1 Introduction
Measured by lack of judicial consensus, one of the hardest questions the modern Supreme Court has faced is what limits, if any, the Establishment Clause places on government religious speech. The Court has decided dozens of cases about government-sponsored prayers and religious symbols, and although some relatively stable patterns can be divined from the holdings, the justices have never settled on a test, principle, or even rationale to guide officials, litigants, and lower courts. From the beginning, the Court drew a line between government-sponsored prayers in public schools (impermissible) and chaplain prayers before legislative sessions (permissible). Eventually, a majority of the Court decided that the touchstone in cases involving religious symbols was whether a reasonable observer would think the government was endorsing religion. Although the Court has recently said it has abandoned the endorsement test, it continues to evaluate government-sponsored religious symbols by whether they express support for religion.1 The question of the government’s religious speech has been a mess from the beginning, and the Court does not appear to have marked a path out of it.2
Through more than three decades of profuse and trenchant scholarship on the historical and intellectual underpinnings of the rights of religious freedom, John Witte has illuminated a more complete, complex, and coherent way forward. Embracing an “integrative” approach that merges positivist, naturalist, and historicist jurisprudence, Witte has explored not only the text and historical context of the First Amendment, but also its natural rights background and
I generally agree with Witte’s approach and with these conclusions,6 but I have one hesitation. Perhaps we have reached too hastily the conclusion that mere speech can never interfere with religious liberty. Might there be a form of government religious speech that does not involve strong pressure to participate in religious exercise but nevertheless violates the religious liberty protected by the Establishment Clause? My tentative answer is yes: the Establishment Clause prohibits speech that amounts to a threat of religious discrimination. My hope is that this chapter builds constructively upon the notion of religious liberty, and the integrative approach to jurisprudence, that Witte has persuasively shown and persistently modeled in his scholarship for many decades.
2 The Question
The Supreme Court’s diverse and inconsistent decisions have made it difficult to pin down the precise legal question. Scholars generally sort the cases
This chapter asks the question underlying all of these cases: what limit, if any, does the Establishment Clause place on what the government may communicate, and why? Communication includes straightforward speech and writing, as well as the use of symbols or even, perhaps, the message conveyed by an action. In this sense, the question captures the range of forms of communication that are recognized, and often protected, by the Free Speech Clause when performed by a private party. Put in terms of free speech jurisprudence, the question is whether the Establishment Clause places any content-based, or perhaps viewpoint-based, restrictions on the government’s communication.
The justices appear to agree that the clause forbids government religious speech in a setting that coerces participation or indoctrination. This is the minimum doctrinal reading of the Court’s cases involving prayer, Bible reading, and Ten Commandments displays in public school rooms and events. Some justices have maintained that compulsion should be the only limit on government speech, but the Court as a whole has consistently rejected coercion as the sole touchstone of government religious speech. The trouble is that the justices have never settled on another test.
For a time, a majority of the justices appeared to embrace the view that the government violates the Establishment Clause when a reasonable observer would conclude that its conduct or speech has the effect of “endorsing” religion.7 The Court has even applied this test to cases involving prayer at public school events.8 But the endorsement test was never stable. The justices who embraced it consistently disagreed about its application to specific cases, leading to incoherent and inconsistent results. Other justices have offered various permutations on the endorsement test, but none have yet to gain the same purchase.9
In American Legion v. American Humanist Association, the Supreme Court held that longstanding war memorials in the shape of a Latin cross do not violate the Establishment Clause.10 The decision is laudable for reaching broad agreement about religious displays that have been on government property for a long time.11 But it did nothing to eliminate confusion about what the Establishment Clause prohibits the government from saying or why. In fact, the Court’s analysis may have made things worse. Although the Court said that it was not basing its judgment on whether the cross amounted to an endorsement of Christianity,12 its entire analysis boiled down to whether the original context and the passage of time had sufficiently drained what is plainly a Christian symbol of its religious content.13 On this basis, the Court concluded that old religious displays generally do not violate the Establishment Clause. The necessary implication, of course, is that some religious symbols (to say nothing of straightforward religious speech) may violate the Establishment Clause. Underneath a surface of broad agreement about the result remains a cacophony of views about what the Establishment Clause forbids.
