The Asymmetry of Religious Motivation

abstract. The Supreme Court’s current approach to religious freedom reflects an asymmetric view of religious motivation. Under the Free Exercise Clause, the Court has increasingly embraced a motivational sufficiency principle, according to which a person’s sincere religious motivation is sufficient to render an act religious and therefore eligible for accommodation. Under the Establishment Clause, by contrast, the Court has moved toward a principle of motivational irrelevance: the religious motivations of public officials are treated as immaterial in determining whether governmental action impermissibly establishes religion. These two principles rest on conflicting conceptions of the constitutional significance of religious motivation—favoring it in the context of granting exemptions but disregarding it when the government acts or speaks in religiously charged ways.

This Essay reveals this doctrinal asymmetry through an analysis of Catholic Charities Bureau v. Wisconsin Labor & Industry Review Commission, in which the Court invalidated a state’s reliance on religious criteria other than motivation to determine eligibility for a statutory tax exemption. Although the Court did not explicitly adopt a motivational sufficiency principle, its reasoning is difficult to explain without appeal to that idea. By contrast, in Establishment Clause cases involving government speech and coercion, the Court has abandoned tests that made motivation relevant to constitutional analysis. The resulting asymmetry—recognizing religious motivation when doing so expands religious freedom but ignoring it for purposes of constraining governmental power—contributes to a jurisprudence of religious preference that is likely to persist despite its doctrinal incoherence.

Introduction

The Supreme Court’s current approach to religious freedom under the First Amendment adopts an asymmetric view of religious motivation. On the one hand, with respect to granting religious exemptions, the Court seems to embrace a principle of motivational sufficiency. According to this principle, when people sincerely believe they are engaged in conduct for religious reasons, that fact is sufficient to prompt a legal inquiry into whether the law burdens them and, if it does, whether there are grounds to grant a religious exemption.1 But on the other hand, with respect to limits on government support for religion under the Establishment Clause, the Court appears to reject the idea that religious motivation matters in determining whether conduct by public officials is impermissibly religious. This latter approach might be explained by an underlying principle of motivational irrelevance, which holds that religious motivation is never relevant in limiting government coercion or expression. In other words, to show a violation under the Establishment Clause, it is not enough to claim that public officials acted with religious motivations; the law or policy in question must have some other objective religious content.

But these two principles—motivational sufficiency (which applies to claims for exemptions under the Free Exercise Clause) and motivational irrelevance (which applies to limits on government support for religion under the Establishment Clause)—take contradictory views about religious motivation. In developing this claim, I will focus on the Supreme Court’s recent decision in Catholic Charities Bureau, Inc. v. Wisconsin Labor & Industry Review Commission.2 This case addressed whether Catholic Charities, a nonprofit organization controlled by a Roman Catholic diocese, was entitled to a religious exemption from paying taxes into Wisconsin’s unemployment compensation system under a state statutory provision for church-owned nonprofits that were “operated primarily for religious purposes.”3 The Wisconsin Supreme Court denied an exemption to Catholic Charities.4 Even though the organization was religiously motivated, that was insufficient to show that it was “operated primarily for religious purposes.”5 To have a religious purpose, the Wisconsin court held, an organization had to engage in religious activities, such as worship, proselytizing, or religious education.6 On this interpretation, religious motivation was not sufficient to trigger an exemption claim under the Wisconsin statute. There had to be religious motivation plus something more—some evidence of religious activity in the work or service performed by the organization.7

In Catholic Charities, however, the Supreme Court disagreed.8 It held that defining “religious purposes” to require certain religious activities favored some religious denominations over others, in violation of the Establishment Clause.9 Although the Court did not explicitly embrace a principle of motivational sufficiency—in which religious motivation without more is sufficient to support a claim for exemption—it also did not explain how any other approach could be made consistent with its holding.10 If courts cannot consider other markers of whether conduct is religious, such as whether an activity involves worship, proselytizing, or religious education, then there may be no other religious indicators besides religious motivation that the government can rely on to limit the scope of exemptions.11 And indeed, after Catholic Charities, a motivational test might be the only general trigger for exemptions that will satisfy the Court’s understanding of the First Amendment.12 Any attempt to constrain exemptions, even when they are granted as a matter of legislative discretion, cannot turn on nonmotivational aspects of religion, at least when those are contested between different religious denominations. As a criterion for determining eligibility for accommodations, religious motivation is the only safe harbor.

Now compare the role of religious motivation when the government engages in expression or when it regulates private conduct, either by compelling or forbidding it. To take a current example, consider state prohibitions on abortion.13 Does it matter whether public officials are religiously motivated to enact a ban? More than forty years ago, in Harris v. McRae, the Court rejected an Establishment Clause challenge to restrictions on public funding of abortions, reasoning that even if the restrictions were justified on religious grounds, that would be irrelevant to their constitutionality.14 More recently, the Court has cast doubt on the idea that the Establishment Clause requires that laws and policies be justified by a secular purpose. In Kennedy v. Bremerton School District, the Court declared that it had “long ago abandoned”15 its test in Lemon v. Kurtzman,16 which held, in part, that laws must have a “secular legislative purpose.”17 If Lemon is no longer good law, then there may be no constitutional requirement that laws have a nonreligious motivation. In other words, to violate the Establishment Clause, a law’s religious motivation would not be sufficient to make its demands religious. There would have to be some other outward indication or manifestation of religion. Accordingly, banning abortion solely for religious reasons would not be religious conduct, because an abortion ban conceivably could be justified on other nonreligious grounds. But a policy to compel some religious activity—for example, prayer in public school—does count as religious coercion, regardless of whether it is motivated on religious grounds. Under what seems to be the Court’s emerging understanding of the Establishment Clause, motivations are irrelevant. When the government speaks or when it regulates conduct, its decisions will be forbidden only when they entail nonmotivational religious content.

