Introduction
A revolution has occurred in the law of religious freedom. At this point, the picture is reasonably clear. The Supreme Court has greatly expanded the scope of the Free Exercise Clause. In almost every Term over the past decade, it has taken cases involving free exercise challenges and sided with religious parties, requiring equal access to public funding1 and authorizing exemptions from a wide range of state and federal laws.2 It is no exaggeration to say that the Roberts Court has been more solicitous of free exercise claims than any of its predecessors.3 At the same time, the Court has also dramatically curtailed or altogether abandoned precedents that had limited government religious expression and taxpayer s…
Introduction
A revolution has occurred in the law of religious freedom. At this point, the picture is reasonably clear. The Supreme Court has greatly expanded the scope of the Free Exercise Clause. In almost every Term over the past decade, it has taken cases involving free exercise challenges and sided with religious parties, requiring equal access to public funding1 and authorizing exemptions from a wide range of state and federal laws.2 It is no exaggeration to say that the Roberts Court has been more solicitous of free exercise claims than any of its predecessors.3 At the same time, the Court has also dramatically curtailed or altogether abandoned precedents that had limited government religious expression and taxpayer support for religious institutions.4 It has been, to borrow a phrase, “[d]isestablishing the Establishment Clause.”5
If the picture is clear, however, it nevertheless emerges from a kind of jurisprudential pointillism. An overall pattern forms from distinct decisional dots, which may appear disconnected, but which merge into a more meaningful image, at least when seen from a distance. Many of the Court’s recent religious freedom decisions are based on the revival or resuscitation of individual precedents that had not been invoked in decades and that had limited, if any, extensions or applications.6 But rather than explain how these precedents are connected, and rather than draw clear lines between them, the Court has moved from one to the next, leaving the future of each undeveloped and uncertain.7 And even when it has drawn a clear line, the Court has stopped abruptly and without explanation.8
Still, stepping back, any observer can see that the era of “strict separation” between church and state is over.9 In the mid- to late twentieth century, the Supreme Court had constrained government sponsorship of religion, both symbolically and financially, and it had rarely authorized constitutionally mandated religious exemptions.10 But the Roberts Court has performed an about-face. It has worked — sometimes incrementally,11 but increasingly in lurches12 — to transform constitutional doctrines shaping the relationship between government and religion. On one hand, the Court now requires equal treatment of religious individuals and associations with respect to funding and access to other publicly available benefits,13 and, on the other, it requires special treatment in the form of accommodations from laws that burden religious beliefs and practices.14 When these doctrines come together, they create a structure of preference for religion — what we have called structural preferentialism15 — which results in constitutional favoritism for religious over nonreligious views and, despite the Court’s recent protestations to the contrary,16 for some religious denominations over others.
The Court’s embrace of this paradigm — equal treatment for benefits and special exemptions from burdens — has proceeded largely under the banner of religious neutrality. The concept of neutrality has long played a prominent and even central role in the Court’s decisions under the Religion Clauses of the First Amendment.17 And it is easy to see why. The idea that the state should not take sides among competing and conflicting religious denominations — or among religious and nonreligious views — resonates within the liberal constitutional tradition and draws support from basic principles of toleration, fairness, and nondiscrimination.18 But as critics have long argued, without further specification, the concept of religious neutrality is ambiguous and susceptible to inconsistent application.19 Some have gone further to argue that neutrality is impossible,20 but the Court has never accepted that skeptical view. Instead, the Justices continue to invoke various conceptions of neutrality, without offering a systematic account of the principle or its applications. The result is a pattern of decisions that are often — although as we shall see, not always — justified in terms of neutrality but that together reveal a preference for religion.
We have previously analyzed the doctrine from an external perspective that accounts for these decisions as if they were the products of conflicts among competing interest groups.21 Here, we add an internal criticism of how various approaches to neutrality operate in the doctrine and in justifications offered for it. In our analysis, we do not defend a particular conception of neutrality and instead aim to show how the Court’s own logic works across doctrinal domains.
To develop our claim, in Part I, we show that the Court has relied on principles of neutrality and nondiscrimination in revolutionizing its religious freedom jurisprudence. But different conceptions of neutrality are at work in different parts of the doctrine, and, as the Court shifts between them, it creates a legal structure that favors religion. This is how neutrality gives way to preference. In Part II, we argue that the Court’s decisions from this Term involving religious freedom — Catholic Charities Bureau, Inc. v. Wisconsin Labor & Industry Review Commission,22 Mahmoud v. Taylor,23 and Oklahoma Statewide Charter School Board v. Drummond24 — extend this pattern in the Court’s pointillistic approach to the Religion Clauses. Last, in Part III, we ask: What, if anything, is wrong with structural preferentialism? Although the separationist consensus has collapsed, its warnings against religious establishment are especially salient in this antiliberal moment, marked by intense polarization, rising religious disaffiliation, and increasing calls for ethnoreligious nationalism. The case for disestablishment — requiring neutrality not only among religious denominations but also between religious and nonreligious views — may have fallen on hard times. But the arguments for it, which we renew, warn against the expansion and entrenchment of religious preferentialism.