In Kennedy v. Bremerton School District, the Court held that a public school violated the Free Speech and Free Exercise Clauses when it fired a football coach for praying publicly at the fifty-yard line after games.14 Although the Court claimed that it had already abandoned the endorsement test,15 the holding did not rely on that assertion. The Court determined that the prayers were attributable to the coach in his individual capacity, and therefore protected. Since the school was not responsible for the prayers, the Establishment Clause—implemented by whatever test—simply did not apply. In fact, the Court made it clear that it was not unsettling any of its prior school-prayer decisions.16 While the case helpfully clarifies the rights of school employees
The question is not going to disappear. Consider some examples of straightforward government religious speech that is not obviously coercive. The official motto of the United States is “In God We Trust,” and it appears on U.S. currency.17 Adults and children alike routinely pledge allegiance “to the Flag of the United States of America, and to the Republic for which it stands, one Nation under God.” Some government entities are named for sectarian religious beliefs, such as Los Angeles, California, and Corpus Christi, Texas.
The government occasionally goes further. President Ronald Reagan declared 1983 the “Year of the Bible,” asserting that “[t]here could be no more fitting moment than now to reflect with gratitude, humility, and urgency upon the wisdom revealed to us in the [Bible]” and “encourag[ing] all citizens, each in his or her own way, to reexamine and rediscover its priceless and timeless message”?18 Government-designated clergy members offer prayers before a variety of events, such as the presidential inauguration, legislative sessions, and Supreme Court hearings (“God save this honorable Court”). Could the government go even further? Could the state of Texas declare that a personal relationship with Jesus Christ is the path to spiritual salvation? Or could the state of Michigan print on its official seal that “There is no God but Allah, and Muhammad is his messenger”? Would religious confessionalism satisfy the Establishment Clause so long as the state does not coerce observers into religious faith or practice?
Whatever limits the Establishment Clause places on religious symbols derives from the limits it places on speech more generally. The challenge posed by symbols is that they are usually more vague or multivalent than words. But the difference is one of degree, not of kind. Speech can be vague, too. When Reagan declared that wisdom was “revealed to us in the Bible,” was he attributing such revelation to God, or to human authors, the way “wisdom is revealed” in secular texts from Plato to Shakespeare? Does encouraging citizens to
Whatever vagueness there might be, deliberate or otherwise, in speech, it is inherent, and often compounded, with symbols. Perhaps judicial prudence would suggest drawing a line of administrability between government speech and religious symbols—whatever outright speech the Establishment Clause may prohibit. But such a line would be the product of judicial prudence, not the Establishment Clause. Sometimes courts underenforce constitutional norms because of concerns about the separation of powers, federalism, or institutional competence. And sometimes they overenforce them, especially if violations are hard to detect. Indeed, as we shall see, overenforcement may be one of the stronger arguments for policing noncoercive government speech.
Knowing that the Establishment Clause prohibits the communication of X does not, of course, resolve the constitutionality of any given communication. Regardless whether the communication is with words, symbols, or both, or whether it is written, spoken, broadcast locally or widely, there will always be a crucial question of interpretation or semiotics: what does this speech or speech-act or symbol mean? What does it communicate? The proper hermeneutic for that analysis is itself a question of law. Whose opinion about the meaning matters, and why? The government’s? The claimant’s? Some hypothetical objective observer? Only once the meaning of the government’s communication is settled can the constitutional rule be applied to determine whether the communication violates the Establishment Clause. This chapter does not have space to fully address these issues, but it is important to recognize that they are logically posterior to the foundational question, which is what sort of communication does the Establishment Clause prohibit, and why?
It is also worth distinguishing the Establishment Clause issue from the question of standing. A federal court has constitutional authority only to decide “cases or controvers[ies]” in which the claimant has standing to sue, which requires the claimant to show a unique and concrete injury.19 Some justices have argued that cases involving mere religious speech, without any coercion, do not give the vast majority of claimants standing to sue because symbols do not create a concrete injury: the only injury is emotional.20 It is important to keep standing conceptually discrete from the merits question, for it is logically and legally possible for the Establishment Clause to prohibit more than anyone would have standing to challenge.
A hallmark of John Witte’s scholarship is the judicious implementation of what he calls an integrative approach to law. This approach is a way of judgment, of discerning both what the law is and what it ought to be, that seeks to marry disparate sources of law, and sometimes divergent legal principles, to provide a coherent approach to understanding and evaluating legal questions.
As defined by Witte’s mentor and friend Harold Berman, “integrative jurisprudence is a legal philosophy that combines the three classical schools: legal positivism, natural-law theory, and the historical school.” The idea is that “each of these three competing schools has isolated a single important dimension of law, and that it is both possible and important to bring the several dimensions together into a common focus.”21 In contrast to modern positivism, now so dominant among scholars, integrative jurisprudence considers both morality and the historical trajectory of the law within a political community. In contrast to a jurisprudence that gives pride of place to the judge’s moral sentiments, it considers natural law to be the universal norms that provide the basis for a political community’s particular solutions to problems of justice, fairness, and peace. And in contrast to originalism, it recognizes that “law is an ongoing historical process, developing from the past into the future.”22 Integrative jurisprudence brings the habits of mind of Aquinas and Althusius, Story and Savigny into the twenty-first century.
