If conservative and liberal church-state scholars agree on one thing, it is that the Supreme Court’s religious liberty jurisprudence is a disaster. No single rule exists to guide decision making. The various doctrines employed are, at best, inconsistent and, at worst, blatantly contradictory. Divisions on the Court run so deep that actions demanded by “free exercise” according to some Justices violate “no-establishment” according to others. The result is an ever shifting, case-by-case jurisprudence based on narrow factual questions that encourages neither the rule of law nor a robust protection of religious freedom.

The history of contradictory decisions and doctrinal uncertainty could come to an end, however, if the Supreme Court acts decisively during the 2003-2004 term. Two cases on the docket will provide the Court a remarkable opportunity to reconsider its religious liberty precedents. If it can muster the will, the Court could reconstitute its religion jurisprudence, unifying the no-establishment and free exercise provisions into a coherent whole that recognizes the legitimate concerns of both sides of the debate while, at the same time, respecting our nation’s founding heritage.

In potentially one of the most explosive church-state cases ever, the 2003-2004 Court will decide whether public school teacher-led recitations of the Pledge of Allegiance violate the First Amendment’s ban on laws “respecting an establishment of religion.” In 1954, Congress added the words “under God” to the Pledge. When signing the act into law, President Eisenhower declared, “Millions of our school children will daily proclaim in every city and town, every village and rural schoolhouse, the dedication of our nation and our people to the Almighty.” That very reasoning led Michael Newdow, an atheist lawyer with a daughter in a California public school, to challenge the Pledge’s constitutionality. Newdow argued that although his daughter was not required to recite the Pledge, she was injured “when compelled to watch and listen as her state-employed teacher in her state-run school led her classmates in a ritual proclaiming that there is a God, and that ours is ‘one nation under God.’” In mid-2002, a federal Ninth Circuit appellate panel agreed, ruling that the added words constituted an establishment of religion. President Bush and members of Congress from both sides of the aisle loudly decried the decision, which led the Ninth Circuit immediately to delay its implementation and to move for reconsideration. In a surprising show of resolve, however, a second Ninth Circuit panel reaffirmed the original ruling in early 2003, again declaring the Pledge unconstitutional.

Given the public and political outcry, it is not surprising that the Supreme Court agreed to hear the case on appeal. It may be more difficult than it seems, however, to save the amended Pledge. For starters, the Ninth Circuit’s decision is remarkably cogent, follows established Supreme Court precedents, and does not possess any of the characteristics typically associated with judicial mischievousness. The carefully crafted opinion demonstrates that under any and all of the Supreme Court’s three leading Establishment Clause precedents”“Endorsement,” “Lemon,” and “Coercion””the 1954 Pledge violates the Constitution.

The Ninth Circuit began with the “Endorsement” test, a doctrine championed by Justice Sandra Day O’Connor and most recently employed in 2002 by Chief Justice William Rehnquist to uphold the use of school vouchers at private religious schools. As its name suggests, the test prohibits the state from “endorsing” religion over irreligion. The Ninth Circuit concluded that “under God” implicitly endorses a particular religious concept, the existence of a singular deity. The facts in Newdow , moreover, closely parallel those in Wallace v. Jaffree (1985), one of the first and most seminal “endorsement” precedents. In that case, the Supreme Court struck down an Alabama moment of silence law because the state legislature added the words “for prayer or meditation” to the original statute, which said only that the public school day should begin with a “moment of silence.” The Ninth Circuit reasoned that if adding words suggesting that students may pray impermissibly advances religion, then adding explicitly religious words to a teacher-led recital clearly constitutes an endorsement.

The Ninth Circuit then moved to analysis under the first prong of the “Lemon” test, a three-part test originally set forth by Justice Warren Burger in the 1971 case Lemon v. Kurtzman . Justice O’Connor’s “Endorsement” doctrine was designed to replace “Lemon,” but Burger’s framework has shown remarkable staying power despite being frequently ignored. The test’s first prong requires legislation to have a valid secular purpose”that is, the state must be able to offer a nonreligious rationale for its action. Because the words “under God” were not a part of the original text, the Ninth Circuit examined the reasons for the addition instead of looking at the Pledge as a whole. The legislative history of the 1954 Act, they found, clearly indicates that Congress intended to acknowledge America’s dependence on God and to distinguish American republicanism from atheistic communism. Such intentions patently violate “Lemon’s” secular-purpose prong.

