Tag: Anti-discrimination Law

The Supreme Court’s decision in Carson v. Makin is the third in a trilogy of cases dramatically upending the meaning of the First Amendment’s Religion Clauses. Beginning with Trinity Lutheran v. Comer in 2017 and followed by Espinoza v. Montana Department of Revenue in 2020, the Court has moved forward with an aggressive project of transforming the Religion Clauses into a broad anti-religious-discrimination clause. In this paper, I trace this doctrinal devolution and argue that the Court’s novel reinterpretation is deeply misguided. By design, the Religion Clauses require discrimination—religion is to be treated differently from non-religion in a broad range of state action. The contemporary Supreme Court, however, has inverted this most basic insight. The Court’s new Religion Clause jurisprudence is also on a collision course with its burgeoning government speech doctrine. This doctrine recognizes that in a democratic polity, every policy choice entails paths not chosen. Government must be able to select its own message, and in turn, discriminate against those messages it wishes not to communicate, tempered by accountability at the ballot box. Granted, to say that discrimination is sometimes required under the Religion Clauses and the Government Speech Doctrine is not to say discrimination against religion is always constitutional. Protections against objectionable discrimination remain as vital as ever. The Court’s public forum doctrine, for example, protects free expression of religion from content-based discrimination when the government itself is not speaking. The heart of the Court’s recent Religion Clause decisions, however, is a jurisprudentially backward constitutional mandate that government actively subsidize religious speech to avoid a Religion Clause “discrimination” claim. It is a command that government express ideas it may not wish to express. The Court’s reimagining of the Religion Clauses is inconsistent with the First Amendment’s original meaning, potentially harmful to both government and religion, and in direct tension with the Government Speech Doctrine.

What a difference five years makes. In 2017, I feared that the Court was “lead[ing] us . . . to a place where separation of church and state is a constitutional slogan, not a constitutional commitment.” Today, the Court leads us to a place where separation of church and state becomes a constitutional violation.

—Justice Sonia Sotomayor¹Carson v. Makin, 142 S. Ct. 1987, 2014 (2022) (Sotomayor, J., dissenting).

INTRODUCTION

The Religion Clauses of the First Amendment require discrimination. Such an assertion may appear counterintuitive in an era prone to viewing subjects of controversy through a lens of equality, but by their very terms the Free Exercise Clause and the Establishment Clause demand that religion be treated differently from other objects of government attention. Today, however, the Supreme Court tells us a different story. Despite the clear language in the Constitution, the Court’s most recent jurisprudence suggests that the religion clauses do something very different than what the words chosen by their framers would suggest.

Beginning in 2017, the Court moved forward with an aggressive project of transforming the Religion Clauses into a broad anti-religious discrimination clause. In this paper, I trace this doctrinal misadventure and argue that the Court’s novel reinterpretation is deeply misguided. This approach, I contend, is precisely backwards. The Religion Clauses are not the Equal Protection Clause. The Court’s conflation of the Religion Clauses with anti-discrimination principles directly contravenes the design and intended function of this critical part of the First Amendment. It is also antithetical to a core principle of popular sovereignty: that a state—and hence, the people—must be able to choose its own priorities and be held accountable for the choices it makes, a key premise underlying the Court’s government speech doctrine.

There are many reasons to find fault in Constitutional doctrine. But whether it is substantive due process and the meaning of the word “liberty” in the Fourteenth Amendment or the right to keep and bear arms in the Second Amendment, such critiques typically boil down to this: the Court is either reading too far into the language of the Constitution or not far enough. It is either finding more meaning then is there, or too little. With the Court’s most recent turn in its religion clause jurisprudence something very different has occurred. Instead of going too far or not far enough, the Court has effectively inverted the very purpose of the Religion Clauses. These clauses, as designed by framers with an understanding of the weighty historical role religion has played in society and governance, carve out religion for a uniquely nuanced, one-of-a-kind treatment. Religion is special. It receives an unusual and distinctive protection from government intervention and is subjected to unusual and distinctive limitations on government support. In between these two constitutional poles established by the Religion Clauses, governments have discretion to make religion-related policy choices. But the unique Janus-faced design of the Religion Clauses sends a clear message: the Constitution requires that religion be treated differently.

Up until 2017, critics of the Court’s religion clause jurisprudence generally fell into the standard camps. They argued, for example, that the Court was restricting too much government activity that “respect[s] an establishment of religion,” as Justice Stewart did in his dissent in Engel v. Vitale addressing a nondenominational school prayer.²Engel v. Vitale, 370 U.S. 421, 444–50 (1962) (Stewart, J., dissenting). Such an exercise, to Stewart, simply did not rise to the level of establishing an “official religion.”³Id. at 450 (Stewart, J., dissenting). Other critics have argued that the Court was not capacious enough in defining what it means to “prohibit” the free exercise of religion, such as Justice Brennan’s dissent in Braunfeld v. Brown, in which he asserted that making a religious practice “economically disadvantageous” should be a sufficient free exercise claim.⁴Braunfeld v. Brown, 366 U.S. 599, 616 (1961). These cases turned on the unique status of religion––and the extent to which government was treating it differently, as required by the Constitution. And in some contemporary cases, it is still taking this approach, moving the needle much more aggressively than in the past, siding with critics who have supported expansive, and distinctive “free exercise” protection.⁵See Kennedy v. Bremerton Sch. Dist. 142 S. Ct. 2407 (2022). But as of 2017, the Court also started asking an entirely different, and contradictory question. Inexplicably, differential treatment of religion went from a Constitutional mandate to a Constitutional infraction. 

The first sixteen words of the U.S. Constitution’s First Amendment are straight forward: “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof . . . .”⁶U.S. Const. amend. I. The constitutional historian Leonard Levy has asserted that “Nowhere in the making of the Bill of Rights was the original intent and meaning clearer than in the case of religious freedom.”⁷Leonard W. Levy, The Establishment Clause: Religion and the First Amendment xv (1986). On its face this language prohibits the federal government from making or enforcing laws that do either of two independent things: respect an establishment of religion or prohibit the free exercise of religion. For over three-quarters of a century, this language has been understood to have been incorporated by the Fourteenth Amendment, and thus to apply with equal vigor to the states as to the federal government.⁸See Cantwell v. Connecticut, 310 U.S. 296 (1940).

In addition to defining precisely what is included in the category of laws “respecting an establishment of religion” or “prohibiting the free exercise,”⁹Philip B. Kurland, The Origins of the Religion Clauses of the Constitution, 27 Wm. & Mary L. Rev. 839, 856 (1986) (quoting U.S. Const. amend. I, cl. 1). the key interpretive challenge of these two clauses has been their inherent tension. In devising the unique structure of the religion clauses (or, we might say religion clause, singular, to emphasize the interdependence of the anti-establishment and free exercise principles) the framers left behind a distinctive jurisprudential task for courts, incomparable to any other part of the Constitution. What is required or implicitly encouraged by one clause might appear to be prohibited by the other––and in between there may be a zone—what the Court has long referred to as a “play in the joints”¹⁰Walz v. Tax Comm’n, 397 U.S. 664, 669 (1970).—where a government may, but is not required, to advance the interests of free exercise or anti-establishment without being prohibited from doing so by the countervailing clause.

The precise contours of the religion clauses continue to be worked out. The drafting history of the religion clauses—particularly, the meaning the framers intended to give to an “establishment of religion”—leaves us with gaping holes in our understanding.¹¹Levy, supra note 7, at 84. One notable area of disagreement in the late twentieth century, for example, has been the debate among jurists and scholars as to whether establishment demands so-called strict separation between church and state or mere nonpreferentialism, that is, not preferring one sect or religion over another.¹²David Reiss, Jefferson and Madison as Icons in Judicial History: A Study of Religion Clause Jurisprudence, 61 Md. L. Rev. 94, 126 (2002). Various justices on the Supreme Court have long presented differing framings of history in their Religion Clause jurisprudence, confirming that the historical “record does not speak in one voice.”¹³Id. at 144. But regardless of where one falls in these debates, and however “religion” may be defined, one thing seemingly remained a constant: the religion clauses of the First Amendment single out a thing called “religion” for disparate treatment. While the debate was not definitively settled over precisely where the lines of impermissible establishment or prohibition on free exercise should be drawn, what was clear was that the Constitution established unique lines for religion, prohibiting both governmental favoritism as well as active suppression.

This idiosyncratic Constitutional status of religion vis-à-vis government, which may be seen as a form of mandatory discrimination, is grounded in a set of founding-era philosophical beliefs about the need to protect religion from government and government from religion. As Thomas Jefferson wrote in an 1802 letter to the Danbury Baptist Association, the First Amendment “buil[t] a wall of separation between Church & State.”¹⁴Thomas Jefferson’s Letter to the Danbury Baptists (Jan. 1, 1802), https://www.
loc.gov/loc/lcib/9806/danpre.html [https://perma.cc/2D4H-5ZGJ]. And while some scholars have disputed the significance of Jefferson’s famed “wall of separation” metaphor, in 1947 the Supreme Court affirmed Jefferson’s reading in forceful terms. It did not merely agree that “[t]he First Amendment has erected a wall between church and state,” ^{15Everson v. Bd. of Educ., 330 U.S. 1, 18 (1947).} it emphasized that the “wall must be kept high and impregnable.”¹⁶Id.

 In 2022 however, the Court issued an opinion that was nothing short of radical. For the Religion Clauses, it was a world turned upside-down. This is not to say that changes had not been on the horizon. Before the recent seismic leap, the Court’s religion jurisprudence had been on a steady retreat from the Jeffersonian vision, particularly as we passed into the new millennium. But Carson v. Makin, capping off a trio of cases that began in 2017, was of a different magnitude.

 Thomas Jefferson’s “wall of separation” between church and state has gone from a route impeded by a barrier “high and impregnable” twenty-five years ago, to one riddled with easily breached fissures shortly thereafter, to an obstruction not merely demolished but replaced and paved over by a wide road—with a shuttle bus travelers are compelled to ride and a fare they are compelled to pay. In Carson the Court did not merely backtrack from its longstanding prohibition on the expenditure of government funds on sectarian schooling; it held, for the first time, that the Free Exercise Clause prohibits a state from not using taxpayer money to fund religious education.¹⁷Carson v. Makin, 142 S. Ct. 1987, 2010 (2022). A government, in other words, may be constitutionally obligated to do, what for most of the Court’s jurisprudential history addressing the religion clauses it had been forbidden from doing: paying for religious education.

This was so despite the glaring objection that such funding directly conflicts with a straight-forward, textual reading of the Constitution’s prohibition of any law “respecting an establishment of religion.”¹⁸U.S. Const. amend. I. Government may be required to utilize taxpayer funds to pay for religious education in spite of the strong belief of the First Amendment’s framers “that no person, either believer or non-believer, should be taxed to support a religious institution of any kind.”¹⁹Everson, 330 U.S. at 12. Government may be compelled to provide an affirmative benefit to religion, despite an absence of evidence that it is in fact “prohibiting” a religion’s free exercise. Indeed, a state may be required to pay for religious education even in the face of its own strong policy reasons for not doing so. How did this happen? And what is the constitutional basis for this revolutionary reformulation?

Carson v. Makin and the two cases leading up to its holding transformed the Free Exercise Clause into an anti-religious-discrimination clause. The Religion Clauses however, were designed to produce the very opposite result, to ensure that religion was treated differently. Religion receives special treatment in the Constitution, precisely because the framers appreciated its unique power. Religion has the ability to inspire, to shape humankind’s deepest and most intimate sense of meaning and well-being, to establish and frame social obligations that supersede or conflict with civic commitments, and to ignite wars, social instability, and bloodshed. In the words of Roger Williams, the theologian and founder of Rhode Island whose religious advocacy for separating church and state left an indelible imprint during America’s colonial era, “[t]he blood of so many hundred thousand souls of Protestants and papists, spilled in the wars of present and former ages for their respective consciences, is not required nor accepted by Jesus Christ the Prince of Peace.”²⁰Mark A. Graber, Foreword: Our Paradoxical Religion Clauses, 69 Md. L. Rev. 8, 9 (2009) (quoting Roger Williams, The Bloudy Tenent of Persecution for Cause of Conscience 1 (Edward Bean Underhill ed., The Society 1848) (1644)).