Witte has rarely discussed the theory of integrative jurisprudence, but his work consistently applies it. In an early book review, he praised the authors for taking an integrative approach to the religion clauses.23 Witte demonstrated the power of the approach in a 1996 article that laid the basis for what became the standard one-volume work on the history and doctrine of the religion clauses (now coauthored with Joel Nichols and Richard Garnett). With a politically, philosophically, and theologically sensitive account of the late eighteenth century, Witte showed that the overarching historical purpose of the religion clauses was to protect an American conception of religious liberty that combines “a variety of principles,” namely “liberty of conscience, free exercise, pluralism, equality, separation, and disestablishment.”24 These
A challenge for integrative jurisprudence is reconciling, or perhaps choosing among, sources or principles of law in tension with one another. Unlike pure positivism, an integrativist cannot simply dismiss contrary claims of morality or the lived experience of society. Witte identifies this problem, and a possible solution, in an essay devoted to understanding the historical, and proper, role of arguments for religious liberty from the principle of church-state separation. Contrary to Philip Hamburger’s claim that separation of church and state was not an important feature of American religious liberty contestation in the late eighteenth century,25 Witte shows that Americans frequently deployed various (sometimes inconsistent) conceptions of church-state separation. But contrary to some strict separationists (including some members of the Supreme Court), he argues that separation was never an end in itself: it was always in service of the broader principle of equal religious liberty. On this basis, he praises the Court’s use of separationist arguments “to extend the ambit of religious liberty, especially for minority faiths,” but he also chides the Court for invoking the concept to “erode the province of religious liberty by effectively empowering a single secular party to veto popular laws touching religion that cause him or her only the most tangential constitutional injury.” In those cases, it seems, the principle of separation is in conflict with principles of free exercise, liberty of conscience, and pluralism; a categorical requirement of separation would shut down various forms of governmental accommodation of religion. When principles conflict, courts should remember that each of them “serves religious liberty best when it is used prudentially not categorically.”26
Witte has not given sustained attention to the problem of government religious speech, but his work touching on the issue suggests he does not think it threatens the essentials of religious liberty. In several pieces, he offers a characteristically subtle and insightful analysis of the caselaw, culling the often-inconsistent Supreme Court decisions for “rules of thumb” to guide courts, lawyers, and officials through the jurisprudential thicket.27 This suggests that
This view is amply supported by American practice before the middle of the twentieth century. There is little evidence that Americans before then believed that disestablishment forbade the government from sponsoring certain forms of religious speech, whether through government chaplains, the expression of faith by officials, the use of “ceremonial Deism,” or the incorporation of religious symbols into government architecture and design.29
Yet there are countercurrents in the history of disestablishment sounding in the principles of equality, liberty of conscience, pluralism, and separation that together comprise religious liberty. In the first government-religious-speech case, Engel v. Vitale, the Supreme Court invalidated a New York law requiring public-school officials to begin the school day with a scripted prayer. The Court pointed to political disputes in England over the Book of Common Prayer and the view of the American founders that “one of the greatest dangers to the freedom of the individual to worship in his own way lay in the Government’s placing its official approval upon one particular kind of prayer or one particular form of religious services.”30 The following year, the Court leaned heavily on the principles of separation and religious equality (styled “neutrality”), distilled from nearly fifteen years of Establishment Clause decisions, to declare that the Establishment Clause forbids laws that have the purpose or effect of advancing religion. On the basis of this rule, the Court invalidated state laws that required devotional Bible reading in public schools.31 From there, it was not such a stretch to invalidate a state law requiring the passive display of the
Witte’s approach provides guidelines for a more thorough evaluation of the religious-speech tests proposed by justices over the past four decades. First, history and tradition matter. The long history of government religious speech with few to no complaints counsels strongly in favor of understanding the Establishment Clause to place no limits on noncoercive speech. Second, positive law counts, too. The Court has invalidated a variety of noncoercive forms of government religious speech, and those decisions are still good law. Third, there are multiple important and interlocking principles that together inform the religious liberty protected by the religion clauses. Each of them reflects an important value the Constitution sought to implement. Fourth, each of those principles can be exaggerated and should be understood not as categorical requirements but rather valid to the extent that they support religious liberty. A Wittean, integrative approach to the religion clauses would consider whether the subsidiary principles, in light of the history and practice of religious liberty in the American tradition, support a conception of religious liberty that might prohibit some forms of noncoercive government religious speech.