The Ninth Circuit concluded with analysis under the “Coercion” test, Justice Anthony Kennedy’s proposal for establishment jurisprudence. Teacher-led recitations of the Pledge coerce religious practice, the Ninth Circuit explained, because they place students in the untenable position of choosing between protesting or participating in an exercise with religious content. Given that students did not have to recite the Pledge, the Ninth Circuit seems to be taking an extremely expansive view of coercion, interpreting it to include actions that could make a student feel psychologically uncomfortable. Yet this is the exact interpretation set forth by Justice Kennedy in his precedent-setting Lee v. Weisman (1992) opinion. In that case, the Supreme Court held that exposing a junior high student to an officially approved nondenominational graduation prayer constitutes undue psychological coercion of religious practice. The Ninth Circuit reasoned that if merely being present while others pray at a single, non-mandatory, end-of-the-year school function is coercive, so too is listening to a daily recital containing religious language.

Given the rules that the Supreme Court has laid down and how they have been applied, the Ninth Circuit’s decision is not outlandish. A candid evaluation must admit that it lies within a fair reading of Establishment Clause precedents. Therein lies the problem and the opportunity for the 2003-2004 Supreme Court.

The Pledge case reveals that something has gone drastically wrong with Establishment Clause jurisprudence. If the Pledge is unconstitutional, so too are teacher-led recitations of the Gettysburg Address. Lincoln claimed “that this nation, under God, shall have a new birth of freedom.” Teaching public school students that the Declaration of Independence is true”that our rights are, in fact, “endowed by our Creator” and that the American Revolution was just according to the “Laws of Nature and of Nature’s God””would violate the Constitution. Even an invited performer signing “God Bless America” at a government-sponsored event, like a local county fair, would be constitutionally suspect. Newdow confirms what critics have long claimed: that pushed to its logical conclusion, the various “wall of separation” constructions of the Establishment Clause are hostile toward religious sentiment and drive religion out of the public square. The case demonstrates that the current interpretations of the Establishment Clause are not neutral and are unworkable and thus fit the criteria for being overturned.

If the Supreme Court does rethink its establishment jurisprudence, it should do so with an eye toward religious free exercise. In the past, the First Amendment’s two religious provisions (note, there is only one clause) have been read independently of one another. If the First Amendment is internally consistent, however, any plausible interpretation of establishment ought to be consistent with free exercise. Free exercise, in fact, is the more fundamental value. Too often it is forgotten that the reason why Congress and the states (since incorporation) are prohibited from making an establishment is that religious establishments tend to abridge religious liberty.

Religious free exercise, including the right not to exercise a religion, is the end; no-establishment is a means toward fulfilling that end.

Perhaps providentially, the Court will hear a blockbuster free exercise dispute during the 2003-2004 term. The case, Davey v. Locke , was filed by Joshua Davey, a college student in the state of Washington. Davey won a “Promise” scholarship during his senior year in high school, a renewable $1,125 state-funded grant that could be used at any public or private accredited college in the state. Davey enrolled at Northwest College, an accredited Bible college affiliated with the Assembly of God, and declared a double major in Pastoral Ministries and Business Management. This prompted a letter from the state informing him that because he elected to major in Pastoral Ministries and because his college educates students from a distinctly Christian point of view, he would forfeit his scholarship. The statute authorizing the “Promise” scholarship explicitly states, “No aid shall be awarded to any student who is pursuing a degree in theology.” The provision was included to ensure compliance with the Washington State Constitution’s “Blaine Amendments,” which declare, “No public money or property shall be appropriated for or applied to any religious worship, exercise, or instruction, or the support of any religious establishment,” and that, “All schools maintained or supported wholly or in part by the public funds shall be forever free of sectarian control.”

Davey seems to present an open-and-shut case of religious discrimination and thus a clear violation of religious liberty. Yet if precedent holds, the Court will not invoke free exercise in its decision; in fact, the Court may not even address the case as a religious liberty controversy. For the past ten years, the Supreme Court has adjudicated religious discrimination cases as free speech questions. This strategy has provided religious citizens significant victories, but at the cost of the near elimination of free exercise as a significant constitutional right. In sum, not only has the Court developed a misguided conception of establishment, it has failed to develop a coherent conception of free exercise.