Under the Religion Clauses, then, religious motivation is sufficient to convert otherwise ordinary or secular actions into protected religious conduct for purposes of granting exemptions. But religious motivation is not sufficient to convert otherwise ordinary or secular government actions into impermissible religious conduct for purposes of disestablishment. To continue with the abortion example, an organization that objects for religious reasons to funding abortions engages in religious activity, but a government that prohibits abortions for religious reasons does not.

Here, then, is both a normative and doctrinal question: what, if anything, justifies this asymmetry in the role of religious motivation? In other words, why is religious motivation sufficient to characterize conduct as religious with respect to free exercise, but not with respect to the disestablishment of religion? My claim is that there is no good answer to this question. If religious motivation matters—if it changes the meaning of an action—then it does so regardless of whether the relevant agents are private or public actors. Assuming that the Court is committed to the principle of motivational sufficiency, which seems especially plausible after Catholic Charities, a principled jurisprudence would have to recognize that laws and policies that are motivated on religious grounds are, for that reason, impositions of religion. The Court might nevertheless hold that such impositions are permissible under the Establishment Clause, but it could not describe religiously motivated government action as anything other than religious conduct.

To develop these claims, this Essay proceeds as follows: Part I elicits the emerging principle of motivational sufficiency from the Court’s holding in Catholic Charities, which is consistent with its deferential approach to exemptions under the Free Exercise Clause. Part II introduces the principle of motivational irrelevance and develops the doctrinal asymmetry between the two principles, arguing that they are in contradiction. Part III anticipates and responds to several objections to the claim that doctrinal consistency requires recognizing the relevance of religious motivation in the exercise of governmental power. A principled account of religious motivation would make it relevant for both free exercise and disestablishment. This would entail shoring up a secular purpose requirement to guard against religious coercion and to protect constitutional rights against reliance on religious purposes that might be asserted to undermine them. Yet, given its recent decisions under the Religion Clauses, the Court is more likely to persist in holding an asymmetric view that accounts for motivation in benefiting religion but not in placing constitutional limits upon it—an approach that contributes to a jurisprudence of religious preference.18

I. the sufficiency of religious motivation

Catholic Charities was something of a sleeper in a Supreme Court term that featured two other important religious freedom cases: Mahmoud v. Taylor involved a free exercise claim for opt-outs from participation in LGBTQ-related readings in public school curricula,19 and Oklahoma Statewide Charter School Board v. Drummond was a potential blockbuster that fizzled when the Court deadlocked, four-to-four, on whether to require that state charter-school programs include religious schools.20 Compared to those cases, Catholic Charities might have seemed less flashy or important. It did not implicate major culture war issues or threaten to disrupt longstanding rules governing public education. The case raised a relatively arcane question about state unemployment tax systems, which had not received much attention or been heavily litigated. Moreover, the Supreme Court answered that question unanimously,21 without the partisan rancor often seen in Religion Clause cases. And yet, in this Part, I argue that the Court’s decision in Catholic Charities has broad significance, in part for what it implies, or perhaps entails, about the role of religious motivation in the Court’s approach to exemptions. More than any other recent case, Catholic Charities demonstrates the gravitational force of a principle of motivational sufficiency, the claim that religious motivation—and nothing more—establishes eligibility for religious accommodation.

A. The Motivational Question in Catholic Charities

As presented to the Supreme Court, Catholic Charities focused centrally on the role of religious motivation in granting exemptions. Like most states, Wisconsin requires nonprofit organizations to pay taxes into an unemployment compensation system, with the aim of protecting employees from the “economic burdens resulting from unemployment.”22 But in language tracking a religious exemption provided in the Federal Unemployment Tax Act (FUTA),23 Wisconsin law exempts nonprofit organizations if they are “operated primarily for religious purposes” and “operated, supervised, controlled, or principally supported by a church.”24

In proceedings below, Catholic Charities had claimed that under a plain reading of this provision, it qualified for an exemption.25 As the “social ministry arm” of the Roman Catholic Diocese of Superior, Wisconsin, Catholic Charities had no difficulty in establishing that it was controlled by a church. This fact was undisputed.26 But Catholic Charities also claimed that it and four of its subentities,27 which provided various types of social services, were “operated primarily for religious purposes” because “the Diocese of Superior’s motivation is primarily religious, i.e., their charitable works are carried out to operationalize Catholic principles.”28

Rejecting this claim, the Wisconsin Supreme Court held that Catholic Charities erred by interpreting the exemption provision to refer solely to the organization’s “purposes” in terms of its motivation or intent.29 On the court’s reading, the statutory text required that “both activities and motivation must be considered.”30 The term “operated” meant looking not only at motivation, but also at the organization’s activities or functions—not only at “why the organization acts,”31 but also at how it does so.32 The court reasoned that because the exemption provision covered entities controlled by a church, those entities would always be operated by an organization with religious motivation. Interpreting the provision solely in terms of motivation would thus render it superfluous.33 A motivational reading would also greatly expand the scope of the exemption, which the court held was contrary to the “narrow construction” given to exemptions that would undermine the legislative purpose of protecting workers from the harms of unemployment.34

Having interpreted the statutory exemption to require both religious motivation and religious activities, the Wisconsin Supreme Court then faced the problem of identifying which activities count as demonstrating a religious purpose. Without claiming to offer an “exhaustive” list or “necessary conditions,” it pointed to several “hallmarks of a religious purpose,” including “whether an organization participated in worship services, religious outreach, ceremony, or religious education.”35 Even if an organization was religiously motivated, it would not be eligible for exemption if its activities were “secular in nature” rather than religious.36

