2079.

Levy, Leonard W. 1986 The Establishment Clause: Religion and the First Amendment. New York: Macmillan.

Mc Connell, Michael 1992 Religious Freedom at a Crossroads. University of Chicago Law Review 59:115–194.

About this article

Lemon Test (Update)

Updated About encyclopedia.com content Print Article