To understand why free exercise has all but disappeared from constitutional litigation requires understanding two related jurisprudential developments: the rise and fall of the exemption doctrine and the absorption of free exercise into free speech. Starting in 1963, the Supreme Court constructed free exercise to grant religious citizens exemptions from laws and regulations that, absent an overriding compelling state interest, significantly burdened religious life. The classic exemptions case involved Amish parents’ refusal to send their children to school past the eighth grade in violation of mandatory school attendance laws. In Wisconsin v. Yoder (1972) the Court ruled that though the state had a legitimate right to pass a school attendance law and though the law was facially neutral and did not target the Amish religion, the Amish had a right to an exemption from the law because it significantly burdened their religious practice. The exemption doctrine prevailed until Justice Antonin Scalia’s 1990 majority opinion in Employment Division v. Smith . In that case, the Court rejected the appeal of members of the Native American Church who had sought an exemption from drug laws that made illegal their religious use of peyote. Reversing its earlier jurisprudence, the Court ruled that free exercise does not grant religious citizens the right to exemptions from neutral but burdensome laws. Justice Scalia went so far as to say that a presumptive right to a judicially granted exemption from a neutral law itself violates the rule of law.

Smith made clear what free exercise does not include, but it left unresolved what it does protect. If free exercise does not demand exemptions, what does it demand? In 1993, the Court had the opportunity to define its post- Smith conception of free exercise using two disputes that came before it. The first case, Church of the Lukumi Babalu Aye, Inc. v. Hialeah , involved a Florida city’s attempt to make the practice of the Santeria religion illegal. The local city council crafted elaborate regulations that did not mention Santeria by name but clearly were intended to outlaw the religion’s animal sacrifice rituals. A unanimous Supreme Court ruled that laws designed to suppress specific religious practices violate free exercise. The decision received little acclaim, in part because blatant acts of religious suppression are so rare.

The second free exercise case before the court that year, Lamb’s Chapel v. Center Moriches Union Free School District , had a more lasting impact. Lamb’s Chapel, a private religious organization, challenged a local New York school district’s regulations for after-hour use of public school property. The regulations permitted social, civic, and recreational uses, but explicitly prohibited groups from using facilities for religious purposes. School officials, accordingly, rejected Lamb’s Chapel’s application to use school grounds to show a series of James Dobson videos on Christian family values and Christian child rearing. The Supreme Court ruled in favor of Lamb’s Chapel nine to zero, but it rested its decision on the First Amendment’s protection of free speech, not free exercise. The Court reasoned that the school district employed unconstitutional viewpoint discrimination when it rejected an otherwise permitted facility’s use because Lamb’s Chapel approached its subject matter from a religious point of view.

The victory of Lamb’s Chapel encouraged other religious groups suffering from perceived religious discrimination to litigate under free speech. In 1995, a Christian newspaper at the University of Virginia successfully appealed its denial of funding. Following Lamb’s Chapel , a five-member Court majority ruled in Rosenberger v. University of Virginia that the university’s policy of funding all school newspapers except those published from a religious perspective constituted unconstitutional viewpoint discrimination. In a 2002 case, Good News Club v. Milford Central School , six Justices found a free speech violation by a school district whose community use policy explicitly denied the use of school facilities for religious purposes.

From a results-oriented perspective, Lamb’s Chapel , Rosenberger , and Good News Club were victories for religious freedom. In each case, the Court ruled in favor of religious citizens who were denied public funding because of their religious viewpoints. From a constitutional perspective, however, these three cases have had a devastating effect on free exercise. Aside from the rare case in which a specific religious practice is suppressed directly, religious free exercise has lost its independent value. The Court has transformed religion into a subspecies of speech. The effects of this change have been hidden because of the results the Court has reached. But should the Court reconsider what constitutes “viewpoint neutrality,” an inherently ambiguous concept, legal protection of religious free exercise would all but evaporate. More fundamentally, not all exercises of religion can be classified as speech. To the extent religion is more than speech, it possesses almost no constitutional protection under current precedent.