When the Wisconsin Supreme Court applied this approach to Catholic Charities and its subentities, the court found that, although the nonprofits were religiously motivated, their activities were “primarily charitable and secular.”37 Neither Catholic Charities nor its subentities “infused” their services “with teaching, evangelism, and worship,”38 nor did they limit participants in their programs to members of the faith.39 The court also noted that before coming under the umbrella of Catholic Charities, the subentities had not been religiously affiliated. The services they continued to offer “would not differ in any sense” if they were “provided by organizations with either religious or secular motivations,”40 which the court took to be a “strong indication” that the organizations were not “operated primarily for religious purposes.”41

In dissent, Justice Rebecca Grassl Bradley blasted the majority’s approach as “discriminatory”42 and “incoherent.”43 The crux of her reasoning was that religious motivation transforms the meaning of otherwise secular conduct. As she put it, “[R]eligious activities cannot be separate from religious purposes. It is the underlying religious motivation that makes an activity religious.”44 Free exercise case law provides numerous examples, and Justice Bradley easily rehearsed them: “[A]nyone—religious or irreligious—could use peyote, kill animals, grow a 1/2-inch beard, or use Saturday as a day of rest.”45 According to the dissent, the majority’s approach was incoherent because it failed to recognize that these activities are not “inherently religious,” but rather become religious when they are done for religious reasons.46

Justice Bradley expressed the core of a motivational principle when she wrote that “[s]uch activities are religious activities only if motivated by religious beliefs.”47 In other words, motivation is necessary to turn an otherwise secular activity into a religious one. And, further, motivation is sufficient, in the sense that an activity need not satisfy any other criteria to count as religious. Indeed, in the dissent’s view, adding any further criteria for determining what counts as religious conduct invites discrimination based on judicial attitudes about which practices are thought to be “religious in nature.”48 The activities emphasized by the majority—worship, proselytizing, and religious education—are no more religious than other religiously motivated activities, including the provision of charitable services.49 For the dissent, in determining whether an activity is religious for purposes of claiming an exemption, motivation—and only motivation—is what matters.

In its appeal from the Wisconsin Supreme Court, Catholic Charities kept the issue of religious motivation front and center. As in the courts below, it argued that by inquiring into whether its activities were religious in nature and not only religiously motivated, Wisconsin violated the Religion Clauses in several ways. First, Catholic Charities claimed that refusal to grant its exemption conflicted with the “principle of church autonomy.”50 The argument here was that if Catholic Charities had merged with the Diocese and been part of its corporate structure, rather than separately incorporated, it would have been covered under a different provision of Wisconsin’s law that exempted churches.51 By granting exemptions to churches but not to their charitable affiliates, the state pressured the Roman Catholic Church to alter its structure of governance, which is protected from interference under church autonomy doctrine.52 Second, Catholic Charities argued that in attempting to determine whether its activities were religious, the Wisconsin Supreme Court impermissibly entangled itself in religious affairs, contravening precedents under both Religion Clauses.53 Third, and finally, Catholic Charities asserted that Wisconsin’s approach discriminated among religious denominations by preferring nonprofits that engaged in “‘typical’ religious activities,” such as worship, proselytizing, and religious education, while groups that avoided engaging in those activities were penalized, even if they had religious reasons for doing so, as Catholic Charities had claimed.54

These First Amendment arguments raised the stakes of the case. Catholic Charities moved from the realm of state statutory interpretation into federal constitutional litigation, with potential significance for dozens of states and for the interpretation of FUTA, which contains the same language exempting organizations “operated primarily for religious purposes.”55 But what should not be lost is that all this statutory and constitutional argument by Catholic Charities was aimed at one thing: guaranteeing that the religious exemption at issue tracked religious motivation and nothing more.

B. From Motivation to Denominational Neutrality

The Supreme Court’s decision in Catholic Charities, which held that the Wisconsin Supreme Court had violated the Establishment Clause of the First Amendment, seemed to shift focus away from religious motivation and toward the “principle of denominational neutrality.”56 That principle prohibits the government from discriminating among religious denominations.57 Applying this principle, the Justices reached unanimity in determining that Wisconsin had discriminated against Catholicism and in favor of other religious denominations.58 But a careful analysis shows that this holding reinforces the role of religious motivation by preventing governmental reliance on other criteria—such as whether conduct involves worship, religious education, or proselytization—for purposes of granting exemptions.

The holding in Catholic Charities was relatively narrow and closely tied to the facts of the case. Writing for the Court, Justice Sotomayor announced that the bar on denominational discrimination was “[t]he clearest command of the Establishment Clause.”59 Here, she relied on Larson v. Valente,60 a precedent that the Court had not applied to require an exemption in more than forty years.61 Larson involved an Establishment Clause challenge to a state law granting a religious exemption from registration and reporting requirements for nonprofit organizations engaged in charitable solicitation.62 Religious groups that received more than half of their contributions from members were exempt, while groups that raised the majority of their funds from solicitations were subject to regulation.63 The Court held that this “fifty percent rule”64 facially discriminated among religious groups and had a disparate impact that favored “well-established churches” and disadvantaged those that were “new and lacking a constituency” that would have lessened their need for outside solicitations.65 The prohibition on this type of “denominational preference” under the Establishment Clause, which the Court linked to the “continuing vitality of the Free Exercise Clause,”66 was held to be “absolute.”67

Applying the principle of denominational neutrality from Larson, Justice Sotomayor held that the Wisconsin Supreme Court had similarly discriminated on denominational grounds.68 It did so by determining that Catholic Charities was not qualified for an exemption unless it engaged in religious activities—such as worship, proselytization, or religious education—or limited its services to members of the faith.69 Sotomayor reasoned that these criteria were facially religious, that is, they “explicitly differentiat[ed] between religions based on theological practices,”70 to the disadvantage of Catholic Charities and other religious organizations that do not infuse their charitable services with religious content or restrict whom they employ or serve based on religious affiliation.71