4 Religious Speech Tests Proposed by the Supreme Court
The Supreme Court has sometimes evaluated government religious speech by an Establishment Clause doctrine or method that applies generally to all government conduct, and sometimes by a rule that focuses on the content of the government communication. Before turning to the speech-specific tests, it is worth considering how the Court’s holistic approaches to the Establishment Clause have contributed to the confusion surrounding the issue.
The secular-purpose requirement. Until recently, the Court has held that a law, including one that authorizes or requires religious speech or symbols, must have a sufficiently secular purpose.34 The requirement does not directly answer the question about the content of government speech; it focuses instead on the government’s objectives, whatever the content of the law or government-sponsored communication. The secular-purpose requirement does not have a strong basis in the original understanding of the religion clauses or the traditional understanding of disestablishment. At the founding, an establishment consisted principally of the unequal distribution of rights, privileges, and immunities on the basis of religious belief and practice to promote religious conformity. It also often entailed government control over religious doctrine, clergy, and institutions.35 I know of no instance in which a law or policy was considered to be part of an establishment merely because the government’s purpose was to promote religion or one religion over another without including an unequal distribution of legal rights and privileges.
In light of the history of disestablishment, the secular-purpose requirement was perhaps best understood as a prophylactic heuristic to avoid the end of a religious establishment (induced religious conformity) by avoiding its beginnings (a purpose to promote such conformity). The only Supreme Court decisions invalidating government religious communications for lack of a secular purpose involved public-school classrooms or events with a high risk of religious conformity.36 The Court has recently repudiated the secular-purpose requirement without overruling any of the cases applying it, which suggests that those cases may now be best understood to rest upon a concern about coercion, not illicit government purposes.37 If so, in addition to having little support in the history and tradition of religious liberty, the secular-purpose requirement now has little support in positive law.
The history and tradition of government practice. The Supreme Court has recently abandoned the purpose-and-effects test in favor of history and tradition: the Court determines whether government action violates the Establishment Clause by comparing it to past government conduct that was understood
Yet an integrative jurisprudence does not simply rubber-stamp the past. It also considers the positive law and the moral principles underlying that law. The Court’s decisions invalidating government religious speech are part of the positive law of the land. And perhaps for good reason—the Court has consistently pointed to historically grounded principles of equality, pluralism, and liberty of conscience in religious-speech cases. An integrative approach should take those concerns seriously and seek to synthesize them with the weight and trajectory of history, and with the norms of constitutional law and judicial review more broadly.
4.2 Rules against Specific Kinds of Government Speech
No endorsement of religion. Members of the Supreme Court have proposed various rules against specific kinds of government speech. The rule that has gained the most support, serving as the basis of several decisions, is the rule against government endorsement of religion. The question is whether “the challenged governmental action is sufficiently likely to be perceived by adherents of the controlling denominations as an endorsement, and by the non-adherents as a disapproval, of their individual religious choices.”40
The government rarely endorses religion in so many words, so applying the test ordinarily requires interpretation of government speech, conduct, and symbols. This entails a close analysis of the physical setting of the speech and its historical and immediate political context to determine whether a reasonable
The rationale for the endorsement test sounds in several of the core principles of religious liberty. “Endorsement,” the Court writes, “sends a message to nonadherents that they are outsiders, not full members of the political community, and an accompanying message to adherents that they are insiders, favored members of the political community.”42 The rule seeks to implement principles of equality, pluralism, separation, and liberty of conscience. It prevents the government from announcing a preference for one religious belief or group over others (including those who reject religion). The doctrine seeks to ensure that the government treats religions equally (none of them get preference), acknowledges the reality—and perhaps the good—of religious pluralism, ensures strict separation of church and state, and protects the conscience of members of the religious majority and dissenters who would prefer their government to remain silent on religious matters. All of these principles are vital for a robust regime of religious liberty.
But the endorsement test is unfortunately at odds with other principles of religious liberty, with the historical and traditional understanding of religious liberty in the United States, and with the ordinary principles of constitutional law. Some government religious expressions facilitate the exercise of religion, sometimes in a way that is respectful of pluralism, such as an ecumenical ceremonial prayer. Some, such as holiday religious displays, allow members of the public to see their religious beliefs respected and acknowledged in the public sphere. Indeed, it is impossible for the government to treat religions entirely equally. Many examples of government religious speech are longstanding, and eliminating them now from the public sphere would strike some as unduly hostile toward religion. Even with respect to new forms of religious speech, the government cannot be entirely neutral. Some religious groups believe the government has an affirmative duty to honor God; if the government were mute about the divine, members of those groups would rightly conclude that they were outsiders.