While a vast sphere of human activity is open to government control, establishing an array of rules determining, for example, the boundaries of criminal and civil conduct, Williams emphasized that religion is different. “God requires not a uniformity of religion to be enacted and enforced in any civil state; which enforced uniformity, sooner or later, is the greatest occasion of civil war, ravishing of conscience . . . .”²¹Id. at 10. According to historian Leonard Levy, James Madison believed that state “[e]stablishments produced bigotry and persecution, defiled religion, corrupted government, and ended in spiritual and political tyranny.”²²Levy, supra note 7, at 55. It was clear that religion, in short, must receive special treatment in its relation to the state.

Carson, however, tells us that this is all wrong. Religion is instead to be treated the same as other human endeavors—at least, for certain purposes. Instead of standing as a mandate for distinctive treatment of religion, the newly reconfigured twenty-first century Religion Clauses, prohibit distinctive treatment. How did the Court justify such a profound—and some might say bizarre—reversal? One explanation is that the Court was drawing on the post-Civil War legacy of the Fourteenth Amendment and modern-America’s strong ethic of opposing inequality and discrimination in its many forms.

Perhaps some justices were also responding to an underlying feeling that religious adherents are looked down upon by societal elites and had not been invited onto the equality train with the same gusto as other identity groups; perhaps this jurisprudential turn was their chance at a ticket. Justice Scalia made such feelings clear in a 2004 dissent when he complained that

 An aggrieved Justice Thomas, in his recent Espinoza concurrence, points the finger directly at other justices, lamenting that “this Court has an unfortunate tendency to prefer certain constitutional rights over others . . . The Free Exercise Clause . . . rests on the lowest rung . . . .”²⁴Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2267 (2020) (Thomas, J., concurring).

It is possible that the Court is taking its cues from grievances such as these. But whatever the motive, the Court had decided, without acknowledging that it was doing so, to completely reimagine the Religion Clauses. In the Carson trio the First Amendment’s religion clauses are framed, not as they have been traditionally construed, as granting religion a unique constitutional status, but as a demand that religion effectively be placed on the same plane as everything else. Granted, this insistence on anti-religious discrimination is not evenly applied. As we shall discuss further, its flattening of the religion clauses is selective. In other contexts, the Court—in the very same term it decided Carson—concluded that a state employee, while acting within his official duties, has special rights of religious expression and practice that he would not possess outside of the religious sphere.²⁵See Kennedy v. Bremerton Sch. Dist. 142 S. Ct. 2407 (2022).

I.  WHY DISCRIMINATION?

Although it may not be commonly acknowledged, government is in the discrimination business. It discriminates every time it “establish[es] Justice, insure[s] domestic Tranquility, provide[s] for the common defence, promote[s] the general Welfare, and secure[s] the Blessings of Liberty.”²⁶U.S. Const. pmbl. All of these ends, eloquently laid out by the founding fathers in the Preamble of the U.S. Constitution, necessarily require America’s government of “we the people” to make choices. There are many routes to realizing, maintaining, and even defining domestic tranquility, the general welfare, and core liberties. And for every policy choice, there are paths not chosen. In a world of scarce resources and fierce disputes over how to allocate those resources to best achieve societal goals, government must not merely decide how much to allocate to particular goals, but which goals are worthy of its energies in the first place.

In a functional and sustainable democracy, this process of discrimination ideally keeps a polity on a trajectory of responsiveness and improvement. Government discrimination allows the state to make discerning choices that take into account an array of complex interests and counter-interests. It allows for action rather than paralysis in light of the needs and pressures coming from a multitude of directions, the often overwhelming and conflicting demands that are part and parcel of having to accommodate a large, diverse, and pluralistic population. Government discrimination in a working democracy means that hard choices will be made; costs will be weighed against benefits. But ultimately, if democracy is functioning in its ideal form, these choices will generally reflect societal values, interests, and goals, while helping correct for the errors of the past as they become evident.

Being “discriminating” thus may be associated with thoughtful, careful, decision-making. And indeed, the Constitution itself not only invites relatively open-ended policy-based discrimination rooted in democratic deliberation, but the document in many places calls for particular kinds of discrimination. It tells us in Article II that we must discriminate against those who are not “natural born” when choosing a president.²⁷U.S. Const. art. II, § 1. While the federal government has the power to “lay and collect Taxes,”²⁸U.S. Const. art. I, § 8, cl. 1. certain kinds of taxes are explicitly verboten (or “discriminated against”) such as tariffs “laid on Articles exported from any State.”²⁹U.S. Const. art. I, § 9, cl. 5. A similar form of discrimination characterizes the Religion Clauses of the First Amendment; religion is explicitly designated as a subject of government regulation to be treated differently from non-religion in a broad range of state action.

Granted, this was not the case under the initial conception devised at the 1787 Constitutional Convention in Philadelphia. The framers’ original notion was one in which the federal government was to be inherently limited to powers enumerated in Article I. Since regulation or establishment of religion was not explicitly included among these powers, a discriminatory carve-out for religion was thought to be superfluous. A bill of rights, including such special treatment for religion, was initially deemed unnecessary because as Madison explained, “[t]here is not a shadow of right in the general government to intermeddle with religion.”³⁰3 Jonathan Elliot, The Debates in the Several State Conventions on the Adoption of the Federal Constitution 330 (2d ed. 1836). Other founding era notables however, remained skeptical. Many states conditioned their support of the new charter on a pledge to make the implicit, explicit.³¹ Steven D. Smith, The Religion Clauses in Constitutional Scholarship, 74 Notre Dame L. Rev. 1033, 1038 (1999). Madison was ultimately persuaded of the merits of this alternative view held by many Anti-Federalists. He became concerned that

under the clause of the constitution, which gave power to congress to make all laws necessary and proper to carry into execution the constitution, and the laws made under it, [congress may be] enabled . . . to make laws of such a nature as might infringe the rights of conscience, or establish a national religion . . . .³²Id. at 1039 (quoting James Madison).

Madison realized that even under a regime of limited government in which federal powers are circumscribed by their enumeration in Article I, the government may use its lawful powers in ways yet unanticipated––and that this exertion of power may bleed into religious establishment or the freedom of individual exercise. Because the constitutional structure that limited government power could not be relied upon as the sole guarantor that church and state would be confined to separate spheres, as with other discrete topics, insurance in the form of the Bill of Rights was deemed expedient. And with religion, the remedy was especially distinctive. The two clauses of the First Amendment do not merely single out religion, but do so in an unusual Janus-faced manner suited to the sui generous dilemma that plagued the history of church-state relations. As Justice Robert Jackson explained, “the Constitution sets up [a difference] between religion and almost every other subject matter of legislation, a difference which goes to the very root of religious freedom . . . .”³³Everson v. Bd. of Educ., 330 U.S. 1, 26 (1947) (Jackson, J., dissenting).

Religion-related practices receive special discriminatory free exercise benefits exempting them from targeted restrictive governmental regulation that in non-religious spheres would constitute an ordinary part of democratic governance. A wide array of behaviors are targeted by government for distinctive kinds of punishment, prohibition, or penalty, but actions that relate to religion—unlike these other realms of behavior—may not be targeted. They receive a free pass from the Free Exercise Clause of the First Amendment. At the same time that government is generally free to choose to partner with, endorse or incorporate a diverse range of philosophical worldviews, values, or private institutions into its operations, religion may not be among them. Religion is uniquely burdened by the Establishment Clause’s distinctive prohibition on intermingling religion and government.

The two religion clauses simultaneously work together and are at odds with one another. On one hand, they may be said to serve similar ends. “An establishment, Madison argued, ‘violated the free exercise of religion’ and would ‘subvert public liberty.’ ”³⁴Levy, supra note 7, at 168. On the other, they appear in direct tension, one seeming to facilitate religious practice by specially prohibiting government interference and the other seeming to discourage religious practice by specially withholding government largess. The clauses both demand unique benefits for, and impose distinctive burdens on, religion––and sometimes these simultaneous constitutional commands overlap, producing puzzling and uncertain results. It is this apparent paradox that set the stage for a peculiar species of constitutional doctrine, an anomalous and sensitive area of jurisprudence with one common baseline: religion must be treated differently.

II.  DEFINING DISCRIMINATION

The Britannica Dictionary defines “discriminating” as “able to recognize the difference between things that are of good quality and those that are not.”³⁵Discriminating, Britannica Dictionary, https://www.britannica.com/dictionary/

discriminating [https://perma.cc/5UN5-LKX3]. However, discrimination is a word with more than one definition. Discrimination may simply describe the act of “recogniz[ing] a difference between things.”³⁶Discriminate, Britannica Dictionary, https://www.britannica.com/dictionary/discriminate [https://perma.cc/46AX-56WV]. Today, the word “discrimination” is commonly understood as a pejorative. A country founded on egalitarian ideals, with a shamefully inegalitarian past and a present in which identity politics are paramount, has given the word “discrimination” toxic properties. This contemporary understanding aligns with another definition, courtesy of Oxford: to “discriminate” is “to treat one person or group worse/better than another in an unfair way.”³⁷Discriminate, Oxford Learner’s Dictionary, https://www.oxfordlearnersdictionaries.
com/us/definition/english/discriminate [https://perma.cc/PBJ3-LR48].

The First Amendment demands that government treat religion in ways that are arguably both “worse” and “better” than the treatment of other subjects garnering the government’s attention. However, such differential treatment is arguably the epitome of “fairness,” that is if one believes applying clearly stated rules of the U.S. Constitution with principled consistency may generally be understood to be a paradigmatic example of “fairness.” Thus, this latter––pejorative––definition is inapposite to the religion clauses. Yet, merely attaching the word “discrimination” to any government action—whether it be in a political speech, a New York Times op-ed, a Fox News commentary, or a Supreme Court opinion—casts reflexive doubt on that act’s legitimacy. Thus, the irony: use of the phrase “government discrimination against religion”—a constitutional mandate serving the interests of both government and religion—will likely strike the average listener as a nefarious wrong.

There are of course many forms of discrimination that are rightfully prohibited by the Fourteenth Amendment, such as invidious differential treatment based on an individual’s race, gender, or sexual orientation. Other forms of identity-based discrimination, including discrimination rooted in religious animus, may be precluded by statutory anti-discrimination laws. However, the existence of unfair or unjust forms of discrimination—that in some cases are forbidden by the Constitution—should not be used to create the misleading impression that the vital government discrimination required by aspects of the First Amendment is in fact an inherent evil that must be stamped out. Acknowledging that parts of the Constitution require or permit some forms of discrimination does not detract from the continued need (or ability under the law) to combat bigotry.

Granted, in certain contexts, evidence that a government is “discriminating” against or in favor of a particular religion may expose a potential Religion Clause violation. But this is not because the clauses contain a general anti-discrimination principle comparable to the Equal Protection Clause or statutory anti-discrimination law, rather, it is because they demand religion be treated differently from other objects of governmental attention.³⁸See, e.g., Est. of Thornton v. Caldor, 472 U.S. 703 (1985). With other government action, the default is that a democratic state generally must be able to make discriminating distinctions in its policy and enforcement choices.

Discrimination is a baseline for an effective governance. It is the stuff of democratic and legal contestation. Thus, for purposes of this article, I will generally use discrimination in its non-pejorative form—as a mere act of recognizing distinctions between different classes of things resulting in some form of differential treatment. Yes, “discrimination” can be unjust or unfair. “Anti-discrimination” laws and the scrutiny courts apply to invidious discrimination under the Equal Protection Clause of the Fourteenth Amendment have long been directed at such unjust forms of discrimination. However, discrimination can also suggest a kind of discernment that is more typically lauded—such as the ability to distinguish a Matisse or a rigorous scientific study at a top research university from the work produced by a seventh grader in their art or science class. In between the extremes there is enormous room to debate as to whether particular distinctions drawn and differences applied are beneficial or harmful, unfair or justified. And, the Court had historically made such “breathing room” between the discrimination required (or merely allowed) under the Free Exercise Clause and the discrimination required (or merely allowed) under the Establishment Clause, a central component of its religion clause jurisprudence.³⁹See, e.g., Walz v. Tax Comm’n, 397 U.S. 664, 669 (1970).