Continue Reading in the Full PDF
* Hardy Cross Dillard Professor of Law, University of Virginia School of Law.
* * Walter L. Brown Professor of Law, University of Virginia School of Law.
* * * Jane M.G. Foster Professor of Law, Cornell Law School. For comments and discussion, we thank Alan Brownstein, Michael Dorf, Chad Flanders, Frederick Gedicks, Andrew Koppelman, Douglas Laycock, Katherine Metheny, Vincent Phillip Muñoz, Melissa Murray, Doug NeJaime, James Nelson, James Oleske, Alan Patten, Laura Portuondo, Richard Re, Zalman Rothschild, Amy Sepinwall, Elizabeth Sepper, Nomi Stolzenberg, Xiao Wang, and the editors of the Harvard Law Review.
Footnotes
^ See Carson v. Makin, 142 S. Ct. 1987, 1997 (2022); Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2262–63 (2020); Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 467 (2017).
^ See, e.g., Fulton v. City of Philadelphia, 141 S. Ct. 1868, 1875, 1882 (2021) (granting an exemption from antidiscrimination rules); Tandon v. Newsom, 141 S. Ct. 1294, 1297 (2021) (per curiam) (public health regulations); Roman Cath. Diocese of Brooklyn v. Cuomo, 141 S. Ct. 63, 65–66, 69 (2020) (per curiam) (same); Our Lady of Guadalupe Sch. v. Morrissey-Berru, 140 S. Ct. 2049, 2055 (2020) (employment discrimination law); Masterpiece Cakeshop, Ltd. v. Colo. C.R. Comm’n, 138 S. Ct. 1719, 1723–24 (2018) (invalidating application of antidiscrimination law); see also Little Sisters of the Poor Saints Peter & Paul Home v. Pennsylvania, 140 S. Ct. 2367, 2373 (2020) (upholding statutory religious exemption from contraceptive mandate); Zubik v. Burwell, 578 U.S. 403, 405–08 (2016) (per curiam) (applying the Religious Freedom Restoration Act to contraception mandate and vacating judgments below); Burwell v. Hobby Lobby Stores, Inc., 573 U.S. 682, 688–90, 692 (2014) (same); Holt v. Hobbs, 574 U.S. 352, 356 (2015) (applying the Religious Land Use and Institutionalized Persons Act to prison grooming standards).
^ See Lee Epstein & Eric A. Posner, The Roberts Court and the Transformation of Constitutional Protections for Religion: A Statistical Portrait, 2021 Sup. Ct. Rev. 315, 323–24 (2022) (“The popular notion that the Roberts Court represents a break in the development of the jurisprudence of the religion clauses is amply supported by the data.”). In fact, even setting aside free exercise challenges to funding restrictions, the Roberts Court has granted more religious exemptions in the past decade than all its predecessors combined. From the incorporation of the Free Exercise Clause in 1940 through 1990, the Court allowed accommodations in only a handful of free exercise cases. See Leo Pfeffer, The Supremacy of Free Exercise, 61 Geo. L.J. 1115, 1126–30 (1973) (collecting free exercise cases up to Sherbert v. Verner, 374 U.S. 398 (1963), and finding that “whenever free exercise stood alone it was unsuccessful”); James E. Ryan, Smith and the Religious Freedom Restoration Act: An Iconoclastic Assessment, 78 Va. L. Rev. 1407, 1414 (1992) (noting that after granting an exemption in Sherbert, the Court “rejected thirteen of the seventeen free exercise claims it heard”). Between 1990 and 2012, the Court affirmed only one free exercise challenge. See Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 524 (1993). The Court has since recognized free exercise claims in more than a dozen merits cases. See Adam Liptak, Will Religion’s Remarkable Winning Streak at the Supreme Court Continue?, N.Y. Times (Mar. 30, 2025), https://www.nytimes.com/2025/03/30/us/politics/supreme-court-religion.html [https://perma.cc/GZ4F-HDD6] (“Since 2012 . . . the pro-religion side has won all but one of the 16 signed decisions in argued cases that concerned the First Amendment’s [Religion Clauses].”). The number grows substantially if one counts cases on the Court’s “shadow docket” during the pandemic. See Stephen I. Vladeck, The Most-Favored Right: COVID, the Supreme Court, and the (New) Free Exercise Clause, 15 N.Y.U. J.L. & Liberty 699, 702 (2022) (surveying pandemic-related free exercise challenges resolved by the Court on emergency motions).
^ See Am. Legion v. Am. Humanist Ass’n, 139 S. Ct. 2067, 2074 (2019) (government religious display); Kennedy v. Bremerton Sch. Dist., 142 S. Ct. 2407, 2427 (2022) (abandoning the test from Lemon v. Kurtzman, 403 U.S. 602 (1971)).
^ Frederick Schauer, Disestablishing the Establishment Clause, 2022 Sup. Ct. Rev. 219, 219 (2023).