The endorsement test is also at odds with the history and tradition of religious liberty in the United States. As discussed above, there is little evidence that Americans believed that government endorsement of religion, without
The endorsement test is also a constitutional anomaly. The history and tradition of religious liberty in the United States has focused on ensuring that the government does not distribute different legal rights, privileges, and immunities on the basis of religious belief (or nonbelief) or practice. The endorsement test, however, protects hypothetical observers from the psychological or emotional experience of feeling like “an outsider,” of not being able to “take pleasure in seeing the symbol of their belief given official sanction and special status.”43 That is, it protects them from feeling like their religious status somehow affects their political status in the community—even when it has absolutely no effect on their legal rights, privileges, and immunities. Such protection is not only unusual for religious liberty, it is a constitutional idiosyncrasy. No other constitutional restriction on governmental power has been implemented by a doctrine that prohibits the government from making observers feel like it has exceeded that power. Such a doctrine gives individual claimants extraordinary power over democratically enacted laws and policy that have no tangible effect on their beliefs or conduct. Moreover, whatever some observers may feel or believe about their political status as a result of government religious speech, judicial intervention only makes matters worse by either validating that belief (yes, you are right to feel like an outsider) or rejecting it (no, you aren’t); it exacerbates political alienation without eliminating genuine threats to the freedom to believe and exercise religion according to one’s own conscience.
As discussed above, the Court has never truly repudiated the endorsement test. In American Legion, the Court articulated what it suggested was a new rule for symbols cases: symbols that have been in the landscape for a sufficient amount of time, without generating political controversy, are presumptively valid. But to articulate this rule, the Court applied a vague form of … the endorsement test. It considered the history and tradition of war memorial crosses generally, and the specific history of the Bladensburg cross, to determine that time and secular use had drained the cross of its religious meaning. The American Legion rule creates a useful default rule for longstanding religious displays, but that rule is nothing more than a generalized conclusion
The endorsement test protects feelings, not religious belief and exercise. But maybe there is a better justification for the test. Perhaps it operates as a prophylactic against official discrimination or reasonable fears of discrimination on the basis of religion. Under this theory, the endorsement test overenforces the Establishment Clause, but for a good reason: official discrimination is often difficult, if not impossible, to detect, but concerns about future discrimination may affect religious liberty by creating an incentive for dissenters to change their religious beliefs or conduct to avoid discrimination.
Assuming the best case for the endorsement test is that it serves as a prophylactic against religious-exercise-altering official discrimination, the test is still not the best way to police that concern. There are plenty of ways the government may endorse religion without threatening discrimination. Prohibiting such endorsement therefore overenforces the Establishment Clause in the teeth of the history and tradition of religious liberty and general constitutional jurisprudence.
No endorsement of sectarian religion. Justice Scalia championed a permutation of the endorsement test “where the endorsement is sectarian, in the sense of specifying details upon which men and women who believe in a benevolent, omnipotent Creator and Ruler of the world are known to differ (for example, the divinity of Christ).”45 This test would allow endorsement of religion in general, and even of monotheism, but not the endorsement of a doctrine unique to a religious denomination.46 This rule seeks to blend the American tradition of government religious speech with concerns about equality and pluralism. Permitting the government to invoke generic religious norms, Scalia thought, promotes social unity by drawing on beliefs and practices shared by most Americans.
No taking a position on religious questions. The Supreme Court has also said the Establishment Clause prohibits the government from “appearing to take a position on questions of religious belief.”47 The Court subsequently subsumed this concern into the endorsement test, basing it on the same rationales. Professor Andrew Koppelman, however, has offered a more robust defense of a similar rule in the name of neutrality: “The Establishment Clause forbids the state from declaring religious truth,”48 from declaring “any particular religious doctrine to be the true one,” and from enacting “laws that clearly imply such a declaration of religious truth.”49 The state may treat religion, and the private exercise of religion and search for religious truth, to be human goods, but it may not take sides. Koppelman offers a variety of justifications for this rule, many of which sound in founding-era principles of religious liberty, including the state’s incompetence to discover religious truth, the risk of political controversy, and the risk that state support will corrupt religion.50 Koppelman argues that this rule is consistent with the American tradition of religious neutrality: as America has grown more religiously pluralistic, the government’s religious speech has likewise become more ecumenical. American pluralism now limits the government to communicating that religion in general is, or might be, a good thing.
