Defining Religion in the Court

Here is a snapshot that captures an increasingly important issue in law and religion in the United States: In August 2021, four parents sued a school district outside Philadelphia for violating the free exercise rights of their children. In response to the Covid-19 pandemic, the district had required all students to wear masks and refused to allow religious exemptions. The parents—from four separate households—argued that covering the face violated their religions, and that the district’s requirement therefore violated the First Amendment.

In one sense, Geerlings v. Tredyffrin/Easttown School District was nothing new. Several lawsuits across the country had claimed, with varying degrees of success, that anti-Covid measures violated the Free Exercise Clause. But the Pennsylvania lawsuit was atypical in one important respect. Unlike the plaintiffs in most cases, who based their objections to anti-Covid measures in the teachings and practices of traditional religious communities—churches and synagogues—the Geerlings parents raised objections that were rather idiosyncratic.

Two of the parents belonged to no religion but described themselves as “spiritual.” Of these two, one believed in “something else out there,” not “just us,” which convinced him that his daughter should not wear a mask; the other maintained that masks mocked “the gift of life.” Two parents were Christians with unusual beliefs. The first, a deacon in a local Presbyterian church that had itself required masks during the pandemic, stated that, in her view, masks dishonored God. The other, an Episcopalian whose parish likewise had required masks during the pandemic, maintained that masks violated the “temple” of the body. Both of these plaintiffs apparently stopped attending services when their churches began to require masks.

With some hesitation, the court concluded that none of these claims qualified as “religious” for First Amendment purposes. Even if the parents’ objections were sincere, Judge Mitchell Goldberg wrote, they were too idiosyncratic to merit constitutional protection. Goldberg conceded that a claimant need not belong to an organized religion or hold the same beliefs as others in order to raise a free exercise claim. But there were limits. The First Amendment did not require exemptions for plaintiffs who objected to legal requirements on the basis of personal intuitions that failed to correspond with the teachings of any religion.

Judge Goldberg’s hesitation was understandable. The Supreme Court has never settled on a definition of religion, and its decisions down the centuries point in different directions. Some of the Court’s rulings indicate that idiosyncratic personal convictions can qualify as religious; others suggest the opposite. Until recently, the question has been mostly academic. Although solitary seekers like Henry ­David Thoreau have always been with us, most Americans have tended to identify with specific religious traditions. Few have sought legal exemptions based on purely personal spiritual commitments.

That seems likely to change. A fast-growing percentage of Americans, the so-called “Nones,” now tell pollsters that they have no religious ­affiliation—roughly 30 percent of the population, up from only 6 percent a generation ago. Like the first two Geerlings plaintiffs, most Nones are what one might call unaffiliated believers, people who reject organized religion but believe in “something else out there” and follow their own spiritual paths. Moreover, many Americans retain formal religious affiliations but combine orthodox teachings with other commitments, like the Christian parents in Geerlings who objected to face coverings on ­spiritual grounds. Tara Isabella Burton refers to this group as “religious hybrids.” Together with unaffiliated believers, she estimates, they may amount to more than 50 percent of the American population.

Geerlings is not the only recent case in which unaffiliated believers and religious hybrids have sought religious exemptions—and it will not be the last. Simply put, there are a lot more Henry ­David Thoreaus than there used to be, and some of them seem quite litigious. The time is ripe for the Supreme Court to clarify the meaning of “religion” for free exercise purposes—specifically, to clarify whether it covers purely individualistic spiritual pursuits.

Of course, many American constitutional scholars argue that religion is too amorphous a concept to be defined at all, or that privileging religion at the expense of other deep personal commitments—making religion “special,” for constitutional ­purposes—is immoral. I lack space to address those arguments here. Suffice it to say that the Free Exercise Clause itself speaks of “religion,” not of “deep personal commitments,” and the Supreme Court has made clear that it continues to see religion as a distinct category meriting special constitutional protection. Avoiding a definition of religion for First Amendment purposes is not sustainable, and recent changes in American religion make the issue only more salient.

Little evidence exists of what the Framers meant by the word “religion” in the Free Exercise Clause. In 1791, when the Clause was adopted, “religion” could refer to conventional faith communities, especially Protestant ones, but also to non-institutional beliefs such as Deism. Nonetheless, the drafting history suggests that the Framers intended the Clause to cover communal rather than purely idiosyncratic phenomena. The Framers famously chose to protect the “free exercise of religion” rather than the “rights of conscience.” Although the Framers did not explain why they preferred the former phrase, “religion” connotes a community of believers, as Michael McConnell has observed, whereas “conscience” suggests individual judgment. The fact that the Framers chose a word with communal connotations over the more personal “conscience” suggests they had collective phenomena in mind.