In reaching this conclusion, Justice Sotomayor emphasized that Catholic Charities was guided by Catholic teaching, which prohibits proselytization and discrimination based on “race, sex, or religion” in the provision of charitable services.72 In other words, Catholic Charities was not only religiously motivated to provide charitable services—which the Wisconsin court had acknowledged73—but it was also religiously motivated to provide those services without performing the types of religious activities that the Wisconsin court had identified as necessary to qualify for an exemption.74 If Catholic Charities and its subentities had engaged in those activities and not been religiously motivated to do otherwise, they would have received the exemption. In drawing a distinction between religious organizations that turned on their “theological choices,” Wisconsin had “impose[d] a denominational preference.”75

Having found a violation of the principle of denominational neutrality, the Supreme Court then applied strict scrutiny to Wisconsin’s interpretation of its religious exemption.76 Wisconsin asserted two justifications for considering the religious activities of charitable service providers. The state sought, first, to guarantee unemployment benefits for workers and, second, to avoid entanglement with religiously informed employment decisions.77 The Court held that Wisconsin’s approach was overinclusive with respect to its first interest, because there was no record evidence that Catholic Charities would fail to provide benefits for its employees, and also underinclusive, because Wisconsin already “exempted over 40 forms of ‘employment’ from its unemployment compensation program.”78 And Wisconsin’s reliance on religious activities was not narrowly tailored to its anti-entanglement interest, because the Court found that the state could have limited its exemption to employees with ministerial responsibilities, rather than excluding entire organizations.79 The Court thus concluded that Wisconsin’s policy of inquiring into the religious nature of an organization’s charitable activities was unjustified. It reversed the decision of the Wisconsin Supreme Court and remanded for proceedings consistent with its constitutional analysis.80

Formally, the Court’s holding in Catholic Charities appears narrow and limited to rejecting Wisconsin’s decision to rely on “theological choices” about how religious charitable organizations provide their services.81 The Court told Wisconsin what it could not do, which is to require that religious organizations engage in outwardly religious activities, such as proselytizing or limiting their services to members of their communities. But aside from this constraint, the Court did not provide further guidance about how Wisconsin could interpret its exemption provision, leaving open several possibilities. One option is that anything other than a motivational approach would run afoul of the Court’s principle of denominational neutrality. Another is that the state could define its exemption in terms that do not turn on religious motivation or on the religious nature of the services provided by religious nonprofits, but rather on some other feature of their work. If this second option (or some version of it) were viable, then Catholic Charities might face difficulties on remand. But the gravitational pull of religious motivation in the Court’s decision likely undermines that possibility.82

C. Why Motivational Sufficiency Is Required

Any approach to eligibility for religious exemptions that relies on religious considerations other than motivation faces serious and perhaps insurmountable difficulties after Catholic Charities. These problems are apparent in Justice Jackson’s functional interpretation of the exemption provision in both the Wisconsin statute and in FUTA, from which Wisconsin borrowed its language.83 Jackson concurred because she agreed with the Court’s holding that Wisconsin had discriminated against Catholic Charities on religious grounds.84 But she argued that another approach was available to Wisconsin and to interpreters of FUTA. Rather than understand “religious purposes” in terms of an organization’s motivation or the manner in which it carries out its activities, in her view, courts should focus on the types of activities that an organization performs.85 According to Jackson, it was a mistake to read FUTA’s text as requiring a “motivations-only interpretation” into “why” organizations provide services.86 And it was a mistake to inquire into “how” they provide those services, asking, as the Wisconsin court had, about whether their activities were infused with religious content.87 Instead, Jackson argued that FUTA’s religious-purposes provision is better interpreted as focused on “what the entity does”—on whether entities “exist to perform religious functions.”88

In support of this interpretation of FUTA’s religious exemption, Justice Jackson pointed, as the Wisconsin Supreme Court had,89 to legislative history showing that Congress had not intended to exempt various religiously motivated entities. The House and Senate Reports on FUTA indicated that “a church related (separately incorporated) charitable organization (such as, for example, an orphanage or a home for the aged) would not be considered . . . operated primarily for religious purposes.”90 By contrast, organizations dedicated to religious functions, such as a college devoted to training ministers, “a novitiate,” or a “house of study training candidates to become members of religious orders” would be eligible for exemption.91 On Jackson’s reading, the distinction between these entities was based on what services they performed and not on why or how they performed them.92

A functional approach to FUTA’s exemption language would treat similarly situated religious and secular charitable entities alike. To borrow a well-known example from Christopher Eisgruber and Lawrence Sager, a church-owned soup kitchen would pay the same taxes as a soup kitchen owned and operated by a secular Ms. Campbell.93 Ex hypothesi, the two kitchens perform the same function, provide the same services to the same people, and present the same risks of workers’ need for unemployment compensation. That one is religiously motivated is irrelevant, on this view. All that matters is that they provide the same type of service.94

Justice Jackson’s concurrence suggests that this functional approach is consistent with the Court’s opinion in Catholic Charities. If her view is correct, then on remand, the Wisconsin Supreme Court could adopt a functional interpretation, while avoiding both a motivational approach and the now constitutionally forbidden inquiry into whether a provider’s activities are sufficiently religious in content. And if the Wisconsin court did follow this reasoning, presumably it would reach the same outcome as it did before, denying an exemption to Catholic Charities.95

But there are at least two reasons to be skeptical about this possibility. The first, and most obvious, is that none of the other Justices joined Justice Jackson’s concurrence. If others shared her view that the functional approach was available, they could easily have signaled as much. That they did not suggests at least ambivalence, if not outright rejection, of her view. Second, and more importantly, the functional approach conflicts with the Court’s reliance on motivation to show a violation of denominational neutrality. In explaining why it was impermissible to require that activities have additional religious content beyond religious motivation, Justice Sotomayor pointed to Catholic Charities’s compliance with Catholic doctrine, which prohibited proselytizing and religious discrimination in the provision of charitable services.96 The Court recognized that Catholic Charities was religiously motivated not to engage in those activities. That motivation had the effect of turning its omission—its refusal to engage in proselytization or discrimination—into a form of religious conduct, namely, the provision of social services not otherwise infused with religious content.