I share many of Koppelman’s concerns about the government taking a position on religious truth, but it is not clear that the rule is either necessary or sufficient to protect religious liberty. America is certainly more religiously
The rule against taking a position on religious truth can also be difficult to apply. Consider President Reagan’s declaration of the “year of the Bible.” The declaration recognizes the Bible’s various secular merits while carefully avoiding a position on religious questions like whether the Bible reveals religious truth and whether it was divinely inspired. Consider, too, the symbols cases. Koppelman admits they are hard because they “lie precisely on the line between permitted ceremonial Deism and forbidden state endorsement of religion.”51 The reason they are hard, though, is because symbols, standing alone, do not declare religious truth. They trade on religious ideas, but vaguely so. It is unclear whether they are simply acknowledging religion’s role in society, attempting to transform a religious message, or making a point that is adjacent to a religious practice. Does a war-memorial cross represent the Christian belief that Christ’s death was a self-sacrifice with spiritual dimensions, or does it trade on the Christian tradition to make a broader point about the value of service members’ self-sacrifice? The latter takes a position on moral truth—the self-giving of service members is a valuable contribution to our society—but not about religious truth. Koppelman thinks prayers offered by legislative chaplains are plainly unconstitutional because they require officials to choose the chaplains, thus “decid[ing] disputed points of theology.”52 But legislatures and other government bodies can and do choose chaplains without opining on theology by adopting a religiously neutral procedure for people to offer invocations.53 And even if the government selected a chaplain from one faith tradition, it is unclear why that would amount to a declaration of religious truth
No proselytization. Justice Kennedy generally maintained that religious speech is problematic only when it is accompanied by coercion, but he also suggested that “[s]peech may coerce in some circumstances.” For example, he said that the Establishment Clause “forbids a city to permit the permanent erection of a large Latin cross on the roof of city hall” because “such an obtrusive year-round religious display would place the government’s weight behind an obvious effort to proselytize on behalf of a particular religion.”54 The idea seems to be that proselytization, or perhaps “an obvious effort to proselytize,” amounts to the sort of coercion the Establishment Clause prohibits. Kennedy never teased out this line of thought, but he seemed to be suggesting that even permanent symbols may affect religious liberty by directly affecting an observer’s religious beliefs and exercise.
But what would make proselytization problematic? As John Locke noted long ago, proselytization, without the application of force, is just an appeal to reason. The magistrate, he argued, has the same access to persuasion by argumentation as anyone else.55 Some might respond that the government has greater resources, more money, more ability to hold the observer’s attention than other people do. Yet large corporations, including religious organizations, have great resources too. What makes governmental proselytization, or at least the reasonable belief that the government is proselytizing, different?
The difference is that the government has a monopoly on the lawful exercise of force. The government alone can withhold public rights and privileges based on one’s response to its message. The government’s effort to proselytize, depending on the circumstances, may reasonably be understood as a threat to rights and privileges on the basis of religious beliefs and exercise. A tacit threat, unlike mere endorsement or a declaration of religious truth, could reasonably induce an observer to change his or her religious beliefs or practices to avoid discrimination.
The justices’ attempts to articulate an Establishment Clause limit on government religious speech have sought to implement many of the principles of religious liberty—separation, equality, liberty of conscience, religious pluralism—but they have ignored the overarching concern of the religion clauses: equal liberty. The historical and traditional acceptance of the vast majority of government religious speech does not mean that none of it violates the Establishment Clause, but it ought to be a clue that Americans generally have not considered it to be inconsistent with the clause’s purpose. The clause was meant to end specific sorts of laws and practices because they had the purpose and effect of inducing religious conformity. The question, then, for those who seek to integrate the positive law, the natural law norms of religious liberty, and the tradition of religious liberty in the United States is, what kind of religious speech might directly affect private religious belief and practice.
A tentative answer is lurking within the rules already on offer: mere government speech (whether in word or symbol) violates the Establishment Clause when it threatens discrimination with respect to legal rights and privileges on the basis of religion. The reason is that such a threat could reasonably induce someone to change their religious membership, belief, or exercise to avoid official discrimination. Whether the threat is understood to be coercive itself, or to rely on future coercion, what matters is that it is likely to induce religious conformity, and that is the evil, the mischief, that the Establishment Clause was meant to prohibit.
Much more needs to be done to flesh out the rule. For instance, must the government intend to threaten discrimination? Or is it sufficient for a reasonable observer to infer such a threat? What sorts of evidence ought to be necessary to prove that a communication amounts to a threat of discrimination? Must there be evidence of past discrimination, or some action in addition to the message?