Nonetheless, words are powerful things. They can be used to manipulate, as well as elucidate. Unfortunately, conflating various definitions of “discrimination,” which is all too common today, may serve the former end. A casual use of the word may create the false impression that particular differential treatment is morally or normatively suspect, when in fact it may be socially desirable––or even a legal requirement. Regretfully, the Supreme Court has gotten in on the act. The Religion Clauses of the First Amendment, very much unlike the Equal Protection Clause, mandate discrimination. Yet, as we shall see, recent religion jurisprudence has mischaracterized “discrimination” as a Constitutional wrong, instead of a Constitutional imperative.

III.  HOW WE GOT HERE: THE LOCKE DISSENT FORESHADOWS A NEW FIRST AMENDMENT

A state may have free reign when it comes to establishing an official state bird, flower, or song, but the Establishment Clause insists that religion is different. That same state may not establish Buddhism or Zoroastrianism or Christianity as its official religion. And the Constitution commands not merely that government shall “make no law respecting an establishment of religion,” it may not prohibit “the free exercise thereof” either.⁴⁰U.S. Const. amend. I. The state, in the guise of its police powers, may regulate, prohibit, punish, and penalize a full spectrum of human behavior, unless that behavior it is targeting constitutes “an exercise of religion.” While political and constitutional theorists may debate the reasons for this mandatory discrimination––many, including Madison, suggest it serves both the interests of the government and the respective religion that may not be established by the government.⁴¹Reiss, supra note 12, at 103. While one might debate the extent and nature of the qualitative benefits Madison foresaw, it is “discrimination” loud and clear.

The Court acknowledged this plain reading of the First Amendment as recently as 2004 in a decision by then Chief Justice Rehnquist. He pointed out, in the context of potential state funding for religious training, that the First Amendment’s unique approach to religion “find[s] no counterpart with respect to other callings or professions. That a State would deal differently with religious education for the ministry than with education for other callings [in other words, that it would discriminate] is a product of these views, not evidence of hostility toward religion.”⁴²Locke v. Davey, 540 U.S. 712, 721 (2004). It was not surprising that his opinion allowing for a selective government scholarship program that excluded theological training read like an exercise in constitutional common sense, with just two dissenters. After all, it had only been two years since the Court, in a controversial 5–4 Establishment Clause decision, first allowed a school voucher program that provided tuition aid to private religious schools to stand.⁴³Zelman v. Simmons-Harris, 536 U.S. 639 (2002).

However, beginning with Trinity Lutheran Church of Columbia v. Comer in 2017, followed by Espinoza v. Montana Department of Revenue in 2020, and most recently, in Carson v. Makin in 2022, the Court radically inverted this natural and widely accepted reading of the Religion Clauses. Admittedly, this novel interpretation of the religion clauses did not appear out of the ether. In that same case in which Chief Justice Rehnquist issued his short ten-page majority opinion rejecting the free exercise inspired demand that the State of Washington pay for a student’s post-secondary religious schooling, Justices Scalia and Thomas dissented and articulated the view that would become the approach of a Court majority beginning in 2017.⁴⁴Locke, 540 U.S. at 726–34.

For many decades prior to this decision, the Court had interpreted the Establishment Clause as an outright bar on state funding of religious exercise.⁴⁵Carson v. Makin, 142 S. Ct. 1987, 2012 (2022). However, beginning in the late 1990s, with the case of Agostini v. Felton,⁴⁶Agostini v. Felton, 521 U.S. 203 (1997).and culminating in Zelman v. Simmons-Harris in 2002, the Court orchestrated what Professor Nelson Tebbe has called a “contemporary turnabout” in its antiestablishment law.⁴⁷Nelson Tebbe, Excluding Religion, 156 U. Pa. L. Rev. 1263, 1265 (2008). Indirect aid to parents for vouchers to pay for religious education, and even some direct aid to religious institutions, was now a constitutional policy option for legislators across the nation.⁴⁸Id. at 1266. 

With Zelman, the Court’s religion jurisprudence had just jumped from a world in which government funding of religious education had been presumed to be unconstitutional under the Establishment Clause, to one in which a closely divided Court tenuously held that it was permitted under certain narrow circumstances. In Locke, two dissenters, just two years later, were arguing that such funding was not merely allowed, but required under the Free Exercise Clause, foreshadowing the even more radical changes that were soon to come. Effectively, these two dissenters were arguing—in contravention of the well-established conventional textual reading—that rather than requiring religion be treated differently, the religion clauses instead imposed a broad anti-discrimination mandate. In just a decade and a half, this trial run of the anti-religious-discrimination Free Exercise Clause would transform into the majority view on the Court.

Here was the dissenters’ proposed statement of the rule: “When the State makes a public benefit generally available, that benefit becomes a part of the baseline against which burdens on religion are measured; and when the State withholds that benefit from some individuals solely on the basis of religion, it violates the Free Exercise Clause . . . .”⁴⁹Locke, 540 U.S. at 726–27 (Scalia J., dissenting). The logic might run as follows: “[P]rohibiting the free exercise” of religion under the First Amendment involves imposing some form of “burden” on such exercise. After all, a “prohibition” imposed by a savvy public official seeking to harm or diminish religion would not typically come in the form of a straight-forward law banning a particular religion or religiosity outright; the more strategically astute approach would be a law that indirectly makes certain elements of a religious practice more difficult, or impossible. Laws with an indirect impact on religion may burden religion. The question then becomes, how do we determine whether there has been such a “burden?” It would seem that to the Locke dissenters, if a “generally available” public benefit is not available to all, we may deem those to whom it is not available, “burdened.”

The dissent utilizes an unexpected dose of post-modern relativistic logic that puts government in the foreground. Their reasoning effectively suggests that it is outside forces—in this case the government—that establish reality for religious practitioners. A burden may be inflicted on religion not just by virtue of what government does to religion, but by virtue of what government does elsewhere. It is as if the dissenters were looking to Article III’s demand that compensation of federal judges “not be diminished during their Continuance in Office,”⁵⁰U.S. Const. art. III. and reasoning that a change in tax law reducing the mortgage interest deduction is unconstitutional because it makes purchasing a home for a judge more expensive, thereby “diminishing” the relative value of their compensation. The baseline for judging whether free exercise has been burdened is not the unique, longstanding, and deeply rooted practices of the particular religion affected by the government action (or inaction), it is government policy and the relative benefits it provides to various other societal actors. With this peculiar logical maneuver, a constitutional provision that on its face demands religion be treated differently—and protected in ways that other life philosophies or practices are not—is inverted to become one that prohibits religion from being treated differently.

At the same time, the dissent begs the question, what does “generally available” mean? Clearly, all public benefits are subject to rules dictating who is, and who is not eligible. A scholarship fund for post-secondary education will presumably not be available to five-year-olds, nor to those who wish to self-educate in isolation in the woods. The concept of “general availability” requires some sort of limiting principle. If “generally available” simply means that the public benefit at issue is offered in accordance with a relatively fixed non-discretionary rule for some category or categories of non-religious purposes or beneficiaries, this anti-religious-discrimination principle would have virtually limitless application. Considering the ubiquity of government in modern society, it would be an invitation for courts to mandate government-funded religion in virtually all spheres of public life. 

For most of the jurisprudential history of the religion clauses, the Court’s primary challenge, considering the inherent tension between the Establishment and the Free Exercise Clause, has been to craft doctrines determining when, and how much discrimination is required. Must religion be discriminated against when public funds incidentally benefit religious institutions in a way that is comparable to how other (secular) institutions benefit, or only when the funds exclusively target and support a particular religion? Must an anti-discrimination law be discriminatorily applied, exempting hiring and firing decisions by religious organizations from the anti-discrimination mandates that otherwise would apply?⁵¹See, e.g., Hosanna-Tabor Evangelical Lutheran Church & Sch. v. Equal Emp. Opportunity Comm’n, 565 U.S. 171 (2012). If so, must such discriminatory exemption apply to just religious ministers, or to all employees of a religious organization? These are the sorts of questions the Court previously asked: to what extent, in what manner, and in what settings do the differential treatment rules of the religion clauses apply to religious organizations and practitioners? The two Locke dissenters inverted the doctrinal question in Religion Clause cases, reframing them as an anti-discrimination mandate.

To critique the dissenter’s approach is not to deny that a violation of the Free Exercise Clause may involve discrimination against a religion or a religious practitioner. A legal ban on Rosary Beads would both arguably prohibit the free exercise of religion for practicing Catholics and at the same time discriminate against Roman Catholicism, treating it differently from other religious practices and secular owners of beaded jewelry. A straight-forward reading of the Free Exercise Clause however, would suggest that it is the prohibition on religious exercise and not the differential treatment that constitutes the constitutional infraction.

 As the Supreme Court has itself emphasized, in the Free Exercise Clause, “[t]he crucial word . . . is ‘prohibit’: ‘For the Free Exercise Clause is written in terms of what the government cannot do to the individual, not in terms of what the individual can exact from the government.’ ”⁵²Lyng v. Nw. Indian Cemetery Protective Ass’n, 485 U.S. 439, 451 (1988). Not only is there no evidence of a general anti-discrimination principle in the text of the Free Exercise Clause, as mentioned earlier, there is an explicit pro-discrimination principle. That is, when a broad legal restriction impacts both religious and non-religious actors, it may be that as to those affected religious individuals “free” religious “exercise” is literally being “prohibited,” entitling them, but not the non-religious affected individuals, to a discriminatory exemption from the law.

Granted, the Court has not been consistent on the question of required accommodations under the Free Exercise Clause. In a 1972 case addressing a state’s compulsory high school education law that was at odds with the practices of a particular religious community, the Court concluded that the Free Exercise Clause demands an exemption.⁵³Wisconsin v. Yoder, 406 U.S. 205 (1972). In contrast, the 1990 case of Employment Division v. Smith suggested that such required differential treatment under the Free Exercise Clause should be construed narrowly.⁵⁴Emp. Div. v. Smith, 494 U.S. 872 (1990). Then in 2012, a unanimous Court—citing both the Free Exercise and Establishment Clause—concluded that religious institutions are entitled to a ministerial exemption that allows them to fire a teacher of secular and theological subjects, even if such firing would otherwise contravene applicable anti-discrimination law.⁵⁵Hosanna-Tabor, 565 U.S. at 171. The Court explained that “imposing an unwanted minister . . . infringes the Free Exercise Clause, which protects a religious group’s right to shape its own faith and mission through its appointments.”⁵⁶Id. at 188.

Regardless of the uneven application over the years, the pro-discrimination implications of the Free Exercise Clause are clear. As O’Connor points out in her Smith concurrence, “A person who is barred from engaging in religiously motivated conduct is barred from freely exercising his religion . . . regardless of whether the law prohibits the conduct only when engaged in for religious reasons, only by members of that religion, or by all persons.”⁵⁷Smith, 494 U.S. at 893 (O’Connor J., concurring). The free exercise remedy, if it is to apply, would only benefit the religious practitioner—freeing him or her up from an otherwise application restriction—while leaving non-religious individuals burdened. It would, in other words, discriminate between religion and non-religion, treating them differently.

The new anti-religious-discrimination interpretation, in contrast, ignores these basic mechanics of the religion clauses. Scalia’s dissenting opinion in Locke is riddled with surprisingly sloppy reasoning. To support his reading of the Religion Clauses, he draws on an analogy to racial discrimination, yet fails to mention that the Court’s jurisprudence there is rooted in an entirely different part of the Constitution, with completely different language, structure and purpose.⁵⁸Locke v. Davey, 540 U.S. 712, 728 (2004). The Equal Protection Clause of the Fourteenth Amendment provides that “No state shall . . . deny to any person within its jurisdiction the equal protection of the laws.”⁵⁹U.S. Const. amend. XIV, § 1. It was not designed with the doctrinally formidable Janus-faced structure (and resulting built-in tension) of the religion clauses—which has led the Court to acknowledge a “play in the joints” between impermissible laws “respecting an establishment of religion” and unconstitutional measures “prohibiting” religion’s “free exercise.”⁶⁰Walz v. Tax Comm’n, 397 U.S. 664, 669 (1970).