On the motivational view, the distinction between why an entity performs some action and what that action is (or what it means) collapses. Consider some standard examples involving other religious exemptions. The use of peyote or hoasca is a crime under the Controlled Substances Act,97 but if its use is religiously motivated, then the conduct is part of a religious rite and protected by statutory exemptions.98 Similarly, if a dozen people congregate in their home in the midst of a pandemic to socialize, that might be a breach of public-health regulations. But if their gathering is religiously motivated, they are engaged in a protected activity under the First Amendment.99 If a corporation refuses to provide coverage for FDA-approved forms of contraception, it could be subject to onerous regulatory fines. But if its refusal to provide that coverage was religiously motivated, then it might be characterized as the avoidance of religious complicity—a form of free exercise—protected under federal law.100 Or if a person is fired from their job because they refuse to work on Sundays in order to engage in some secular activity, such as caring for their children or elderly parents, they have no claim for a workplace accommodation. But if they are religiously motivated to worship on that day, then they may be entitled to relief under federal employment discrimination law.101

In all these examples, religious motivation for engaging (or refusing to engage) in some conduct changes the legal meaning of that action (or omission). The reasons for doing something are relevant in determining what is being done.102 On the motivational view that the Court seems to embrace across a wide range of exemption cases, including Catholic Charities, the why of an action matters for—and may indeed be constitutive of—what that action is or what it means for purposes of granting exemptions under various statutory and constitutional doctrines.103 In short, motivation cannot be severed from function. What an entity does turns, at least in part, on why the entity is doing it.

If something like this view is at work, even implicitly, in Catholic Charities, then the functional approach proposed by Justice Jackson cannot be an option for applying religious exemptions that track FUTA’s “religious purposes” provision. The functional interpretation will be undermined by the claim that religious motivation alters an entity’s function. Justice Jackson wrote that on the functional view, “Soup kitchens feed the hungry. Shelters house the homeless.”104 But on the motivational view, what a religiously motivated entity “actually does” is transformed by its reasons.105 A church-owned soup kitchen fulfills a religious duty. It engages in what the Court calls a “theological practice[].”106 Under this interpretation, motivation and function cannot be easily separated. To know what an entity’s function is may require knowing its reasons for action.

To be sure, the Court was not explicit about adopting this motivational interpretation, which had been urged upon it by Catholic Charities,107 by the United States as amicus curiae,108 and by the dissent in the Wisconsin Supreme Court.109 But acceptance of this view helps to explain the logic of the Court’s holding with respect to denominational neutrality, including its emphasis on Catholic Charities’s religious reasons for not engaging in various forms of religious conduct. This view also provides a plausible rationale for why other Justices might not have joined in support of an alternative, function-based interpretation of FUTA’s (and Wisconsin’s) religious exemption.110

II. motivational irrelevance

My claim up to this point is that Catholic Charities reflects the Court’s acceptance of a deeper view about religious motivation, pointing toward a principle of motivational sufficiency. According to this principle, a private individual’s (or entity’s) religious motivation is sufficient to trigger a claim for religious exemption.111 To say that motivation is sufficient means that no other religious conduct can be required, at least not without running afoul of the Court’s principle of denominational neutrality, which, as we have seen, forbids government discrimination among religions.

While religious motivation plays this crucial role in exemption doctrine, the Court seems to have adopted a radically different principle with respect to its significance for government action. When public officials exercise their governmental authority based on religious reasons, their motivations seem to have no bearing on the constitutionality of their actions under the Establishment Clause. In these circumstances, the Roberts Court now appears to embrace a principle of motivational irrelevance, according to which religious motivation is never constitutionally relevant when the government regulates private conduct or engages in its own speech or expression.

In this Part, I consider whether the Court has, in fact, embraced the irrelevance of religious motivation in the context of government support for religion under the Establishment Clause. I argue that, in signaling its rejection of a secular purpose requirement, the Court seems to have moved sharply in this direction. I then examine the doctrinal asymmetry that arises from adhering simultaneously to the principles of motivational irrelevance and sufficiency. Finally, I suggest that despite the resulting incoherence, the Court is likely to persist in maintaining these contradictory approaches to religious motivation.

A. Whither Secular Purpose?

Has the Roberts Court, in fact, adopted a principle of motivational irrelevance, such that whether the government has acted on the basis of religious motivation has no bearing in determining the constitutionality of its action? If the Court has embraced this principle, or some version of it, then the Court has taken an asymmetric view about when motivation matters. In the context of granting religious exemptions, motivation is sufficient to make actions religious, but in the context of government speech and coercion, motivation would be irrelevant. I have argued above that the Court is committed to a motivational view with respect to exemptions. Here, I offer some evidence that the Court may also accept the irrelevance of motivation when it evaluates government action.

The argument for this conclusion turns on the Court’s recent rejection of the Establishment Clause framework that it had applied for decades under Lemon v. Kurtzman.112 The Lemon test, which was much derided by conservative jurists and scholars,113 held that a law violated the Establishment Clause unless it satisfied three requirements: “First, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion; finally, the statute must not foster ‘an excessive government entanglement with religion.’”114 The first part of this test recognized a “secular purpose requirement,”115 which the Court applied on several occasions in finding violations of the Establishment Clause.116 Under weak versions of this requirement, the government was forbidden from being “motivated wholly by religious considerations.”117 In later decisions, the Court relied on stronger formulations, according to which a secular purpose had to be the “preeminent” or “primary” purpose,118 or at least that the “secular purpose . . . ha[d] to be genuine, not a sham, and not merely secondary to a religious objective.”119 But notice that, under all of these variations on Lemon’s secular purpose requirement, it was possible for the government’s religious motivation to be relevant under the Establishment Clause. Even under the weakest version, the government had to point to some “adequate secular object.”120 If a law was “entirely motivated by a purpose to advance religion,” then it lacked a constitutional basis.121 At the least, if the government’s only reason was religious, then under that limited circumstance, its motivation mattered to the permissibility of its action.122 In short, for decades—before and after Lemon—the Court recognized that the government’s motivation was relevant under the Establishment Clause.