Davey represents a unique opportunity because it reveals the contemporary crisis in free exercise jurisprudence. A creative Supreme Court might interpret Washington’s confiscation of Davey’s scholarship to violate his free speech rights. But to consider the reception of a state-funded scholarship a matter of speech seems to strain ordinary common sense. Davey did not lose his scholarship for anything he said. It is doubtful that he considered his decision to major in Pastoral Ministries to be an exercise of expressive activity. He lost his scholarship because he sought to study religion from a religious perspective. The state did not punish a certain type of speech; it discriminated against religion. His case, accordingly, should be decided under the Constitution’s protection of religious liberty.

Given its standing establishment precedents, however, the Court cannot invoke free exercise to protect Davey from religious discrimination. The Court turned to free speech to protect free exercise in the first place because much of its establishment jurisprudence required state actors to discriminate against religion. The “Lemon” test formally required that government actions have a legitimate secular purpose, neither advance nor inhibit religion, and avoid an excessive entanglement between church and state. In practice, “Lemon” was interpreted to prohibit public money from flowing to religious schools or organizations. The Court reasoned that since money is fungible, government funding for secular purposes could be used by religious organizations for sectarian ends. Justice O’Connor’s “Endorsement” test and Justice Kennedy’s “Coercion” standard were meant to abate the harsh tendency of “Lemon,” but in their own ways they too required state actors to be hostile toward religion. Following the logic of these precedents, many government agencies drafted policies excluding religious organizations from receiving public money or benefits. If only to avoid possible (and costly) Establishment Clause litigation, school boards and university officials adopted policies explicitly discriminating against religion.

In Lamb’s Chapel , Rosenberger , and Good News Club , the Court’s majorities lacked the will to overturn these precedents. Without confronting them directly, however, the Court could not interpret free exercise to prevent the exclusion of religious groups from public programs on account of religion. If they had interpreted free exercise to prevent religious discrimination, free exercise would have prohibited the exact same thing that no-establishment demanded. Turning to free speech allowed the Court to reach the results it desired without overturning precedent. This move, however, only masked and complicated the contradictions in the Court’s First Amendment jurisprudence. Currently, free speech prohibits viewpoint discrimination based on religion. No-establishment, at least according to some standing precedents, demands such discrimination in order to prevent state funding of core religious activities. Free exercise, meanwhile, sits idly by, silently watching its warring neighbors.

With judicious decisions in Newdow and Davey , the Supreme Court could unify the First Amendment’s two religion provisions into one coherent principle of religious freedom. In doing so, it could both bury those establishment precedents that are hostile toward religion and resurrect the right to free exercise. To some extent the Court already has taken small steps in this direction. In 2000, the Court allowed private religious schools to participate in a government-funded program for acquiring computer and library services. Likewise in 2002, the Court upheld the participation of private religious schools in Cleveland’s school voucher program. In both cases, however, narrow Court majorities evaded, somewhat disingenuously, earlier establishment precedents. The Court did not overturn “Lemon,” “Endorsement,” or “Coercion,” which means all three doctrines can still be employed to demand affirmative discrimination against religion.

Overturning these precedents will require the cooperation of Justices O’Connor and Kennedy. In the past, they have been unwilling to break from their favored establishment doctrines, “Endorsement” and “Coercion” respectively. Yet the Pledge case pushes these doctrines as no prior case has before. It shows that they drive religion out of the public square; that they are neither neutral nor fair. Newdow also will bring with it an immense amount of public scrutiny, factors that Justices O’Connor and Kennedy have taken into account in the past. Given the nature and profile of the case, it is hard to imagine a moment when establishment jurisprudence has been more ripe for reconsideration.

If the Court can muster a majority to reconsider no-establishment, it should start by invigorating free exercise. As mentioned above, the First Amendment was drafted to protect religious liberty. Free exercise is the end; no-establishment a means. In fact, without a prior conception of free exercise it is nearly impossible to understand why religious establishment should be avoided at all.

To reformulate free exercise, the Court could do worse than turn to the writings of James Madison. Not only did Madison introduce the Bill of Rights into Congress, he has long been cited as the authoritative guide for the meaning of the First Amendment’s protection of religious liberty. Madison offers a clear and simple statement, moreover, summarizing what “free exercise” means. When editing the religious freedom amendment to Virginia’s state Bill of Rights, Madison proposed the following:

That religion or the duty we owe to our Creator, and the manner of discharging it, being under the direction of reason and conviction only, not violence or compulsion, all men are equally entitled to the full and free exercise of it accord[in]g to the dictates of conscience; and therefore that no man or class of men ought, on account of religion to be invested with peculiar emoluments or privileges, nor subjected to any penalties or disabilities.