In between the two clauses, in other words, there must be some room for laws that promote anti-establishment values but do not violate free exercise, and vice-versa. This is because laws aimed at avoiding establishment—in the direct sense—will almost invariably diminish free exercise; and laws intended to promote free exercise inevitably move toward establishment. It is a conundrum by design, built upon the Framers understanding of the precarious balance needed to maintain a safe buffer between church and state. The boundaries established by Court doctrine on either side necessitate judicial intervention into matters of religion that are not required of other spheres of government action. Yet, completely disregarding this unique structure of the religion clauses, Scalia instead drew a direct analogy to equal protection. To drive home his point, he argued that “A municipality hiring public contractors may not discriminate against blacks or in favor of them; it cannot discriminate a little bit each way and then plead ‘play in the joints’ when haled into court.”⁶¹Locke, 540 U.S. at 728 (Scalia, J., dissenting). But unlike the Equal Protection Clause this is precisely what the religion clauses require––discrimination—a delicate dance between anti-establishment and free exercise in which religion is given special treatment on both ends.

History is riddled with religious wars and instability. The Framers’ innovative formulation in the First Amendment was an attempt to protect the new nation from this same fate. Including only an Establishment Clause would have risked a government so intent on divorcing itself from religion that it would end up stymieing it—generating resentment and potentially violent revolt from passionate religious adherents who felt their free exercise was being choked. Include only a Free Exercise Clause and the danger for government and religion falls on the opposite end of the spectrum; a government openly facilitates and becomes intertwined with religious practice risking its politicization, and the perception (and likely reality) that the state is choosing favorites. Bitterness and backlash among those sects not granted politically favored status would naturally result. As an integrated whole, the two religion clauses were a Goldilocks solution.

As Professor Steven D. Smith observes, “[t]he words . . . ‘establishment of religion’ [and] ‘free exercise’—served to define the substantive area over which Congress was disclaiming jurisdiction.”⁶²Smith, supra note 31, at 1045. It was that simple. There is nothing in the First Amendment demanding that if non-religious governmental benefits are distributed, religious institutions should be entitled to equivalent goodies. Quite the contrary. The Equal Protection Clause of the Fourteenth Amendment and the Religion Clauses of the First Amendment are not the same.

IV.  THE RISE OF “NEUTRALITY”

How then to explain the dissenters’ conflation of principles from these two very different amendments in the Constitution—the Religion Clauses in the First Amendment and the Equal Protection Clause of the Fourteenth? It would seem that Scalia in his Locke dissent was drawing on the “neutrality” principle rooted in certain of the Court’s Establishment Clause decisions. In the seminal 1947 decision Everson v. Board of Education the Court upheld New Jersey’s reimbursement of bus transportation costs to parents sending their children to private schools, including those with a religious affiliation.⁶³Everson v. Bd. of Educ., 330 U.S. 1, 3 (1947). The Everson Court recounted the context in which the Framers’ drafted the religion clauses, stressing that early American settlers sought to escape the compulsion in Europe that they financially support churches favored by the government.⁶⁴Id. at 8. It emphasized—and included in full in the appendix—James Madison’s Memorial and Remonstrance, a tract written in opposition to a Virginia law that would have imposed a tax on its residents to support the established church.⁶⁵Id. at 11–12.

Despite ultimately rejecting the Establishment Clause challenge, the Everson Court insisted that “New Jersey cannot consistently with the ‘establishment of religion’ clause of the First Amendment contribute tax-raised funds to the support of an institution which teaches the tenets and faith of any church.”⁶⁶Id. at 16. It simply found that here, “[t]he State contributes no money to the schools. It does not support them. Its legislation, as applied, does no more than provide a general program to help parents get their children, regardless of their religion, safely and expeditiously to and from accredited schools.”⁶⁷Id. at 18. Under these circumstances the state was “a neutral in its relations with groups of religious believers and non-believers,”⁶⁸Id. at 17–18 (emphasis added). not unlike if it were providing police assistance for children crossing the street––some of whom happen to be traveling to or from a religious school.

“Neutrality,” in other words, was a way of distinguishing innocuous general welfare laws that just happen to have, among their many beneficiaries, religious individuals or institutions, from those constitutionally problematic laws that “respect an establishment of religion” by using taxpayer funds for targeted support of religion. If anything, neutrality as used in Everson is about understanding that religion must be treated differently, that while government has broad discretionary power to single-out and benefit all-sorts of respective groups or individuals through the policy distinctions it makes, the one exception is religion. The existence of neutrality (that is, that benefits are provided without regard to the religious status of the beneficiaries) provides support for the conclusion that it is not the kind of law that unconstitutionally respects an establishment of religion. Neutrality suggests that government is not targeting religion qua religion for a specific benefit in violation of the Establishment Clause.

 Neutrality was the principle that Scalia seemed to rely upon when he drew an analogy to equal protection in his Locke dissent, explaining that “[i]f the Religion Clauses demand neutrality, we must enforce them, in hard cases as well as easy ones.”⁶⁹Locke v. Davey, 540 U.S. 712, 728 (2004) (Scalia, J., dissenting). But, as we have seen, the “neutrality” of Everson is nothing like the general anti-religious discrimination rule the Locke dissenters portray it to be. The fact that the Court has turned to neutrality as a consideration in particular Establishment Clause settings does not transform the Religion Clauses more broadly, and particularly the Free Exercise Clause, into sweeping prohibition of religious discrimination.

The neutrality principle laid out in Everson is one evidentiary standard, among many, for determining whether or not a particular state may be targeting religion in a manner that is inconsistent with the Establishment Clause. Indeed, it is a method of determining when discrimination may be constitutionally required. Considering the fact that most policy choices by government will have some effect on some religious actors, neutrality is simply a device for separating the wheat from the chaff. By providing reimbursement of transportation costs for all schoolchildren—attending secular and religious schools alike—a state is no doubt promoting free exercise of religion. It is making it more affordable for religious parents to freely exercise their religion by educating their children at the religious school of their choice. The question then becomes: under these circumstances does the other religion clause demand discrimination, mandating that religion be treated differently and be denied, unlike the secular schools, this benefit?

Neutrality may be a useful tool in some establishment cases, but it is one that the Court has used only when appropriate, and not with consistency. Indeed, illustrating just how far the Court has moved on religion clause issues, we might observe that Everson itself was a closely contested 5–4 decision. Four dissenters were not convinced that a state should be allowed under the Establishment Clause, as part of a neutral public service program available to all parents, to reimburse families for the cost of sending their children to religiously affiliated schools.

The Locke dissent never explains why, by laying out a standard of “neutrality” in a narrow Establishment Clause context, Everson should now be understood to impose an equality rule under the Free Exercise Clause—requiring the Court to mandate, what in Everson, it just barely allowed. As we shall explore further, while establishment and free exercise may represent two ends of a tension rod, respectively they impose distinct kinds of constraints on government. Scalia, in his Locke dissent, conflates establishment and free exercise.

Justice Gorsuch utilized this conflation to profound effect in his 2022 majority decision in Kennedy v. Bremerton School District.⁷⁰Kennedy v. Bremerton Sch. Dist. 142 S .Ct. 2407 (2022). There he analyzed a public prayer by a public school coach at a public school event as largely a free exercise issue—whereas in the past the issue would almost certainly have been framed along Establishment Clause lines as an unconstitutional instance of a government official injecting his religion into a school-sanctioned activity. As the smoking gun, Gorsuch points out that “[b]y its own admission, the District sought to restrict [the coach’s] actions at least in part because of their religious character.”⁷¹Id. at 2422. It sought to prohibit actions “appearing to a reasonable observer to endorse . . . prayer.”⁷²Id. This was the “gotcha” moment to Justice Gorsuch; a conscientious choice by a school district to comply with the separation of church and state principles articulated in the Establishment Clause becomes damning evidence of a violation of neutrality under the Free Exercise Clause. This is a Religion Clause world turned upside-down.

In Kennedy the Court effectively overruled, indeed inverted, its Establishment Clause precedents recognizing an endorsement test.⁷³Id. at 2427. Public endorsement of religion by government went from prohibited, to prohibited to prohibit. This blowtorch to the Court’s previous jurisprudence, however, cannot alter the fact that the First Amendment, by its very terms, demands discrimination; a state may for legitimate policy purposes designate taxpayer funds to a specific circus school, driver’s education school, agricultural school, or most any other school it deems worthy, except if it is targeting religious education. As we shall discuss in the next Section, consistent with the government speech doctrine, a state has largely unconstrained discretion to choose its own policies and policy messages, except with regard to religion.

V.  THE EMERGING GOVERNMENT SPEECH DOCTRINE

There is some irony in this new muddying of the Religion Clause waters, as the Court has in recent years also moved toward clarification of another part of the First Amendment, one that resonates in the free exercise context: the government speech doctrine. The Free Speech Clause has over time come to incorporate a kind of anti-discrimination principle of its own. Despite reading as a simple across-the-board prohibition that “Congress shall make no law . . . abridging the freedom speech,”⁷⁴U.S. Const. amend. I. modern free speech case law has come to the realization that the most potent threats to expression come in the form of laws that specifically target (or “discriminate” against) particular content or viewpoints. After all, virtually all laws could be said to impact expression; whether it is blocking traffic on an eight-lane highway, setting private property ablaze, or assaulting a police officer in front of the nation’s capital, if human behavior is observable, it may be framed as expressive. Broad exemptions from criminal and civil accountability merely because the harmful behavior at issue happens to be observable would be intolerable; this was clearly not what the framers of the First Amendment had in mind.

The protection of free expression must have some limiting principle. Thus, the Court has come to differentiate between state attempts to silence particular ideas or ideologies from mere content-neutral “time place or manner” restrictions or regulations directed at harmful behavior that incidentally affects expression. Under the Supreme Court’s free speech doctrine, the former discriminatory treatment of certain content or viewpoints is subjected to a much higher level of judicial scrutiny than the latter—neutral regulations that may in some sense be said to inhibit expression, but without regard to content or viewpoint.⁷⁵See, e.g., Reed v. Town of Gilbert, 576 U.S. 155, 172–73 (2015). As the end of the twentieth century approached, the Court began to explicitly come to terms with the inverse principle. When it is the government that is doing the speaking, it must have the ability to discriminate.

In a sense, like religion under the Establishment and Free Exercise clauses, “government speech” under the free speech clause is different. It is a democratic imperative that government be able to discriminate in the ideas it conveys. Government must have the ability to choose its own message. It is the culmination of its messages and expressive actions, after all, for which the people hold government to account at the ballot box. Government “speaks” by, among other things, subsidizing particular activities, employing individuals to propagate particular messages, or installing monuments that convey certain ideas.⁷⁶See Rust v. Sullivan, 500 U.S. 173, 192–93 (1991); Pleasant Grove City v. Summum, 555 U.S. 460, 460 (2009). This is, by its very nature, an exclusionary activity.

As the government chooses to spread one message, it necessarily declines to communicate others. It discriminates based on content or viewpoint. As a new administration takes the helm in response to a shift in voter sentiments, a government will likely change its message. A city government might remove a statue of Robert E. Lee from a public park. It might replace that statue with one depicting the civil rights triumphs of Martin Luther King, Jr. A group of Civil War reenactors may object. However, their recourse is not in a First Amendment that guarantees them a right to have the government send the message they want it to send. It is the political process. The Court made this point succinctly in a 1991 case that would come to be described as the first in a series of cases that form the government speech doctrine.⁷⁷Helen Norton, The Government’s Speech and the Constitution 32–34 (Alexander Tsesis ed., 2019).