But in the last few years, all of this has changed—at least if the Roberts Court’s recent statements about Lemon are to be taken at face value. In Kennedy v. Bremerton School District, which held that a public high school was required to allow a football coach to pray on the field after games,123 the Court declared that it had “long ago abandoned Lemon.”124 Writing for the majority, Justice Gorsuch described Lemon in standard terms, stating that its “approach called for an examination of a law’s purposes, effects, and potential for entanglement with religion.”125 The Court also rejected Justice O’Connor’s “endorsement test offshoot”126 of Lemon’s secular purpose and effects requirements,127 which asked whether a “‘reasonable observer’ would consider the government’s challenged action an ‘endorsement’ of religion.”128 According to Gorsuch, all this had been swept aside. In its place, the Court had adopted an interpretation of the Establishment Clause based on “reference to historical practices and understandings.”129 There was no indication that any part of the Lemon test survived under the Court’s turn toward history and tradition.130 The Court did not qualify its abandonment of Lemon or preserve any aspect of that decision. In dissent, Justice Sotomayor stated that the majority had “overrul[ed] Lemon entirely and in all contexts”131 and “call[ed] into question decades of subsequent precedents.”132 The majority did not contradict or correct her.

If the Court has indeed rejected a secular purpose requirement as part of its “abandonment” of Lemon, the implications would be profound. It would be permissible for the government to act on religious motivations in regulating private conduct, provided its regulations are otherwise secular in their content. If the government were to impose requirements to engage in religious practices, it might be barred from doing so regardless of its motivation or intent. For example, whatever their motivations, public officials cannot compel students to pray or to participate in religious education.133 But if those officials were religiously motivated to ban otherwise secular conduct, such as abortion,134 same-sex marriage,135 gender-affirming care,136 teaching evolution,137 or school dancing138—to name only a few litigated examples139—their decisions would not be subject to challenge under the Establishment Clause. Whether a law, or any other governmental action, is constitutionally permissible would have nothing to do with its religious motivation. Even laws justified solely on religious grounds would be legitimate. That conclusion, at any rate, would follow from rejecting Lemon’s secular purpose requirement—even in its weakest forms—in favor of a principle of motivational irrelevance.

Unless the Court finds its way back to some version of a secular purpose requirement through its approach to “historical practices,” which seems rather unlikely given criticisms of the requirement from conservative Justices,140 the range of permissible, religiously motivated government action could extend to serious infringements on constitutional rights. As Professor Andrew Koppelman has observed, without a secular purpose requirement, nothing prevents the government from offering religious reasons as compelling governmental interests to justify restricting individual liberties.141 Consider, again, a religious prohibition on same-sex marriage. In the years leading up to Obergefell v. Hodges,142 state courts denied that religious purposes were rational grounds for refusing to recognize a right to same-sex marriage.143 Perhaps less explicitly, but nevertheless unmistakably, the Supreme Court did the same in Obergefell.144 If Lemon and its secular purpose requirement are dead, however, then it is not clear on what basis the Court could reject laws justified by religious reasons, whether in the context of same-sex marriage or other matters involving rights that conflict with deeply held religious views.145

Carried to its logical conclusion, the abandonment of Lemon, or at least its first prong, would have radical consequences. As Professor Koppelman has argued, “If the state were not limited to secular purposes, the effects would be so far reaching that constitutional law would be unimaginably different from what it is now.”146 The secular foundations of governmental authority, including the protection of basic constitutional rights, would be put in question. For that reason, it might be difficult to accept the Court’s own description of its doctrine. Perhaps the Court’s “abandonment” of Lemon must be taken “seriously, but not literally.”147 The Court has rejected Lemon and its progeny, especially in the contexts of school funding and government religious expression,148 but a literal reading—one that allows for laws based solely on religious reasons—would go too far. And yet, the Court has not given any indication that a literal reading is mistaken. Perhaps it will have an opportunity to do so in the future, as lower courts continue to invalidate laws for lack of an adequate secular purpose.149

B. The Incoherence of Motivational Asymmetry

If the Court has rejected a secular purpose requirement, as it claims, then constitutional doctrine would seem to embrace both the principle of motivational sufficiency, which makes religious motivation crucial for granting exemptions, and the principle of motivational irrelevance, according to which religious motivation has no role to play in determining whether government coercion or expression is constitutional. Yet these two principles are starkly at odds with each other. It is not clear how the Court could accept both simultaneously and maintain a coherent jurisprudence.150 This Section argues that while the Court cannot reconcile these principles, its doctrinal commitments ensure that motivational asymmetry will be a persistent feature of its Religion Clause jurisprudence.

To see the conflict between the principles, we can begin with the underlying premise of the sufficiency principle: that motivation is relevant to—and perhaps even determines—the meaning of actions. As Justice Bradley argued in her Wisconsin dissent, “religious motivation makes an activity religious.”151 Interestingly, and perhaps tellingly, in her list of examples of actions that could be rendered religious by acting for religious reasons, Justice Bradley included that “[o]ne could erect a cross to promote a Christian message or honor fallen soldiers.”152 And here she cited American Legion v. American Humanist Association, in which the Court rejected an Establishment Clause challenge to the Bladensburg Cross—a thirty-two-foot tall Latin cross displayed on public land in Maryland—concluding that it had a “secular meaning” as a war memorial.153 But by Justice Bradley’s logic, if the Cross had been erected to promote a Christian message,154 that would have been a religious act.155 On this view, there is no distinction between the religiosity of public and private actions taken for religious reasons. Whether the actor is a private individual or a government official exercising legal authority, motivations play a role in determining whether their conduct is religious. There might be a further question whether, by engaging in religiously motivated conduct, the government violates the Establishment Clause. On some interpretations, the Clause permits certain forms of religious expression or conduct by the state.156 Even so, there is no question that motivation is at least relevant to whether the conduct at issue is religious.