^ We develop this point at various places throughout, but notice, initially, that when the Court decided Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission, 138 S. Ct. 1719 (2018), in 2018, it relied on Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520 (1993), see Masterpiece Cakeshop, 138 S. Ct. at 1731, which the Court had not applied in the preceding twenty-five years. Fulton v. City of Philadelphia, 141 S. Ct. 1868 (2021), relied on Sherbert v. Verner, 374 U.S. 398 (1963), see Fulton, 141 S. Ct. at 1877, which had not been applied in this way in thirty years. And the cases this Term followed a similar pattern. Catholic Charities Bureau, Inc. v. Wisconsin Labor & Industry Review Commission, 145 S. Ct. 1583 (2025),* *revived Larson v. Valente, 456 U.S. 228 (1982), see Cath. Charities, 145 S. Ct. at 1592–93, a precedent that the Court had not invoked to extend an exemption in the past forty-three years, and Mahmoud v. Taylor, 145 S. Ct. 2332 (2025), relied upon Wisconsin v. Yoder, 406 U.S. 205 (1972), see Mahmoud, 145 S. Ct. at 2352–53, which yielded no progeny over the past fifty-three years. Instead of connecting these cases under some more general framework, the Court reclaimed individual precedents that had no real life until their recent revival.
^ See, e.g., infra note 79 and accompanying text (discussing the Court’s reluctance to rely on Masterpiece Cakeshop).
^ There is a clear doctrinal line in the Court’s school funding cases, but no explanation for why it was not extended this Term in Oklahoma Statewide Charter School Board v. Drummond, 145 S. Ct. 1381 (2025). See infra section II.C, pp. 240–45.
^ See Richard Schragger, Micah Schwartzman & Nelson Tebbe, Reestablishing Religion, 92 U. Chi. L. Rev. 199, 202 (2025) (detailing the Court’s rejection of separationism); Ira C. Lupu & Robert W. Tuttle, The Remains of the Establishment Clause, 74 Hastings L.J. 1763, 1766 (2023).
^ See Ira C. Lupu, Hobby Lobby and the Dubious Enterprise of Religious Exemptions, 38 Harv. J.L. & Gender 35, 49–51 (2015) (showing that the Warren and Burger Courts rarely granted religious exemptions).
^ See Justin Driver, The Supreme Court, 2021 Term — Comment: Three Hail Marys: Carson, Kennedy, and the Fractured Détente over Religion and Education, 136 Harv. L. Rev. 208, 223 (2022) (“Carson[ v. Makin, 142 S. Ct. 1987 (2022),] did not arrive as a bolt from the blue, but instead represented the logical culmination of recent Free Exercise Clause opinions.”).
^ See Jim Oleske, Tandon Steals Fulton*’s* Thunder: The Most Important Free Exercise Decision Since 1990, SCOTUSblog (Apr. 15, 2021), https://www.scotusblog.com/2021/04/tandon-steals-fultons-thunder-the-most-important-free-exercise-decision-since-1990 [https://perma.cc/BE2F-AY7W].
^ Schragger, Schwartzman & Tebbe, supra note 9, at 212–13.
^ Id. at 213.
^ Id. at 208–09 (introducing the idea of structural preferentialism).
^ See infra section II.A, pp. 227–31.
^ See 2 Kent Greenawalt, Religion and the Constitution: Establishment and Fairness 181 (2008). See generally Andrew Koppelman, Defending American Religious Neutrality (2013).
^ See Cécile Laborde, Liberalism’s Religion 72–73 (2017) (exploring the concept of neutrality); Alan Patten, The Normative Logic of Religious Liberty, 25 J. Pol. Phil. 129, 144 (2017) (proposing a principle of fair opportunity for religious freedom); Nelson Tebbe, Religious Freedom in an Egalitarian Age 71 (2017) (defending a principle of fairness toward others).
^ See Douglas Laycock, Formal, Substantive, and Disaggregated Neutrality Toward Religion, 39 DePaul L. Rev. 993, 1008 (1990).
^ See Steven D. Smith, The Illusion of Religious Neutrality, Law & Liberty: F. (Apr. 3, 2012), https://lawliberty.org/forum/the-illusion-of-religious-neutrality [https://perma.cc/TPG4-CMRA] (“[R]eligious neutrality . . . turns out to be impossible.”).
^ See Schragger, Schwartzman & Tebbe, supra note 9, at 203–04.
^ 145 S. Ct. 1583 (2025).
^ 145 S. Ct. 2332 (2025).
^ 145 S. Ct. 1381 (2025) (mem.). Notably, none of this Term’s religious freedom cases were decided on originalist grounds. See Cath. Charities, 145 S. Ct. at 1591; Mahmoud, 145 S. Ct. at 2350. The Court did not engage in any effort to discern the original meaning, intent, or understanding of the Religion Clauses, which extends its record of inconsistent and selective use of originalism in this context. See Caroline Mala Corbin, Opportunistic Originalism and the Establishment Clause, 54 Wake Forest L. Rev. 617, 618–19 (2019).