Rust v. Sullivan involved a government program that appropriated public funds for certain family-planning services.⁷⁸Rust, 500 U.S. at 178. In so doing, Title X of the Family Health Service Act stipulated that none of the allocated funds were to be used in programs that included abortion as a family-planning method.⁷⁹Id. As a plain-vanilla First Amendment free speech issue, one might assume that the government could not prohibit a counselor or physician from merely discussing a legal abortion as a medical option. Such discriminatory censorship directed at particular content might seem, on the most basic level, antithetical to core First Amendment principles. However, when it is the government that is speaking—as is arguably the case with a government program intended to promote certain goals but not others—the First Amendment prohibition on content or viewpoint-based discrimination is flipped on its head. We expect an anti-abortion administration to “be discriminating” when it comes to the messages it chooses to send about this volatile issue, just as pro-abortion rights voters would expect elected officials who run on a prochoice platform to propagate government speech that facilitates, rather than inhibits, the right to choose. To drive home its point, the Court in Rust provided this example: “When Congress established a National Endowment for Democracy to encourage other countries to adopt democratic principles, . . . it was not constitutionally required to fund a program to encourage competing lines of political philosophy such as communism and fascism.”⁸⁰Id. at 194.

Thus, if we return to the Court’s anti-discriminatory religion clause innovation, we can see another glaring tension. Even before this current Supreme Court’s most recent Religion Clause turnabout mandating certain government expenditures on religion, some had expressed concern that the growing prominence of the government speech doctrine might diminish previously viable Establishment Clause challenges—because of their potential framing as government speech.⁸¹Carol Nackenoff, The Dueling First Amendments: Government as Funder, as Speaker, and the Establishment Clause, 69 Md. L. Rev. 132, 147–48 (2009). But under the Court’s new regime, the anti-religious-discrimination doctrine and the government speech doctrine are on a collision course. A constitutional mandate that government subsidize religious speech to avoid a free exercise “discrimination” claim (just because such subsidy is also available to certain non-religious recipients), is a command that it express ideas it may not want to express, using taxpayer money. It is counter-majoritarian, and directly contradicts the principle underlying the government speech doctrine. In Rust, the Court reiterated the common sense conclusion that “[t]he Government can, without violating the Constitution, selectively fund a program to encourage certain activities it believes to be in the public interest, without at the same time funding an alternative program which seeks to deal with the problem in another way.”⁸²Rust, 500 U.S. at 193. It turns out, however, that this is not the case; that is, at least according to the Court’s novel anti-discriminatory religion clause doctrine.

One might respond, however, as pointed out earlier, that religion is different. Might there be something about religion that would justify a diversion from the otherwise applicable government speech principle? Could it be that this difference merits an exception from the intuitive notion that a government—as a representative of “we the people”—should be able to choose which policies or messages to propagate, and which messages not to endorse, or simply not expend taxpayer resources on? The Constitution, after all, already carves out certain areas in which simple majoritarian politics will not do, requiring instead a super majority for policy change. Fifty-one percent of the population, in other words, cannot do away with probable cause; the Constitution would have to be amended.

One might argue, for instance, that as a fundamental constitutional right, the free exercise of religion should be exempt from the baseline government speech rule. This is quite similar to what was argued by the dissenters in Rust. They pointed to the fact that the right to choose abortion under the “liberty” guarantee in the Fifth Amendment was (at the time) a fundamental constitutional right. As such, selective discrimination against the expression of certain medically pertinent information facilitating that freedom of choice, even under the auspices of a government program, was unconstitutional.⁸³Id. at 216.

 The Court, however, rejected this argument. It also left little room to doubt the basis of this rejection. Citing Regan v. Taxation with Representation, a decision in which the Court upheld a narrowly selective subsidy for lobbying by certain types of organizations, it explained that a “legislature’s decision not to subsidize the exercise of a fundamental right does not infringe the right.”⁸⁴Id. at 193. Thus, it would seem that the fundamental constitutional rights argument cannot explain the Court’s new anti-religious-discrimination doctrine. The Court’s government speech precedents directly conflict with today’s Court’s characterization of a failure to fund religious education as a “penalty” imposed on that religion.⁸⁵Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2255 (2020).

VI.  THE RADICAL TRINITY

If a jurisprudential entrepreneur were on the lookout for an ideal test case to sell a radical reformulation of the Court’s approach to the religion clauses, the facts of Trinity Lutheran Church of Columbia v. Comer would certainly fit the bill. On the surface, this case about the re-surfacing of children’s playgrounds in Missouri involved a highly sympathetic petitioner and addressed relatively un-weighty issues of church and state. To promote recycling and benefit children in low income areas, the state government allocated funds on a competitive basis to help nonprofit daycare centers replace older, harder playground surfaces with ones made from recycled tires.⁸⁶Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 454–55 (2017). Unfortunately for Trinity Lutheran Church, it discovered that its preschool and daycare center were ineligible.⁸⁷Id. Article I, Section 7 of the Missouri Constitution provided that “no money shall ever be taken from the public treasury, directly or indirectly, in aid of any church, sect or denomination of religion.”⁸⁸Id at 455. Missouri categorically disqualified religious organizations from receiving grants under the program.⁸⁹Id.

Although the District Court did not mention the government speech doctrine by name, it upheld the Missouri program using reasoning consistent with the doctrine’s underlying principles. It drew an analogy to a case upholding a state’s “mere” choice not to fund a particular “category of instruction,”⁹⁰Trinity Lutheran Church of Columbia, Inc. v. Pauley, 976 F.Supp.2d 1137, 1148 (W.D. Mo. 2013). suggesting that it was within Missouri’s discretion to determine the scope of its programs. This includes the choice not to subsidize playgrounds run by religious institutions with public money. In concisely rejecting a free expression argument, the District Court dismissed any notion that the program was designed as an “open forum” for speech.⁹¹Id. at 1157.

Consistent with the pro-discrimination implications of the religion clauses, it pointed to the state’s “antiestablishment” interests in preventing religious organizations from receiving government funds.⁹²Id. at 1148. Even if Missouri was not required to promote this interest to the extent it did––prohibiting any receipt of funds by religious organizations––significant “play in the joints” exists between what is prohibited by the Establishment Clause and what is required by Free Exercise.⁹³Id. at 1147. The District Court reasoned that Missouri’s more robust prohibition (what we might certainly call “discrimination” against religion), supports the antiestablishment values built into the religion clauses.⁹⁴Id. at 1148. Indeed, according to the Court, it would be patently “illogical” to presume that a choice not to fund religion to avoid potential entanglement with government necessarily reflects a hostility toward religion.⁹⁵Id. The District Court emphasized that the grant here would be paid directly to the religious organization, making the antiestablishment concerns even more compelling than programs designed to sever the direct link between government aid and religious institutions by putting the choice to spend in the hands of private individuals.⁹⁶Id. at 1152.

The Eighth Circuit affirmed the District Court decision, characterizing the appellant as “seek[ing] an unprecedented ruling—that a state constitution violates the First Amendment . . . if it bars the grant of public funds to a church.”⁹⁷Trinity Lutheran Church of Columbia, Inc. v. Pauley, 788 F.3d 779, 783 (8th Cir. 2015). In no uncertain terms, it rejected the notion that a state could be compelled to provide taxpayer funds directly to a church: “No Supreme Court case” it explained, “has granted such relief.”⁹⁸Id. at 784. Moving to an approach in which every generally available public benefit becomes a baseline in which we might scrutinize the denial of comparable benefits to religious actors, would, according to the Circuit Court, constitute “a logical constitutional leap.”⁹⁹Id. at 785. It would fundamentally recast the Free Exercise Clause from a provision that demands religion be treated differently, to one that prohibits discrimination against it. It would require a repudiation of decades of precedent, and of our foundational understanding of how the religion clauses were to function. In blunt terms, the Circuit Court conceded that “only the Supreme Court can make that leap.”¹⁰⁰Id.

But the Supreme Court had indeed changed. Beginning with this unassuming little case about playground surfaces, it was poised to make just such an unprecedented and radical shift in its religion clause jurisprudence. Granted, Chief Justice Roberts, in his majority opinion that overruled the Eighth Circuit in Trinity Lutheran, did not frame his decision in this way. Roberts has developed a reputation for strategic incrementalism, in which the seeds of what will eventually blossom into highly consequential doctrinal change are planted in unassuming soil.¹⁰¹Linda Greenhouse, Justice on the Brink: The Death of Ruth Bader Ginsburg, the Rise of Amy Coney Barrett, and Twelve Months that Transformed the Supreme Court 219 (2021). However, the Trinity Lutheran dissenters did not mince words. Emphasizing the high-stakes of this seemingly low-stakes decision, Justice Sotomayor tells us that “[t]his case is about nothing less than the relationship between religious institutions and the civil government . . . [t]he Court today profoundly changes that relationship.”¹⁰²Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 471–72 (2017) (Sotomayor, J., dissenting).

Roberts’s analysis begins by setting the stage for the Court’s new Free Exercise non-discrimination principle. He cites as a broad rule the rationale of a narrow Free Exercise decision that happened to involve targeted discrimination against a particular religious sect. Granted, the language in the 1993 case Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah¹⁰³Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520 (1993).  gave Roberts a good deal to work with. Although the decision was centrally about, as Justice Kennedy explained in the second sentence of the opinion, the “fundamental nonpersecution principle of the First Amendment,”¹⁰⁴Id. at 523. it was peppered with the ominous suggestion that impermissible religious discrimination was afoot. However, there is no reason to conclude that the mere relevance of discrimination in this case would convert the religion clauses into a general anti-discrimination rule. Here discrimination simply served as evidence that this particular law should be understood as an unconstitutional prohibition of the free exercise of religion. Like the neutrality principle discussed above, the discriminatory nature of the law was highlighted to demonstrate that things were not as they seemed; a law that may have appeared neutral on its face, was in fact targeting a particular religion’s practices, and thus, quite literally, prohibiting “free exercise” of that religion.

The dilemma with the religion clauses, as with free speech, is that there will necessarily be a vast number of laws aimed at addressing a wide range of social ills that have the subsidiary effect of in-part “prohibiting” the free exercise of particular religions (or “abridging” expressive activity). And the Court has never taken the position, for understandable reasons, that all such laws are unenforceable as to religious practitioners (or to those whose actions are, in part, “expressive”). As Justice Scalia opined, in a country of vast religious diversity, adopting a rule that would strictly scrutinize any neutral, generally applicable law that somehow could be said to intrude on a religious practice would be “courting anarchy.”¹⁰⁵Emp. Div. v. Smith, 494 U.S. 872, 888 (1990). The doctrinal parameters of whether, and when, a religious exemption may be required under such circumstances continue to evolve. However, it is clear that laws advancing legitimate, non-religion-related policy ends that incidentally impact the free exercise of certain religious actors are not automatically deemed constitutionally suspect.

No doubt, in drafting the First Amendment the framers sought to prohibit the type of targeted religious persecution that was all too common in the old world.¹⁰⁶Babalu, 508 U.S. at 532. But again, the concern was that government not prohibit free exercise through persecution, not that it refrain from treating religion differently from other subjects (something that it is required to do under a straight-forward reading of the text of the First Amendment). Government persecution might be achieved through direct measures that leave little ambiguity as to the intended objective. However, a government intent on punishing, stigmatizing, or driving away an unpopular religious minority might also use non-religion-related policy justifications as a pretext for doing so. It may craft laws that are intended to impede the practices of certain religious believers but justify those laws on legitimate non-religion-related public policy grounds. Or, a legislature might truly have mixed motives. Determining whether or not there has been a free exercise violation under such circumstances may prove difficult. Thus, in this context, identifying “discrimination” may become a vital tool in sussing out whether intentional religious suppression, or a mere side effect of an unrelated policy goal, is occurring.