For now, it is sufficient to identify several merits of the threat-of-discrimination test. First, it brings the religious-speech doctrine into line with the Court’s approach to the Establishment Clause. The Court has struggled to articulate a religious-speech doctrine. Even the justices who favor the coercion test have admitted that some forms of mere speech might be sufficient to violate the Establishment Clause, but they have struggled to explain what it is about such speech that is problematic, or how to identify it. The problem addressed by the threat test is the same as the problem addressed by the coercion test; it simply acknowledges that threats, though themselves not formally coercive, have the
Second, the threat test coheres with, and supplements, the Court’s approach to discrimination in free-exercise cases. The Court has long made it clear that official discrimination in favor of one religion violates the Establishment Clause, while official discrimination against one religion violates the Free Exercise Clause. Although the Court has spoken of these rules as reinforcing equality or neutrality, it has often missed the underlying rationale for religious equality or neutrality: discriminatory treatment affects religious choice. Restricting rights and privileges to one religious group, or to those who hold a particular religious belief, creates an incentive for outsiders to change their religious beliefs and practice. It is the enemy of religious freedom and its natural byproduct, religious pluralism.
Third, restricting the Establishment Clause limits on religious speech to speech that conveys a threat of religious discrimination bring the religious- symbols doctrine into line with the ordinary rules of standing. There is currently a gap between the religious-symbols doctrine and the ordinary standing doctrine: a claimant may challenge a government religious symbol for endorsing religion when the symbol causes the claimant no individualized or concrete injury. A claimant who is threatened with religious discrimination, however, would have such an injury. A symbol alone may rarely be enough to establish a threat, but if it were, a member of the threatened group would have standing to sue.
A threat-of-discrimination test would not eliminate all of the difficult features of the endorsement test. Cases would still require fact-intensive inquiries into the meaning of government communications, and reasonable jurists would still disagree about its application in hard cases. But any test of government speech would require such an inquiry, including whatever test the Court applied in American Legion. A threat test, however, would greatly reduce the scope of the inquiry. Rather than looking for whether a government is endorsing some relatively vague religious concept, idea, or group, judges would focus on a much sharper inquiry that plays to their strengths: has the government, or an official, threatened to engage in a specific form of conduct.
6 Conclusion
Although the threat-of-discrimination test itself does not have deep roots in the American tradition of religious liberty, it operates as a prophylactic rule
American Legion v. American Humanist Association, 139 S. Ct. 2067, 2082–83, 2089–90 (2019).
John Witte, Jr., Joel A. Nichols, and Richard W. Garnett, Religion and the American Constitutional Experiment, 5th ed. (Oxford: Oxford University Press, 2022), 301.
John Witte, Jr., “The Integration of Religious Liberty,” Michigan Law Review 90 (1992): 1363–83; and Witte, Nichols, and Garnett, American Constitutional Experiment, 1–128.
John Witte, Jr., “The Essential Rights and Liberties of Religion in the American Constitutional Experiment,” Notre Dame Law Review 71 (1996): 371–445, at 376.
John Witte, Jr., “Facts and Fictions about the History of Separation of Church and State,” Journal of Church & State 48 (2006): 15–45, at 42.
Nathan S. Chapman and Michael W. McConnell, Agreeing to Disagree: How the Establishment Clause Protects Religious Diversity and Liberty of Conscience (Oxford: Oxford University Press, 2023), 157–72.
County of Allegheny v. Greater Pittsburgh ACLU, 492 U.S. 573 (1989).
Santa Fe Indep. Sch. Dist. v. Doe, 530 U.S. 290 (2000).
Witte, Nichols, and Garnett, American Constitutional Experiment, 301.
139 S.Ct. 2067 (2019).
Michael W. McConnell, “No More (Old) Symbol Cases,” Cato Supreme Court Review (2018–19): 91–118.
139 S. Ct., 2082–83.
Ibid., 2089–90.
142 S. Ct. 2407 (2022).
Ibid., 2427.
Ibid., 2430.
Several states have similar mottos that appear on their official flags. Florida’s is “In God We Trust.” Colorado’s is “Nil sine Numine” (Nothing without the Divine). The motto of the territory of American Samoa is “Samoa, Muamaua Le” (Samoa, let Atua [God or the gods or the ancestors] be first). Kentucky’s is “Deo gratium habeamus” (Let us be grateful to God). Ohio’s is “With God, all things are possible.” South Dakota’s is “Under God the people rule.” Troy Brownfield, “The 50 State Mottos, Ranked,” Mar. 25, 2019, https://www.saturdayeveningpost.com/2019/03/the-50-state-mottos-ranked.