Another example might help to illustrate the tension between the two principles. Suppose, again, that Justice Bradley is correct, and that religious motivation can determine the meaning of both private and governmental action. Now imagine that the same public officials who erect a cross for religious reasons then refuse to certify same-sex marriages because they believe that doing so would contravene the will of God.157 Such a decision would impose coercive sanctions on couples who seek the legal benefits of marriage. To the extent motivation makes meaning, those sanctions would enforce religious conduct, namely, conformity with a theological prohibition on same-sex marriage. If the Establishment Clause bars coercion of religious conduct,158 then this prohibition would violate that rule. So would religiously motivated bans on abortion, in vitro fertilization, gender-affirming care, the use of preferred pronouns, and other forms of conduct that political majorities find objectionable on religious grounds. If motivational sufficiency holds for such regulations, then the government might, at least under some circumstances, be implicated in religious coercion. A consistent application of the motivational view of religious action would leave open the possibility that such coercion is an impermissible use of government power under the Establishment Clause. And if that is correct, the principle of motivational irrelevance—which holds that religious motivation is never relevant to the constitutionality of government conduct or expression—must be false. If it is impermissible for the government to coerce religious conduct, and if motivation is relevant (let alone determinative) for whether conduct is religious, then motivation must matter, at least in some circumstances, under the Establishment Clause.

One way to resolve this tension would be to reject the principle of motivational irrelevance. In other words, the Court could give up on the idea that motivation never matters for determining whether conduct is religious under the Establishment Clause. If this seems unlikely, given the Court’s “abandonment” of Lemon,159 another solution would be to work in the opposite direction. The Court could reject the sufficiency principle by denying that religious motivation, without more, is enough to establish eligibility for exemptions.

Within the American constitutional tradition, there is a strand of doctrine that adopts something like this latter view. In Reynolds v. United States, the nineteenth-century precedent rejecting a free exercise exemption for Mormon polygamy, Chief Justice Waite asked, “Can a man excuse his practices to the contrary [of the law] because of his religious belief?”160 And he famously answered that “[t]o permit this would be to make the professed doctrines of religious belief superior to the law of the land, and in effect to permit every citizen to become a law unto himself.”161 For the Reynolds Court, religious motivation for illegal conduct was irrelevant under the Free Exercise Clause of the First Amendment: “It matters not that his belief was a part of his professed religion.”162

The Court’s skepticism in Reynolds about religious motivation as a basis for legal exemption was echoed more than a century later in Employment Division v. Smith, which rejected a free exercise accommodation for the use of peyote.163 Justice Scalia wrote that the Court had “never held that an individual’s religious beliefs excuse him from compliance with an otherwise valid law prohibiting conduct that the State is free to regulate.”164 Formally speaking, Smith remains the law with respect to interpreting the Free Exercise Clause, with the Court declining recent invitations to overrule it.165

Yet attempting to minimize the significance of religious motivation runs counter to the recent trajectory of free exercise doctrine. There is little doubt that Smith has been seriously undermined by a series of free exercise decisions over the past decade,166 and its continuing vitality is doubtful.167 In several notable cases, the Court has also granted statutory exemptions for religiously motivated conduct.168 Any reconciliation of the Court’s approach to religious motivation that depends on the view articulated in Reynolds and Smith would be working hard against the prevailing current of free exercise jurisprudence, which is now more solicitous of exemption claims than at any point in American history.169 In short, there is little prospect that the Roberts Court would accept an account of free exercise that diminishes the role of religious motivation in prompting claims for exemptions.

What seems more likely, then, is that the Court will continue to embrace both principles despite their inconsistency. When it comes to exemptions, motivation matters—it transforms the social meaning of conduct in ways that trigger statutory and constitutional protections. But when it comes to the exercise of government power, either through coercive regulations or through official expression, the significance of religious motivation fades away entirely. The result is a jurisprudence that privileges religious motivation when doing so serves religious ends and suppresses it when taking it into account would limit government impositions of religion.

III. against asymmetry

This Part briefly anticipates several objections to the claim that motivation should matter symmetrically across the Free Exercise and Establishment Clauses. In other words, if motivation is relevant to exemptions, it must also matter with respect to government expression and conduct that promotes religion. Some of these objections are familiar from arguments against secular purpose requirements that have been addressed elsewhere.170 But it may be helpful to restate them here, if only because they might be invoked in support of the asymmetric view that, while motivation matters for exemptions, it should be irrelevant in evaluating government speech and coercion under the Establishment Clause.

First, critics sometimes claim that courts are not well-situated to ascertain the motivations of public officials, especially when they are part of multimember bodies like legislatures.171 But this objection runs directly into the Court’s animus jurisprudence,172 including in the free exercise context, where the Justices have readily applied a totality-of-the-circumstances approach in finding hostility toward religion.173 For example, in Masterpiece Cakeshop v. Colorado Civil Rights Commission, which involved a Christian baker who refused to serve a gay couple celebrating their marriage, the Court relied on statements made by civil rights commissioners to show a failure of “governmental neutrality.”174 And in Kennedy, the Court doubled down on this approach, indicating that “[a] plaintiff may also prove a free exercise violation by showing that ‘official expressions of hostility’ to religion accompany laws or policies burdening religious exercise.”175 If the Court can conduct a sensitive inquiry into the statements and views of public officials for purposes of determining religious animus, it is difficult to see why the same inquiry cannot focus on intentions to promote religion, rather than to inhibit it. Moreover, to the extent the Court requires neutrality toward religion, one might expect that it would apply the same methodological approach to determining official purposes, intentions, and motivations—or expressions of them—whether in opposition to religion or in preference of it.