Preventing animal cruelty was the stated policy goal in Church of Lukumi Babalu. Upon investigation however, this facially legitimate objective was found to have been a front for religious animus. The case involved four ordinances in the south Florida city of Hialeah. Together, they prohibited certain forms of animal sacrifice, a practice associated with the Santeria religion.¹⁰⁷Id. at 524–28. The ordinances were apparently spurred on by the imminent prospect of a Santeria church opening in Hialeah and the hostility and discomfort many residents and city council members held toward Santeria and its traditional practices.¹⁰⁸Id. at 541–42.

The city argued that its ban on animal sacrifice was justifiable on non-religious grounds. It cited not just protecting animals from cruel treatment, as mentioned above, but also the health risks involved, the emotional injury to children that might result from witnessing such killings, and the interest in restricting slaughter to particular areas of the city.¹⁰⁹Id. at 529–30. The narrow ban however, was carefully crafted to exclude virtually all animal killing other than religious sacrifice, and even within this category it exempted kosher slaughter.¹¹⁰Id. at 535–36. The Court concluded that “Santeria alone was the exclusive legislative concern. . . . [K]illings that are no more necessary or humane in almost all other circumstances are unpunished.”¹¹¹Id. at 536. This was, as Justice Souter pointed out in his concurrence, “a rare example of a law actually aimed at suppressing religious exercise.”¹¹²Id. at 564 (Souter, J., concurring).

The Court unanimously struck down the ordinances as a violation of the Free Exercise Clause.¹¹³Id. at 546. It was from this unexceptional holding in Church of Lukumi Babalu—prohibiting a legal ban directly targeting practices that were a clear element of the sect’s religious exercise—that the Court in Trinity Lutheran extracts from the Free Exercise Clause a strikingly broad anti- religious-discrimination rule. The new rule requires taxpayer money be used to facilitate the religious mission of an organization—that is, if such funds are available to secular organizations.

Granted, the Church of Lukumi Babalu Court identified, through a close examination of the text of the ordinances at issue and the broader social context, blatantly discriminatory treatment targeting particular practices of a particular religious group. And at times, Kennedy used language to emphasize the significance of such unequal treatment, for example, when he stated that “[a]t a minimum, the protections of the Free Exercise Clause pertain if the law at issue discriminates against some or all religious beliefs or regulates or prohibits conduct because it is undertaken for religious reasons.”¹¹⁴Id. at 532. In this context, this observation simply points out that a restriction on free exercise that is specifically directed toward a particular religion or religious practice is a First Amendment red flag. Such a law presents a sharp contrast to generally applicable laws that affect, and are directed toward, religious and non-religious actors alike. The fact of “discrimination,” in other words, helps courts home in on the most egregious and likely unconstitutional prohibitions on free exercise. Nothing in the decision, however, would suggest that it is the “discrimination” that is the free exercise offense, nor that unconstitutional “discrimination” should be interpreted to encompass a mere choice by a government not to provide financial support to particular religious organizations.

Indeed, Roberts’s reliance upon Church of Lukumi Babalu is particularly curious considering that it was issued just two years after Rust v. Sullivan. As discussed above, this is the seminal government speech case in which the Court explicitly affirmed a government’s power to discriminate—to be selective and make substantive distinctions as to the programs it chooses to fund or not fund.¹¹⁵See supra Part V. What is Chief Justice Roberts’s response to this apparent contradiction? He tells us that “Trinity Lutheran is not claiming any entitlement to a subsidy. It is asserting a right to participate in a government benefit program without having to disavow its religious character.”¹¹⁶Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 451 (2017).

But how is claiming a right to receive government largess by “participat[ing] in a government benefit program” that one is not qualified to participate in, anything but an assertion of an “entitlement to a subsidy?”¹¹⁷Id. The Chief Justice’s artful reframing and rephrasing of Trinity Lutheran’s argument does not alter the fundamental facts. After this decision the government in Missouri is required to use taxpayer money to subsidize what on policy grounds it does not wish to subsidize. The Chief’s attempt to sugarcoat its radical decision notwithstanding, the unelected Supreme Court is telling an elected government how it must legislate and allocate its resources—a command that is in direct conflict with its own government speech doctrine.

The only other ostensibly on-point case cited by the Trinity Lutheran Court as support for its innovative religion clause non-discrimination rule was the 1978 plurality opinion in McDaniel v. Paty.¹¹⁸McDaniel v. Paty, 435 U.S. 618 (1978). Under the Tennessee Constitution, clergy were disqualified from serving as state legislators, and thereby not permitted to serve as delegates to a state constitutional convention.¹¹⁹Id. at 620–21. The Supreme Court struck down the exclusion on Free Exercise grounds. The plurality explained that this exclusion of ministers from state legislatures was a practice that was implemented in seven of the original thirteen States. It was instituted “primarily to assure the success of a new political experiment, the separation of church and state.”¹²⁰Id. at 622.

However, the notion that clergy members should ipso facto be excluded from legislative positions remained controversial. This was so despite the fact that the First Amendment did not at the time apply to the states (it would not be explicitly incorporated until well after the ratification of the Fourteenth Amendment in 1868). Even James Madison, “the greatest advocate for the separation of state and church” ^{121Andrew L. Seidel, The Founding Myth: Why Christian Nationalism Is Un-American 37 (2019).} and primary drafter of the Constitution’s religion clauses suggested (in contrast with Thomas Jefferson’s initial position) that disqualification resembled a kind of unjust punishment reserved for those who happened to choose religious professions.¹²²McDaniel, 435 U.S. at 624. To Madison, the exclusion itself might even constitute a breach of the church-state separation, in that religion was to be exempted “from the cognizance of Civil power.”¹²³Id. at 624. One can thus see the parallel Roberts was attempting to draw with Trinity Lutheran—a law that was arguably “punishing” a playground operator, denying it the opportunity to benefit from a recycled tire resurfacing program, merely due to its religious affiliation.

However, with the help of the government speech doctrine, the distinction between Trinity Lutheran and McDaniel becomes immediately clear. A policy choice as to how the government will use taxpayer dollars—what kinds of interests or schools or playgrounds it will support—is fundamentally different from a law that makes distinctions as to who may legislate in the first place. The former represents a choice as to the policy message the government will communicate, a democratic imperative; the latter represents a choice to exclude certain voices from the possibility of being a part of that government, an anti-democratic exclusion. To suggest that the right to run for office in a democracy is a mere government “benefit” comparable to a government program that helps fund playground resurfacing¹²⁴Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 462 (2017). is to demean a core element of representative democracy. It conflates the ability to select a representative with the naturally selective product of representative democracy; it degrades them both by suggesting that a democracy-affirming Court intervention to prevent limitations on who we may choose as a representative is somehow analogous to a democracy-inhibiting limitation on a government to make policy choices.

McDaniel was also grounded in an individual right to practice one’s religion. The Court explained that “the right to the free exercise of religion unquestionably encompasses the right to preach, proselyte, and perform other similar religious functions, or, in other words, to be a minister of the type McDaniel was found to be.”¹²⁵McDaniel, 435 U.S. at 626. McDaniel’s right to free exercise was being conditioned upon his surrender of democratic political participation, the choice to run for office. His desire to serve as a delegate to a state constitutional convention was not a request to have the state subsidize his religious activity, except to the extent than any government employee’s private activities might be said to be subsidized by a state salary.

Trinity Lutheran in contrast, involved not an individual’s rights, but the rights of a collective entity. It described its Child Learning Center’s mission as “provid[ing] a safe, clean, and attractive school facility in conjunction with an educational program structured to allow a child to grow spiritually.”¹²⁶Trinity Lutheran, 582 U.S. at 455. Trinity Lutheran, in other words, was seeking state tax dollars to advance its religious goals as a collective entity. The loss by Trinity Lutheran of the opportunity to participate in a subsidized playground surface program was nothing like the Hobson’s choice that confronted McDaniel. He was not seeking support from the government for his religious works. For McDaniel, under the Tennessee law he was forced to either forfeit his right to fully participate as a citizen or refrain from free religious exercise. 

Chief Justice Roberts finds commonality in McDaniel and Trinity Lutheran, emphasizing the status-based nature of the discrimination in both cases.¹²⁷Id. at 459. He characterized the policy in Missouri as “expressly discriminat[ing] against otherwise eligible recipients by disqualifying them from a public benefit solely because of their religious character.”¹²⁸Id. at 462. The McDaniel Court similarly stressed the unique way the law in Tennessee disqualified the petitioner from office “because of his status as a ‘minister’ or ‘priest.’ ”¹²⁹McDaniel, 435 U.S. at 627. And indeed, the Court has in recent years frequently conflated the individual and the collective; but there can be good reason to acknowledge the differences between the two.

At the heart of classical liberalism is a respect for the individual. The notion that status-based individual deprivations are particularly repugnant is found in many parts of the Constitution itself—whether it is the prohibition on Bills of Attainder,¹³⁰U.S. Const. art. I, § 9, cl. 3. the demand that “no religious Test shall ever be required as a Qualification to any Office or public Trust under the United States,”¹³¹U.S. Const. art. VI. or that the right to vote shall not be denied “on account of race, color, or previous condition of servitude”¹³²U.S. Const. amend. XV. in the Fifteenth Amendment. Although the Court has extended many individual rights in the Constitution to collective entities, there is reason to be skeptical that the same set of concerns applies here.

Tennessee justified its disqualification of a certain category of individuals from elective office on the basis of the “leadership role” and “full time” promotion of “religious objectives” of those who choose to be ministers and priests.¹³³McDaniel, 435 U.S. at 634–35. Citing its goal of maintaining the separation of church and state, the state emphasized its concern that the religious commitments of ministers and priests would at times interfere with their duties as a state legislator.¹³⁴Id. at 645. Implicit in the plurality decision rejecting this rationale is the understanding that human beings are more than just their chosen avocation. A “unique disability” imposed on an individual because they “exhibit a defined level of intensity of involvement in protected religious activity”¹³⁵Id. at 632. is, quite simply, highly distinguishable from differential treatment of legal entities based upon their respective, narrowly defined legal purpose.

Nonetheless, the Trinity Lutheran Court finds the organization’s status-based disqualification from the recycled tire playground surface program to be relevant, and sufficiently analogous to the disqualification from office faced by McDaniel. As a result, the Court found Trinity Lutheran merited a similar legal outcome. The Court’s focus on the status-based nature of the religious discrimination at issue also served to distinguish Trinity Lutheran from the 2004 decision Locke v. Davey, the seemingly on-point precedent discussed previously in which the Supreme Court reached the opposite conclusion.

In Locke the Supreme Court upheld a scholarship program in Washington State that, although available for a full range of postsecondary education degrees, stipulated funds could not be used by students “pursuing a degree in devotional theology.”¹³⁶Locke v. Davey, 540 U.S. 712, 715 (2004). Roberts reasoned that in Locke the student was not denied the benefit of the program on the basis of his religious status, as was true of Trinity Lutheran, but “because of what he proposed to do—use the funds to prepare for the ministry.”¹³⁷Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 464 (2017). Thus, for the Trinity Lutheran Court, the distinction between religious discrimination based on religious “status” and religious “use” appeared to be determinative.

The silver lining of deciding to have the opinion turn on this questionable analogy between the status-based discrimination against the individual minister in McDaniel and the collective religious institution in Trinity Lutheran, is that it established a rule that would, in theory, still allow for government to make crucial policy distinctions consistent with the government speech doctrine. As long as the government is not declining to spend on the basis of religious status, a government might still decline to draw on finite state resources to fund religious action. A government might conclude, for example, that spending on such religious “use” would be unwise, have benefits that are unsupported by evidence, reflect objectives inconsistent with the state’s current policy goals, or simply on balance represent a less weighty spending priority than other competing governmental aims.

This “status” versus “use” distinction, however, would not have staying power. Locke would ultimately be narrowed dramatically, largely relegated to doctrinal irrelevance. In Trinity Lutheran the status/use test was thrown into question in a concurrence by Justices Gorsuch and Thomas. Gorsuch, foreshadowing the Court’s eventual path in Carson, would have distinguished the contradictory outcome in Locke on the basis of its narrow exclusion of scholarship funds for devotional theology and the “long tradition against the use of public funds for training of the clergy.”¹³⁸Id. at 470 (Gorsuch, J., concurring). For Gorsuch, not only was the status/use distinction likely to be difficult to apply in practice, but it was also irrelevant for the purposes of First Amendment free exercise. The reason? To Gorsuch, “that Clause guarantees the free exercise of religion, not just the right to inward belief.”¹³⁹Id. at 469.