Down the centuries, the Court has failed to offer consistent guidance. The handful of decisions that address the definition of religion point in different directions. In a couple of nineteenth-­century cases involving the LDS Church, the Court indicated that religion entailed a system of beliefs consistent with traditional Christianity. “Religion,” the Court said in Davis v. Beason (1890), “has reference to one’s views of his relations to his Creator, and to the obligations they impose of reverence for his being and character, and of obedience to his will.” The Court scoffed at the idea that the LDS Church could be a religion for constitutional purposes—notwithstanding that it held a belief in God—because it advocated polygamy, a practice condemned “by the general consent of the Christian world.” To call advocacy of polygamy “a tenet of religion,” the Court stated, would “offend the common sense of mankind.”

The Court has never repudiated these cases (though it would surely consider the LDS Church a religion today), yet subsequent decisions cast doubt on the idea that religion denotes traditional, organized religion, much less traditional Christianity. In two conscientious-objector cases from the Vietnam Era, the Court construed statutory language referring to “religious belief” to cover personal, non-theistic convictions. In United States v. Seeger (1965), the Court read language in the Draft Act exempting persons who objected to war on the basis of “religious belief” as covering a “belief in and devotion to goodness and virtue for their own sakes.” The test, the Court held, was “whether a given belief that is sincere and meaningful occupies a place in the life of its possessor parallel to that filled by the orthodox belief in God.”

Seven years later, the Court reverted to a more conventional definition in Wisconsin v. Yoder (1972), a case involving the Amish. ­Unlike the Draft Act cases, which collapsed the distinction between the communal and the personal, ­Yoder insisted on a bright line. “Religion,” the Court indicated, denotes a commitment to a traditional faith community; merely personal convictions would not qualify. The Court offered Thoreau as a counterexample. Thoreau, the Court explained in dicta, had engaged in an “isolated” and “­subjective” quest—an alternative lifestyle all his own. By contrast, the Amish had a three-­hundred-year history as a tight-knit religious community; they lived according to strict and detailed rules enforced by a formal church structure. The Amish exercised a religion for constitutional purposes. Thoreau had not.

Nonetheless, in two cases following Yoder, the Court suggested that religion could indeed be a personal phenomenon—in some ­circumstances and within limits. In Thomas v. Review Board (1981), the Court held that a Jehovah’s Witness who worked at a steel factory could claim a religious exemption from making weapons, even though Jehovah’s ­Witnesses apparently did not consider making weapons to be a sin and another Jehovah’s Witness at the same factory did not object to making weapons. The Free Exercise Clause does not require that religious beliefs be “acceptable, logical, consistent, or comprehensible to others,” the Court explained, or “shared by all of the members of a religious sect.”

Several years later, in Frazee v. Illinois Department of Employment Security (1989), the Court ruled unanimously that the Free Exercise Clause protected a claimant who argued that he could not work on Sundays because he was a Christian—even though he did not belong to a church. “[M]embership in an organized religious denomination, especially one with a specific tenet forbidding . . . work on Sunday, would simplify the problem of identifying sincerely held religious beliefs,” the Court explained. But the Clause did not require a claimant to show that he was “responding to the commands of a particular religious organization.”

Thomas and Frazee are in tension with Yoder, but neither case disavows that earlier ruling, and neither suggests that religion can be an entirely idiosyncratic matter. The Thomas Court explained that some claims might be “so bizarre, so ­clearly non­religious in motivation,” as to fall outside the protection of the Free Exercise Clause. ­Thomas itself did not involve such a claim. The debate among ­Jehovah’s Witnesses about whether one could permissibly work on weapons was a typical “intrafaith” dispute, an apparently good-faith disagreement of the sort “not uncommon among followers of a particular creed.” Similarly, in Frazee, a person who considered himself a Christian claimed that he could not work on Sundays. Though not all Christians shared it, his Sabbatarianism was hardly a “bizarre or incredible” conviction in the context of Christianity. A more singular claim, the Court implied, one that departed more seriously from a religion’s conventional baseline, would receive different treatment.