Ronald Reagan, “Presidential Proclamation 5018,” Feb. 3, 1983, https://reaganlibrary.gov/archives/speech/proclamations-february-3-1983.
Lujan v. Defenders of Wildlife, 504 U.S. 555, 563 (1992).
American Legion, 139 S. Ct. at 2098–2103 (Gorsuch, J., concurring in the judgment).
Harold J. Berman, “Toward an Integrative Jurisprudence: Politics, Morality, and History,” California Law Review 76 (1988): 779–801; and Jerome Hall, “From Legal Theory to Integrative Jurisprudence,” University of Cincinnati Law Review 33 (1964): 153–205.
Berman, “Integrative Jurisprudence,” 795.
Witte, “Integration,” 1363.
Witte, “Essential Rights and Liberties,” 376; and Witte, Nichols, and Garnett, American Constitutional Experiment, 59–92.
Philip Hamburger, Separation of Church and State (Cambridge, MA: Harvard University Press, 2004).
Witte, “Facts and Fictions,” 42.
Witte, Nichols, and Garnett, American Constitutional Experiment, 289–307; and John Witte, Jr. and Nina-Louisa Arold, “Lift High the Cross?: Contrasting the New European and American Cases on Religious Symbols on Government Property,” Emory International Law Review 25 (2011): 5–55.
Witte, “Fact and Fiction,” 44–45.
Chapman and McConnell, Agreeing to Disagree, 157–61; and Stephanie Barclay, Brady Earley, and Annika Boone, “Original Meaning and the Establishment Clause: A Corpus Linguistic Analysis,” Arizona Law Review 61 (2019): 505–60.
Engel v. Vitale, 370 U.S. 421, 429 (1962).
Abington Sch. Dist. v. Schempp, 374 U.S. 203, 222–23 (1963).
Stone v. Graham, 449 U.S. 39 (1980).
Marsh v. Chambers, 463 U.S. 783 (1983).
Lemon v. Kurtzman, 403 U.S. 602, 613 (1971).
Michael W. McConnell, “Establishment and Disestablishment at the Founding, Part I: Establishment of Religion,” William & Mary Law Review 44 (2003): 2105–208.
Santa Fe Indep. Sch. Dist. v. Doe, 530 U.S. 290 (2000) (prayer before football game); Wallace v. Jaffree, 472 U.S. 38 (1985) (period of silence for meditation); and Stone v. Graham, 449 U.S. 39 (1980) (Ten Commandments displays in classrooms).
Kennedy v. Bremerton, 142 S. Ct. 2407, 2427 (2022).
Kennedy, 142 S. Ct. at 2427; and Town of Greece v. Galloway, 572 U.S. 565, 576 (2014).
Marc DeGirolami, “The Traditions of American Constitutional Law,” Notre Dame Law Review 95 (2020): 1123–81.
Allegheny County v. Greater Pittsburgh ACLU, 492 U.S. 573, 597 (1989).
Ibid., 620 (Blackmun, for a plurality).
Lynch v. Donnelly, at 688 (O’Connor, J., concurring in the judgment); see also ibid., 701 (Brennan, J., dissenting) (“The effect on minority religious groups, as well as on those who may reject all religion, is to convey the message that their views are not similarly worthy of public recognition nor entitled to public support.”).
Ibid., 701n7 (Brennan, J., dissenting).
Elk Grove Unified School District et al. v. Newdow et al., 542 U.S. 1, 37 (2004) (O’Connor, J., concurring).
Lee v. Weisman, 505 U.S. 577, 641 (1992) (Scalia, J., dissenting).
McCreary County, 545 U.S. 844, 909 n.12 (2005) (Scalia, J., dissenting) (“The Establishment Clause prohibits the favoring of one religion over another.”).
Allegheny County, 492 U.S. at 594.
Andrew Koppelman, Defending American Religious Neutrality (Cambridge, MA: Harvard University Press, 2013), 6.
Ibid., 3
Ibid., 6, 46.
Ibid., 75.
Ibid., 76.
Town of Greece, 572 U.S., 1816.
Allegheny County, 492 U.S., 661 (Kennedy, J., concurring in part and dissenting in part).
John Locke, A Letter Concerning Toleration and Other Writings, ed. Mark Goldie (Indianapolis: Liberty Fund, 2010), 13, Online Library of Liberty, https://oll.libertyfund.org/title/goldie-a-letter-concerning-toleration-and-other-writings.