A second standard objection is that many facially neutral laws are consistent with religious convictions. That such laws might be religiously motivated is not a reason to invalidate them. A typical example here is a prohibition on murder. Similarly, in rejecting an Establishment Clause challenge to restrictions on public funding of abortion, the Court in Harris v. McRae stated that religious “oppos[ition] [to] stealing does not mean that a State or the Federal Government may not, consistent with the Establishment Clause, enact laws prohibiting larceny.”176 Religious justifications can overlap with secular purposes, and laws that are supported by both are not, for that reason, vulnerable under the Establishment Clause.

But this conclusion does not undermine either a well-formulated secular purpose requirement or the argument above against the principle of motivational irrelevance. Some laws might be justified solely on religious grounds. Under that circumstance, with no independent secular purpose, the objection simply disappears. And indeed, even after McRae was decided in 1980, the Court proceeded to apply a secular purpose requirement in several cases,177 before its recent declaration in Kennedy that it had abandoned Lemon.178 As Justice Souter wrote in McCreary County v. ACLU of Kentucky, “[While] the secular purpose requirement alone may rarely be determinative . . . , it nevertheless serves an important function.”179 That function, according to Justice O’Connor, was to “remind[] government that when it acts it should do so without endorsing a particular religious belief or practice that all citizens do not share.”180 But even if one does not accept O’Connor’s focus on endorsement, her point holds more generally. A secular purpose requirement guarantees that the government has some religiously neutral basis for its action.

There can be disagreement about whether laws motivated primarily by religion have sufficient secular justification. But when laws are motivated solely by religious considerations, their meaning is fixed by that motivation, in a way that the standard counterexamples (of laws against murder, theft, etc.) fail to address. Such laws may be infrequent, but their scarcity may itself be a function of the longstanding Lemon framework, which prohibited laws motivated solely on religious grounds. Without Lemon’s secular purpose requirement, laws motivated primarily or entirely by religious justifications could become more prevalent. But whatever the effect of abandoning that rule, the argument from mixed motives and overdetermined purposes is not sufficient to show that religious motivations are always irrelevant.181 If there are some instances in which governmental actions are entirely motivated by religious convictions, and if the meaning of those actions is determined by their motivations, then motivations are relevant in at least those limited circumstances. And if motivations are relevant in this way, that is sufficient to defeat the principle of motivational irrelevance—the claim that motivations never matter in determining the constitutional permissibility of governmental action.

A final objection to the relevance of religious motivation is that it would be unfair and discriminatory to exclude the religious motivations of religious citizens and public officials in enacting legislation and other government policies.182 Singling out religious convictions for special disability is, according to this argument, an expression of hostility toward religion.183 The effect of this argument is to turn a secular purpose requirement under the Establishment Clause into a violation of the Free Exercise Clause.184

But this argument faces two difficulties. First, under the principle of motivational sufficiency, religious convictions are singled out for special treatment when it comes to exemptions. Secular beliefs are not taken to alter the meaning of actions in the same way.185 But if religious motivations have this effect, it is hard to see why they do not also transform the meaning of government action. And if that is the case, then some account is needed to explain why it is fair to single out religious motivations as relevant for purposes of receiving special legal protections, while being treated like secular political and ideological views as permissible grounds for legal decision-making. The second problem, mentioned above, is that if this fairness argument is taken to its conclusion,186 it is difficult to see how courts could resist reliance on religious reasons, not only as grounds for lawmaking but also as compelling interests for overriding individual rights. The Court has never embraced this view, which has potentially radical and theocratic implications.187 Perhaps this marks a limit to how far the principle of motivational irrelevance can be taken.

Conclusion

It may seem that we have traveled some distance from Catholic Charities. Yet, in the litigation leading to the Supreme Court’s decision, religious motivation played a central role in the arguments for granting and extending exemptions. Although the Court’s opinion focused on denominational neutrality, its analysis presupposed that religious motivation alters the meaning of conduct. The Court did not explicitly reject Justice Jackson’s functional approach to understanding “religious purposes,” but the logic of its decision runs contrary to her view. Religious motivation is sufficient as a basis for triggering exemption claims because why an entity does something affects our understanding of what it is doing. Motivation is relevant to function, even if not necessarily determinative of it.

But at the same time, the Court seems to have embraced an interpretation of the Establishment Clause that makes motivation irrelevant to understanding the meaning of government action—which denies that motivation can inform, or determine, what it is that the government is doing. This asymmetry is difficult, and perhaps impossible, to justify. There is a deep tension, then, emerging in the Court’s doctrine of religious freedom, and resolving it with integrity requires developing a more systematic and consistent view about when motivation is relevant to both the free exercise and disestablishment of religion. Given the Court’s apparent commitment to treating motivation as relevant with respect to exemptions, it should apply the same reasoning with respect to government expression and coercion. Rejecting motivational irrelevance under the Establishment Clause would shore up a secular purpose requirement and guard against the possibility of limiting individual rights and liberties on the basis of religious reasons. But the Roberts Court has shown no indication that it will pursue a consistent approach. Instead, adopting an asymmetric view of motivation reflects and advances this Court’s preference for religion.188

Hardy Cross Dillard Professor of Law, University of Virginia School of Law. For helpful comments and discussion, I thank Leslie Kendrick, Christopher Lund, Ira C. Lupu, James Nelson, Richard Schragger, and Nelson Tebbe. I thank Kelly Scrivner for excellent research assistance and the editors of the Yale Law Journal Forum for their engagement with this Essay.