As this quick survey demonstrates, the Court’s decisions on the definition of religion are muddled. According to the cases, religion entails a conventional belief in God—except when it does not. Religion entails a commitment to a traditional, organized faith community—except when it does not. Religion excludes purely individualistic spiritual convictions—except when it does not. And a court should not evaluate whether a particular belief is “bizarre” or is shared by others in the claimant’s religion—except when it should.

Until very recently, one could dismiss the tension between communal and individualistic religion as peripheral. Notwithstanding occasional litigation like the Draft Act cases, few Americans have ever sought exemptions from civil law on the basis of entirely personal spiritual commitments. Courts could avoid deciding whether the Free Exercise Clause protected idiosyncratic beliefs and practices. The issue almost never came up.

But the rise in unaffiliated believers and religious hybrids has begun to change things. A recent study by scholars Michael Heise and Gregory Sisk reveals that Nones brought about 6 percent of the religious-accommodations claims in federal court from the years 2006 to 2015. About 40 percent of those claims were “loosely religious or spiritual in nature but not associated with a recognized religious grouping.” The numbers are still small, but they will probably increase as unaffiliated believers and religious hybrids become more entrenched in American society. One could imagine all sorts of claims based on idiosyncratic spiritual commitments: a vegetarian diet in prison, for example, or the right to wear certain clothing or insignia in the military. Or, as in Geerlings, an exemption from public health laws.

It’s time for the Court to establish that religion, at its core, denotes communal beliefs and practices rather than idiosyncratic personal commitments. A communal definition makes sense for several reasons. First, the existence of a community captures something important about the social reality of religion. In lived experience, religion suggests a group of people linked with one another, through time, in worship. As sociologist Christian Smith writes, “religions are almost invariably social activities—communities of memory engaged in carrying on particular traditions.” Without a communal structure to give them meaning, religious practices such as prayer, fasting, and so on are incoherent, “simply the strange doings of odd people.”

Second, a focus on community accords with an important goal of religious freedom: the promotion of private associations that encourage cooperative projects and check state power. As Tocqueville explained, the despotic state desires nothing more than for individual citizens to feel isolated from and indifferent to others, so that it can divide and dominate them all. By encouraging people to identify with and look out for one another, private associations militate against self-centeredness and social isolation and help keep the state in check. Religious groups perform this function especially well. No associations have been better, historically, at promoting cooperative social projects and defying state oppression—as dictators down the centuries have learned.

Third, the existence of a religious community reduces the possibility of fraud. Everyone agrees that courts need not honor a religious claim that a litigant does not genuinely hold or raises merely as a pretext. But sincerity, which depends on a claimant’s subjective state of mind, is notoriously difficult for courts to evaluate. The existence of a religious community to which the claimant belongs can provide objective evidence of the claimant’s good faith. A continuous tradition of teaching and worship, and an organized body that enforces discipline, can go a long way toward demonstrating the claimant’s genuineness about his religious convictions.

Fourth, the existence of a community helps ensure that religious commitments are not frivolous and fleeting, but serious and lasting. It is one thing for the state to accommodate a citizen’s profoundly held beliefs. To do so honors the citizen’s dignity and accords him respect. It is another thing to defer to commitments that may be temporary and superficial. Not all idiosyncratic commitments fit that description, of course, and beliefs can be religious even if novel. But the existence of an established religious community can screen out passing whims that the state need not honor.

Finally, making the definition of religion turn at least in part on the existence of a religious community reduces the potential for administrative disorder. Long ago, the Supreme Court warned that if personal spiritual convictions were sufficient to override legal obligations, “every citizen [would] become a law unto himself.” One should not overstate this concern, but the rise of the Nones makes it important today. Requiring a claimant to show that his or her objections are not merely personal but derive from the teachings of an organized body of believers can reduce the potential for conflict with state laws—especially in a society in which very large numbers claim to follow their own ­spiritual paths.

In short, the existence of a community is crucial to a plausible definition of religion. Nonetheless, categorically limiting legal protection to collective beliefs and practices would create two serious problems. First, excluding non-­institutional, idiosyncratic beliefs would contradict a long American tradition of honoring individual religious conscience. Unaffiliated believers like Thoreau and his spiritual descendants reflect something important about the way religion has historically been understood in America and the way it is understood today. Moreover, notwithstanding the problems of alienation, anomie, and unserious claims, honoring individual religious conscience promotes the important values of autonomy, dignity, and personal well-being. Categorically declining to protect individualistic claims would scant those important values and depart too far from our long tradition.