But this is clearly incorrect. The language of the Free Exercise Clause does suggest a “guarantee.” It no more “guarantees” free exercise than the Free Speech Clause “guarantees” free speech or the Second Amendment “guarantees” that each citizen will be supplied with her own private arsenal. It merely prevents the state from interfering with or “prohibiting,” such freedom. Free exercise of religion may be hampered by friends or family, a wide range of private actors, or the free market itself. Practicing one’s religion may be time consuming, expensive, embarrassing, or stigmatizing. Indeed, it is precisely this kind of interpretive line-blurring of the Free Exercise Clause by Gorsuch that the government speech doctrine rejected when it came to the Free Speech Clause. One is not “guaranteed” an equal opportunity to have the government promote your message of x just because it has chosen to run a public service announcement promoting y.

VII.  ESPINOZA AND THE TRINITY LUTHERAN AFTERMATH

Just three years later in Espinoza v. Montana Department of Revenue the Court broadened the applicability of this fallacious reading. The Montana Constitution included a provision that barred government aid to religious schools.¹⁴⁰Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2251 (2020). Under this “no-aid” provision that the State’s Supreme Court had rejected, a private school tuition assistance program that would have granted “a tax credit to anyone who donates to certain organizations that in turn award scholarships to selected students.”¹⁴¹Id. In Espinoza, building on the newly invented anti-religious discrimination principle, the U.S. Supreme Court struck down this provision in the Montana Constitution.

Like Trinity Lutheran, it homed in on the status/use distinction to explain why the analogous Locke holding should not apply.¹⁴²Id. at 2255–57. The Court emphasized that although both Espinoza and Locke addressed government scholarship funds used for religious education, the Montana Constitution prohibited all aid to sectarian schools simply by virtue of their being religious (that is, status) whereas the program in Locke excluded, specifically, just religious training (that is, use).¹⁴³Id. at 2257. This case, the Court explained, “turns expressly on religious status and not religious use.”¹⁴⁴Id. at 2256. It even took the time to refute claims that Montana’s Constitution was in fact about preventing “use” for religious education, responding that “[s]tatus-based discrimination remains status based even if one of its goals or effects is preventing religious organizations from putting aid to religious uses.”¹⁴⁵Id. It asserted that “status-based discrimination is subject to ‘the strictest scrutiny.’ ”¹⁴⁶Id. at 2257. Thus, a reasonable reading of the Court’s opinion would conclude that the status versus use distinction was central to this doctrine.

At the same time that it repeatedly emphasized its significance, however, the Court seemed to be readying itself to discard this distinction in the near future. It provided the caveat that “[n]one of this is meant to suggest that we agree . . . that some lesser degree of scrutiny applies to discrimination against religious uses of government aid.”¹⁴⁷Id. Why then raise this distinction in the first place? As mentioned earlier, Trinity Lutheran was framed as a narrow decision addressing an even narrower, idiosyncratic, and relatively low-stakes set of facts. Allaying fears that it was anything broader than this, Trinity Lutheran’s footnote three had read: “This case involves express discrimination based on religious identity with respect to playground resurfacing. We do not address religious uses of funding or other forms of discrimination.”¹⁴⁸Trinity Lutheran Church of Columbia, Inc. v. Comer, 582 U.S. 449, 465 n.3 (2017). Reliance on this status/use distinction, as well as the inclusion of this qualifying footnote, likely contributed to a majority that was able to bring along two justices (Breyer and Kagan) who shortly thereafter would pull away, dissenting in Espinoza and Carson.

Once the critical break with the religion clause precedent was achieved, like Lucy and Charlie Brown, the Chief Justice quickly pulled that football. It turns out Trinity Lutheran was no minor decision at all. In Carson, decided two years after Espinoza, the Court was clear that it was in fact Locke that was the minor decision. Leaving little ambiguity, Roberts asserted that “Locke cannot be read beyond its narrow focus on vocational religious degrees.”¹⁴⁹Carson v. Makin, 142 S. Ct. 1987, 2002 (2022). Thus, just a short five-year time span had passed between Trinity Lutheran—adopting the status/use device as a central means of justifying its jarring divergence from Locke—and Carson—effectively retracting it. The unfortunate implication is that the status/use distinction served merely as a short-term results-oriented expedient—the proverbial camel’s nose that could push its way, ever so slightly, under the tent—facilitating the Court’s radical transformation of the Religion Clauses. 

In Espinoza, the Court repeatedly stressed the completely inapposite, but rhetorically powerful pejorative conception of “discrimination” to justify its holding, explaining that the Constitution “condemns discrimination against religious schools and the families whose children attend them.”¹⁵⁰Espinoza, 140 S. Ct. at 2262. But even more than Trinity Lutheran, both Espinoza and Carson address a species of governmental action that is inevitably, and necessarily, grounded in discrimination—the state’s choices about education. It is indeed difficult to imagine a more consequential sphere of government speech than the fine-grained discretion involved when a democratically elected government chooses the ideas, ideals, knowledge, and values to impart to future generations. No question, this is most apparent in the field of public education, where states and localities are in the position of determining every last detail of a curriculum. But, unless it is establishing an open public forum, there is no reason to believe that it is less relevant when a state decides which educational alternatives it will choose to subsidize, and which it will not. Such choices are a direct manifestation of the will of the people as exercised by their elected representatives.

Indeed, the only limitation on this foundational majoritarian precept that it is “the people” who decide (indirectly, through elections) on the substance of public education and private educational subsidies, is when it is overridden by the Constitution itself, which, of course, requires a supermajority to overrule.¹⁵¹See Epperson v. Arkansas, 393 U.S. 97, 107 (1968). And one of the most notable examples of this can be found in the requirement of religious discrimination—that religion is subject to differential treatment—in the First Amendment. This requirement of religious discrimination in public education is well established in Court precedent.

In Epperson v. Arkansas, the Court confirmed that the religion clauses carve out an exception to the general and broad discretion a state has over its schools’ curricula.¹⁵²Id. at 104–05. Under Arkansas law, public schools were prohibited from “teach[ing] the theory or doctrine that mankind ascended or descended from a lower order of animals.”¹⁵³Id. at 98–99. The clear motivation behind the law was to thwart teaching that conflicted with the biblical account of the origin of life.¹⁵⁴Id. at 109. Although the Court expressed a general reluctance to involve the judiciary in questions of educational policy, it was unequivocal that “the First Amendment does not permit the State to require that teaching and learning must be tailored to the principles or prohibitions of any religious sect or dogma.”¹⁵⁵Id. at 106. The Court reaffirmed this reading in the 1987 decision Edwards v. Aguillard.¹⁵⁶Edwards v. Aguillard, 482 U.S. 578, 594 (1987). This well-established understanding of the religion clauses, that educational choices which are otherwise within the discretion of state and local government must be judicially curtailed due to their religious nature, was not just contradicted, but inverted by Espinoza and Carson. The problem with “Montana’s no-aid provision” explains the Espinoza majority, is that it “bars religious schools from public benefits solely because of the religious character of the schools.”¹⁵⁷Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2255 (2020).

Indeed, not only is the Court converting a constitutional principle that has always required differential treatment of religion into an anti-religious-discrimination rule, but government inaction—not doing what it was formerly required not to do by a conventional reading of the First Amendment—is understood as potentially coercive. As the Court explains, “[t]he Free Exercise Clause protects against even ‘indirect coercion,’ and a State ‘punishe[s] the free exercise of religion’ by disqualifying the religious from government aid . . . .”¹⁵⁸Id. at 2256. Roberts, in other words, is taking Scalia’s Locke dissent logic one step further: not providing a government benefit is not just a relative “burden” on religion, it is a coercive punishment. Government benefits are so alluring that Jefferson’s separation of church and state is itself unconstitutional. The wall of separation is coercive because the church on one side will see the bag of goodies on the other side and feel compelled to un-church itself––to shed its religious identity so it too can get a hold of those benefits.

VIII.  THE ASYMMETRIC AND INTERDEPENDENT RELIGION CLAUSES

The Alice in Wonderland feel of the Court’s logic may be dizzying. But it is the built-in tension between the two religion clauses that makes the Court’s startling logical backflips possible. The Court is effectively borrowing concepts culled from one side of its religion clause decisions and lending them to the other. Since the two clauses were designed to pull in two different directions and operate in fundamentally different ways, predictably, the results are perverse.

While “neutrality” is drawn from Everson, “coercion” can be found in decisions such as 1992’s Lee v. Weisman. In that case, a student made an Establishment Clause challenge to a public school practice of inviting clergy members to give nondenominational prayers at graduation ceremonies. Although a passionate concurrence by Justices Blackmun, Stevens, and O’Connor argued for a more robust separationist rationale, the majority nonetheless struck down the policy, asserting that “at a minimum, the Constitution guarantees that government may not coerce anyone to support or participate in religion or its exercise.”¹⁵⁹Lee v. Weisman, 505 U.S. 577, 587 (1992). Students, in other words, would feel peer pressure to conform to, and perhaps participate in, the religious exercise. This anti-coercion principle was firmly rooted in the Court’s Establishment Clause jurisprudence; the Court’s sights were set on identifying those types of government actions that cross the unconstitutional line of “respecting an establishment of religion.”

The Lee Court acknowledged that attendance at the ceremony was technically voluntary, but in the eyes of most students, it was a crucial rite of passage.¹⁶⁰Id. at 594–95. The Court explained that “[i]t is a tenet of the First Amendment that the State cannot require one of its citizens to forfeit his or her rights and benefits as the price of resisting conformance to state-sponsored religious practice.”¹⁶¹Id. at 596. The Court in Espinoza and Carson takes this Establishment Clause principle, and applies it as if it were about free exercise. This is a mistake. These two clauses may work in tandem, but they function differently, as their disparate textual construction clearly suggests. The latter simply prevents the government from actively interfering with or “prohibiting” religious practice, whereas the former involves the thornier question of what it may mean for a law to “respect” an establishment of religion. As constitutional historian Leonard Levy explains, “Congress can pass laws regulating and even abridging the free exercise of religion without prohibiting it altogether.”¹⁶²Levy, supra note 7. And not only does the Court, with little theoretical justification, blithely transfer an Establishment Clause test to a free exercise issue, it quietly alters its relative rigor.

As this concept of “coercion” is understood to be ever more capacious on the Free Exercise side of the ledger, including the “indirect” coercion of merely not having one’s religiously informed policy preferences fulfilled, the meaning of Establishment Clause coercion gets appreciably narrower. In Kennedy v. Bremerton School District, decided just one week after Carson, the Court appeared untroubled by establishment concerns because there was “no evidence” that, during a public prayer by an influential school employee at a public school event, “students [were] directly coerced to pray with [the coach].”¹⁶³Kennedy v. Bremerton Sch. Dist. 142 S. Ct. 2407, 2419 (2022) (emphasis added). Thus, in the free exercise context, it would appear that a highly tenuous, and certainly debatable “indirect” form of coercion is sufficient to impose a constitutional demand that taxpayer money be used to fund private religion. At the same time, a popular football coach publicly praying “under the bright lights” of a stadium full of spectators,¹⁶⁴Id. at 2439 (Sotomayor, J., dissenting). while “on duty,”¹⁶⁵Id. at 2437 (Sotomayor, J., dissenting). and implicitly inviting student participation, was not a “direct” enough form of coercion to constitute an Establishment Clause violation. This coach had “made multiple media appearances to publicize his plans to pray at the 50-yard line,”¹⁶⁶Id. at 2437 (Sotomayor, J., dissenting). and was someone from whom students might naturally seek favorable treatment such as extra playing time and recommendation letters¹⁶⁷Id. at 2443 (Sotomayor, J., dissenting).. . Justice Sotomayor, in dissent, characterizes this newly watered down establishment test as “a nearly toothless version of the coercion analysis.”¹⁶⁸Id. at 2434 (Sotomayor, J., dissenting). The effect is to invert the very meaning of the religion clauses, taking what would have been an unconstitutional violation of the Establishment Clause under the Court’s precedents—the injection of religion into the public schools—and transforming it into a constitutional requirement under the Free Exercise Clause.¹⁶⁹Id. at 2441 (Sotomayor, J., dissenting).