Madison interpreted “free exercise” to mean no privileges and no penalties on account of religion. This construction forbids the state from outlawing specific religious practices and forbids state discrimination on the basis of religion. If public school officials open school facilities for after-school use, they cannot single out religious groups for exclusion. Similarly, eligibility for publicly funded programs cannot be conditioned on the absence or existence of religious affiliation. If the State of Washington grants college scholarships to its best students, then it cannot exclude religious individuals who choose to major in theology.

“No privileges, no penalties” also lends itself to establishment jurisprudence. No-establishment would prohibit the state from singling out religious citizens for special legal privileges. Special tax exemptions for religious organizations alone, for example, would be disallowed (although religions could qualify for tax-exempt status as nonprofit corporations or under some other, nonreligious criterion). Legislative chaplains drawing state salaries also would be forbidden. The Founders’ conception of what constitutes an impermissible religious establishment, however, did not include the public recognition of God by state officials. For the Founders, a legal privilege meant receiving material benefits or possessing special rights unavailable to the typical citizen; legal penalties meant fines or imprisonment. By prohibiting religious establishments, the Founders intended to end things like special religious taxes, religious qualifications for public office, and the enforcement of religious orthodoxy through Sabbath-breaking laws. The public recognition of God and voluntary participation in ceremonies or rites that recognize God, even if led by public officials, were never thought to be impermissible. In fact, the Founders sought to solemnize the most important public events by deliberately including religious language and appeals to divine providence in the text of their speeches and documents.

Most of the Founders believed that religious belief supported republican citizenship, and thus, although government could not make an official religious establishment, it could encourage religion in symbolic and non-compulsory ways. Unlike Justice O’Connor’s “Endorsement” test and Justice Kennedy’s “Coercion” standard, then, “no privileges, no penalties” would not require forays into students’ psychological feelings. Judges would not need to inquire if school children feel like “insiders” or “outsiders,” or if a child might perceive the state to be “endorsing” religion, which are necessarily subjective judgments. Courts would only need to ask if, on account of religion, religious citizens as such were granted a material benefit or if nonreligious citizens were subject to a penalty like a fine or imprisonment. For Newdow , the relevant question is: Was Michael Newdow’s daughter subject to some form of disciplinary action because she would not say the Pledge? Since she was not, the Pledge stands.

The advantages of the “no privileges, no penalties” approach are numerous. First and foremost, it is capable of unifying First Amendment religion jurisprudence. No-establishment and free exercise can be the complementary protectors of religious liberty that the Founders intended them to be. The doctrine, moreover, is relatively simple and straightforward. School officials and local governments would no longer face a maze of contradictory rules and regulations. Disputes that do reach the judiciary could be decided more easily and transparently. Perhaps most significantly, “no privileges, no penalties” recognizes the legitimate claims of both no-establishment and free exercise litigants. The Constitution prohibits religious establishments in order to prevent a powerful religious group from legislating special legal privileges for its own members and subjecting nonmembers to particular burdens. Yet in the attempt to prohibit religious privilege, the Court saddled religion itself with particular burdens. Religious groups became the one group against whom legal discrimination was allowed. Religious citizens, in response, rightfully claimed that their right to free exercise was being trampled by extreme antiestablishment decisions. “No privilege, no penalties” prevents any type of religious discrimination, either for or against religion. It affords a robust protection for religious liberty by forbidding religion as such to be a cause for state action.

In the notorious 1947 “wall of separation” case, Everson v. Board of Education , the Supreme Court claimed to set forth the Founders’ interpretation of religious freedom. They missed the mark then and, ever since, the Court has failed to construct a historically accurate, let alone a legally coherent, interpretation of no-establishment or free exercise. The 2003-2004 term offers a unique opportunity for the Court to reconsider its past mistakes. With two decisive decisions that return to Madison’s writings, the Supreme Court could reconstitute its religion jurisprudence, reinvigorating a robust protection for religious liberty.

Vincent Phillip Muñoz is a Civitas Fellow at the American Enterprise Institute and Assistant Professor of Political Science at North Carolina State University.