Second, limiting religion exclusively to communal beliefs and practices would create difficult line-drawing problems. Of course, civil courts lack competence to determine which religious expressions are correct in a metaphysical sense; that question is not at issue. For these purposes, courts would identify religion in an empirical sense—­describing things as they are. Even so, in many cases, courts might have a hard time identifying whether a community exists and what it teaches about a specific question. Most religions, ­especially if they have existed for a long time, have both majority and minority strains. Determining which expression among many qualifies as the “true” representative of a religious tradition can be difficult and may entangle courts in internal theological squabbles. These line-drawing problems will not always pose insurmountable obstacles. Law frequently relies on generalizations, and in many situations, courts should be able to fairly identify a religious community and its teachings as empirical matters. But the problems are significant enough to give one pause about categorically limiting free exercise protection to religious communities.

Thus, the proper constitutional definition of religion would be a flexible one with community at its core. The more a plaintiff can tie his beliefs and practices to those of an established religious community, the more his claims should qualify as religious for constitutional purposes. By contrast, the further a plaintiff diverges from a religious community, the more idiosyncratic his spiritual path, the less plausible his claim to practice a religion becomes.

Thomas and Frazee are suggestive in this regard. Thomas, recall, concerned what the Court characterized as a genuine “intrafaith” dispute among Jehovah’s Witnesses about the permissibility of making weapons. In the context of the religious tradition to which he belonged, Thomas’s objections were not novel and unique to him, but a matter of good-faith disagreement with his fellow believers. The Court was correct in recognizing his objections as religious.

Similarly, Frazee involved a claim by a person who called himself a Christian and maintained that he could not in good conscience work on Sundays. In the context of Christianity, this was hardly a “­bizarre” position, even if most American Christians no longer shared it, and even if the claimant did not formally belong to a church. Purely as a descriptive matter, refraining from work on Sundays is not an idiosyncratic position for someone who calls himself a Christian; it suggests neither fraud nor a lack of seriousness. Once again, given the faith tradition to which he claimed to subscribe, Frazee’s objections were properly characterized as religious.

The claims in Thomas and Frazee had plausible links to religious communities and did not implicate concerns about hyper-individualism, fraud, and lack of seriousness. The Court correctly concluded that the claims were religious. Contrast those claims with the claims in Geerlings, the recent Pennsylvania case involving the Covid mask mandate. The Christian plaintiffs in that case argued that medical masks dishonored God and violated the temple of the body. In the context of the Christian faiths to which the plaintiffs claimed to belong, and speaking only descriptively, those claims do indeed qualify as bizarre. They do not reflect genuine intrafaith disputes and do not cohere with historic beliefs and practices. Judge Goldberg was correctly skeptical that the claims qualified as religious.

Or consider the claim of the unaffiliated believers in Geerlings who sought religious exemptions because of personal views that masks mocked “the gift of life” and intuitions about a vague supernatural force beyond “just us.” Courts should be especially chary of such claims, which lack even a purported connection to a religious tradition. To treat such claims as religious for First Amendment purposes would contravene the common understanding that religion is a communal phenomenon and fail to advance the important associational benefits organized religion can confer. It would increase the potential for fraudulent and trivial claims and administrative disorder, in a country with millions and millions of unaffiliated believers, each of whom could make similar claims about the requirements of “something else out there.”

To be sure, the flexible approach I advocate here would not settle questions in the manner of a categorical rule. It would consider specific facts and depend a great deal on judgment; in any particular case, whether a claim qualifies as religious may be genuinely uncertain. Line-drawing problems will remain. Nonetheless, by placing the existence of a religious community at the core of the definition of religion, this approach would offer the benefits of tying religion to common understandings and avoiding at least some of the problems associated with defining religion in idiosyncratic terms, while remaining true to our cultural and legal traditions and minimizing the difficulties that a more categorical ­approach would entail.

Mark Movsesian is the Frederick A. Whitney Professor of Contract Law and Co-Director of the Center for Law and Religion at St. John’s University. A longer version of this essay, “The New Thoreaus,” appears in a forthcoming issue of the Loyola University Chicago Law Journal.

Image by Alexandra Koch licensed via Creative Commons. Image cropped.