Can the state of Arizona pay an interpreter to sign the rosary for James Zobrest, a deaf student at a Catholic high school? Zobrest's parents claim Arizona misinterpreted the religion clauses of the First Amendment ("Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof") when it decided to put sign language interpreters in secular, but not religious, private schools. At oral argument on February 24, the justices joked about their incoherent doctrine for policing the Establishment Clause and the embarrassing results it has produced. Adopted in a 1971 case called Lemon v. Kurtzman, the aptly named Lemon test forbids government actions that (1) have no secular purpose; (2) have a "primary effect" of advancing or inhibiting religion; or (3) foster an "excessive entanglement" between government and religion.
"Maybe it's time we tried to straighten out some of that," said William Rehnquist. "Be careful!" stage-whispered Byron White. The false ingenuousness of the exchange was cloying. For the past decade Rehnquist and White, now joined by Antonin Scalia and Clarence Thomas, have been trying to scuttle the Lemon test for an alternative that would permit unabashed government support for religion, including school prayer. Their agenda has been swaddled in historical claims that have been thoroughly discredited by scholars. But there is a better alternative to the Lemon test, more deeply rooted in text and history, that would prohibit prayer in public schools while giving Zobrest the interpreter he seeks.
Under Lemon, courts must strike down government programs that have a "primary effect" on religious education. Now that private as well as public schools receive extensive government funds, this means that the state must discriminate against parochial schools when it distributes benefits. Thus, the Court has construed the Establishment Clause to require what the Free Exercise Clause forbids. To make things even more perverse, the Court subsequently decided that "primary effect" really means any "direct and immediate" effect: the state must be "certain" that parochial schools receiving government funds do not use them to promote religion. At the same time, the Court has interpreted the "entanglement" test to forbid the monitoring necessary to achieve this certainty. Thus the entanglement prong seems to forbid what the effect prong requires.
The Court's efforts to apply the Lemon test are even more absurd. Under Lemon, bus trips from home to religious school are constitutional, but bus trips from religious schools to local museums are unconstitutional. (The latter, but not the former, bring government too close to religion, according to the Court.) Standardized tests are O.K., but teacher-prepared tests are not. Government can provide parochial schools with books but not maps, provoking Senator Daniel Patrick Moynihan's quip: "What about atlases?" The Court has invoked Lemon to strike down a nativity scene surrounded by poinsettias and to uphold a nativity scene surrounded by elephants, teddy bears, Santa's workshop and a talking wishing well. This is the Court's idea of equal time for atheists; practitioners call it the "three plastic animals" rule.
If Zobrest's fate hangs on the Lemon test, the outcome is anyone's guess. The lower court's conclusion -- that a publicly funded interpreter would create a "symbolic union" of church and state -- is farfetched. The court relied on the fiction that students will know who pays for the interpreter, and will wrongly conclude that Arizona endorses the religious message. It would be just as plausible, Michael McConnell of the University of Chicago notes, for students to think that government is hostile to religion when it denies aid to parochial schools that is available to all other private and public schools.
The result for Zobrest, in the end, will turn on which line of inconsistent precedents the Court chooses to follow. Some Supreme Court cases say that state and federal aid to the handicapped can be used for religious education, as long as any benefits to parochial schools are the "incidental" result of private choices. Other cases say that whenever public funds are used for religious education, they impermissibly "advance religion." The justices must also wrestle with existential questions about whether a sign language interpreter is more like a hearing aid, which the Court has permitted, or a tape recorder, which the Court has prohibited.
If the Court chooses to re-examine Lemon -- as seven of the nine justices have promised to do -- then the question for Zobrest is which of the flawed alternatives can command a majority. (Thomas said in his confirmation hearings that he had "no quarrel" with Lemon; months later he joined a Scalia dissent calling for its "interment.") The most conservative justices have flirted with a view that Rehnquist, in a 1985 dissent, attributed to the framers of the First Amendment. In 1789, according to Rehnquist, "The Establishment Clause did not require governmental neutrality between religion and irreligion nor did it prohibit the government from providing nondiscriminatory aid to religion."
Of all the historical claims Rehnquist has floated to mask his conservative political agenda, his claims about religion have been the most thoroughly disproved. In 1986 Douglas Laycock of the University of Texas wrote a law review article and Leonard Levy of Claremont wrote a book responding to Rehnquist's claims. Meticulously reviewing the historical evidence, Laycock and Levy independently concluded that the ratifiers of the First Amendment intended to preserve neutrality between religion and irreligion. (Non-preferential language was proposed and rejected in favor of a more absolute ban prohibiting the federal government from promoting or discouraging religion in any way.) No serious scholar has challenged Laycock's and Levy's conclusions, and David Souter cited them in his school prayer concurrence last June. But Rehnquist, Scalia, White and Thomas have refused to answer the overwhelming evidence against them.
The more moderate justices have been less adept in agreeing on alternatives to Lemon. Anthony Kennedy proposed a test that would have prohibited government only from "coercing" citizens to support religion. Sandra Day O'Connor, who prefers mushy balancing tests, wants to ban government aid that "reasonable people" might perceive as "endorsements" of religion.
In Zobrest, the Court has the opportunity to cut the Gordian knot by replacing Lemon with the very test that Rehnquist rejected: neutrality. A neutrality test would be more faithful to constitutional text and history. It might get five votes from the moderate and liberal justices. And it would resurrect the principle that Justice Hugo Black identified in 1963 as the core value of the Establishment Clause. (Chief Justice Warren Burger mindlessly took the wrong catchphrase from Black's opinion when he cut and pasted the Lemon test.) But neutrality does not mean a wall of separation between church and state. Instead, it means that religion is to be left entirely to private choices.
A neutrality test would give appealingly clear answers to the most controversial questions of church and state. All government-sponsored crèches, menorahs and the like would be unconstitutional, with or without talking plastic garnishes. Vulgarity would be no substitute for neutrality. All prayer sponsored by public schools, similarly, would be forbidden, and there would be no need to resort to what Scalia called "psychology practiced by amateurs" to reach this unequivocal conclusion. Moments of silence, on the other hand, would be less suspect, as long as the teacher's introduction to the students is truly neutral. ("You have the right to remain silent," Leon Wieseltier suggests.) And students would be able to use state aid -- including vouchers -- in religious or secular private schools, without talmudic surveillance from the justices at every turn.
Most importantly, a neutrality test would help defuse the false tension that recent justices have created between the two religion clauses of the First Amendment. Separationists on the Burger Court were wrong to suggest that the Establishment Clause requires the religious discrimination that the Free Exercise Clause forbids. Statists on the Rehnquist Court are wrong to suggest that government can promote "religion in general" while refusing to accommodate the unpopular beliefs of religious minorities. Instead, the Free Exercise Clause protects your right to practice your own religion and the Establishment Clause prevents government from urging you to practice anyone else's. James Zobrest's interpreter doesn't threaten either principle, even if he signs a novena.
Jeffrey Rosen is legal affairs editor at The New Republic and president and CEO of the National Constitution Center.