Considering the ubiquity of both law and religion, and the fact that most policy will interact with religion in a multitude of ways, the task of drawing the establishment line is arguably much more difficult and subtle than drawing the free exercise line. On its face, the text of the Free Exercise Clause—a simple ban on governments prohibiting the free exercise of religion—would not seem to support a reading that demands active promotion by government of religion to preempt indirect coercion of religious believers who might feel left out. On its face, the Free Exercise Clause requires answering just two questions: First, how is a particular religion practiced, or exercised? Second, does the law at issue in fact prohibit that religion or its individual practitioners from practicing in such manner? The text of the Establishment Clause, in contrast, suggests that any state activity associated with, or part of a regime of, government establishment, should be subject to judicial scrutiny. The word “respecting” gives the Establishment Clause a degree of play that that the word “prohibiting” in the Free Exercise Clause does not.

As with all constitutional language, textual analysis allows for a range of plausible interpretations; the meaning given to both the word “prohibiting” and “respecting” is not fixed and will naturally be context dependent. As Randy Barnett explains, “[a]lthough most words are potentially vague, we do not face a problem of vagueness until a word needs to be applied to an object that may or may not fall within its penumbra.”¹⁷⁰Randy E. Barnett, Interpretation and Construction, 34 Harv. J.L. & Pub. Pol’y 65, 68–69 (2011). The Janus-faced nature of the religion clauses—pushing in two different directions at the same time—heightens the interpretive challenge. Any doctrinal test by the Court that attempts to put flesh on the bones of the purportedly vague language in one religion clause, what Barnett refers to as a process of constitutional “construction,”¹⁷¹Id. at 69. must remain cognizant of its potential interaction with, impact on, or inconsistency with, the other clause. The Court’s insight of a “play in the joints”—a necessary degree of governmental discretion in enacting policies that promote the principles of one clause without violating the other—is consistent with this penumbral overlap.

Nonetheless, under many factual circumstances the same test simply cannot apply simultaneously under both the Establishment Clause and Free Exercise Clause without producing irreconcilable outcomes. The coercion test, so casually transferred from establishment to free exercise in Espinoza provides an example. The prayer in Lee is a violation of the Establishment Clause’s anti-coercion principle, but under the logic of Espinoza a constitutionally repaired, prayer-free graduation ceremony would be unconstitutionally coercive to religious students under the Free Exercise Clause by depriving them of a government benefit available to secular students. A free exercise anti-coercion rule would suggest that due to this deprivation, religious students would be indirectly coerced to either give up the benefit of publicly funded education and pay to attend a private religious school or relinquish their ability to partake in a religious graduation ceremony.

James Madison emphasized the importance of separation for the good of both government and religion, seeing it as a way of “[guarding against a] tendency to a usurpation on one side or the other, or to a corrupting coalition or alliance between them . . . .”¹⁷²Geoffrey R. Stone, Louis Michael Seidman, Cass R. Sunstein, Mark V. Tushnet & Pamela S. Karlan, Constitutional Law 1438 (8th ed. 2018) (quoting James Madison). Roger Williams focused primarily on the way separation protects the church from control by the state.¹⁷³Id. Yet, the majority in Espinoza dismisses this concern in just a few short paragraphs. Inverting historical reality, it treats Montana’s claim that “the no-aid provision promotes religious freedom”¹⁷⁴Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2261 (2020). as the novel view, and its own recent invention of the anti-discrimination religion clauses as the constitutional baseline.

Consistent with an understanding that extends back hundreds of years, the state argued that “the no-aid provision protects the religious liberty of taxpayers by ensuring that their taxes are not directed to religious organizations, and it safeguards the freedom of religious organizations by keeping the government out of their operations.”¹⁷⁵Id. at 2260. As if this deeply-rooted Madisonian understanding were a fringe perspective, the Court dismissed allowing an “infringement of First Amendment rights” on the basis of what it characterized as “a State’s alternative view.”¹⁷⁶Id. But this is no “alternative view.” The dangers of the politicization of religion, the resentments taxpayer funding of religious institutions may engender, and the pressure governmental oversight and regulation would naturally place on the church, were not lost on the founders.

Effectively dismissing this wisdom in a single paragraph, the Court justifies its decision by emphasizing how its prior cases have allowed programs that provide aid to religious organizations where “attenuated by private choices.”¹⁷⁷Id. at 2261. It then goes on to conflate freedom from government interference—these “private choices” that are rightfully protected under the Free Exercise Clause—and a right to non-discriminatory government benefits—which is, to the contrary, in direct tension with a traditional understanding and reading of the religion clauses. It achieves this slight-of-hand by citing for support its precedents that have “long recognized the rights of parents to direct ‘the religious upbringing’ of their children.”¹⁷⁸Id. Of course, the freedom to opt-out of a majoritarian government program never implied a right to demand that the government offer an alternative version of that program that is tailored to one’s particular tastes.

Yet, in Carson v. Makin this is precisely what the Court requires of the state of Maine. The program at issue there, as discussed previously, differed from Espinoza in that it had limited its applicability based on the substance of the educational content of a school rather than its religious status. The Maine tuition assistance program was available to parents wishing to send their children to private schools in sparsely populated areas of the state where local government does not operate its own secondary school. Funds were ineligible however, if the desired school “promotes a particular faith and presents academic material through . . . that faith.”¹⁷⁹Carson v. Makin, 142 S. Ct. 1987, 2001 (2022). The state explained that the private school option was designed to offer a “rough equivalent” of the secular public schools available in more populous parts of the state. The Carson family, however, wanted to send their daughter to a private school with a “Christian worldview [that] aligns with their sincerely held religious beliefs.”¹⁸⁰Id. at 1994. Under the Court’s new anti-religious discrimination reading, the state was now required to use taxpayer funds to accommodate the family’s religious tastes.

IX.  THE DEMISE OF THE STATUS/USE DISTINCTION

Unless a majoritarian democracy is structured to require unanimity, it is inescapable that some minority of the population will be unhappy with the substantive policy choices the government makes. As Alexander Tsesis has pointed out, “[there are] disagreements about the wisdom of myriad government programs, policies, statutes, and priorities.”¹⁸¹Alexander Tsesis, Government Speech and the Establishment Clause, 2022 U. Ill. L. Rev. 1761, 1771 (2022). A distinct policy choice to fund only private schools with an evidence-based curriculum, is, of course, bound to displease those who prefer a faith-based approach to education. However, a state may have many legitimate policy reasons for declining to fund religious education, and these reasons may be independent of a desire to adhere to a “stricter separation of church and state than the Federal Constitution requires.”¹⁸²Carson, 142 S. Ct. at 1997. Most obviously, a government may conclude that an epistemological approach grounded in faith is in tension with a commitment to the scientific method. Its reasoning, in other words, may relate directly to its judgment as to how it will best fulfill its educational mission. The Court acknowledges that only private schools that “meet certain basic requirements” were eligible to receive the funds under Maine’s program.¹⁸³Id. at 1993. Yet, somehow, four pages later, the Court characterizes it as “a neutral benefit program,” seemingly forgetting that the state established detailed criteria laying out just what attributes schools must have if it is to fund them.¹⁸⁴Id. at 1997.

With Carson, the Court thus ratchets up its novel anti-religious-discrimination interpretation of the religion clauses to include substantive as well as status-based distinctions. As the Court explains, “the prohibition on status-based discrimination under the Free Exercise Clause is not a permission to engage in use-based discrimination.”¹⁸⁵Id. at 2001. The former was at least arguably one step further removed from the kind of policy discretion essential for responsive democratic judgment––a discretion that informs the Court’s own government speech doctrine. In theory, status-based distinctions are also potentially indicative of a substance-free animus or discriminatory impulse against religion. But “use-based discrimination,” as the Court puts it, is just ordinary lawmaking. As preeminent constitutional historian Leonard Levy unequivocally concluded, “the fact is that no framer believed that the United States had or should have power to legislate on the subject of religion.”¹⁸⁶Levy, supra note 7, at 121–22. Yet, perversely, under the Court’s new anti-religious-discrimination doctrine, states now must do so. As of 2022, the substantive educational content a state chooses not to expend its resources on is subject to the Court’s intrusive new religion clause rule.

Although those who want their children to receive a faith-based education are by no means precluded from making this choice, according to the Court the mere fact that they must pay for such education themselves (while the choice to utilize a secular private school would be supported by the state) exerts coercive pressure on their choice.¹⁸⁷Carson, 142 S. Ct. at 1996. A failure to fund faith-based approaches to education does not just result in the natural disappointment felt by those in a democracy whose policy preferences do not go completely fulfilled, such failure to spend “ ‘penalizes the free exercise’ of religion.”¹⁸⁸Id. at 1997. The implications of this conceptualization are quite stunning. The Supreme Court is effectively depriving democratic governments of their discretion to determine their spending priorities in one of the most consequential and democratically hard-fought domains: public education.

X.  THE LIMITING PRINCIPLE PROBLEM

Now, some might be inclined to see the concerns above as alarmist. Carson, after all, addresses just one case-specific state program. However, it is difficult to see the stopping point of the Court’s logic. The Court’s novel anti-religious discrimination rule lacks a limiting principle. In Trinity Lutheran, Roberts seemed at least mildly attuned to this potential concern by emphasizing the purportedly status-based nature of the discrimination. But, consistent with the Chief’s camel’s-nose-under-the-tent approach to doctrinal change, after Carson, any government program might become the next target of an allegation that it is discriminating against religion, and therefore violating the Free Exercise Clause. Under the Court’s newly expansive anti-religious-discrimination rule in Carson, simply not offering a comparable religion-based alternative to any secular state benefit presents a potential constitutional infraction. What might the future portend under such a regime? We might anticipate a kind of constitutionally mandated menu-based governance in which state resources must be shared equally among religious and non-religious options.

For religion, this vision may ultimately prove to be self-defeating. Government resources are limited, as is the tolerance of the populace for ever higher taxes. Constitutionally mandated religious alternatives will become costly and will ultimately be subjected to the same type of politicized, compromise-laden, and messy process that is at the heart of all spending decisions in a democratic polity. As a constitutionally imposed unfunded mandate, religion would lose its prized independence.

Granted, the anti-religious-discrimination impulse is understandable. As mentioned, the new anti-religious-discrimination Free Exercise principle is no doubt rooted to some extent in a broader concern that religion, religious belief, and religious practitioners have been unfairly mistreated and disparaged by a secular society. However, those who would like to see an expanded role for religion in the public sphere, even those who support taxpayer subsidies of religion in certain areas, may ultimately find themselves deeply troubled by the ultimate consequences of the slippery-slope the Court has erected. The Court’s radical re-interpretation of the religion clauses may prove self-defeating, for government and religion. Without a limiting principle, it cannot be contained.

XI.  THE PUBLIC FORUM DOCTRINE TO THE RESCUE

All of this is not to say that there is no place for a constitutional principle prohibiting, in some contexts, discrimination against religion. Just because the religion clauses demand the opposite, does not mean there are not other settings in which religion may be protected from government. The Free Exercise Clause, most obviously, protects religion by forbidding targeted prohibitions on free exercise. But those who would like to see a greater presence of religion in the public sphere have an alternative constitutional hook to grasp. Another First Amendment doctrine, derived from the Free Speech Clause, does include an anti-discrimination rule that serves to protect against forms of religious discrimination.

The public forum doctrine prohibits the government from imposing viewpoints, and sometimes content-based, discrimination on private speech; and the Court has concluded that this restriction extends to religious expression.¹⁸⁹See Lamb’s Chapel v. Ctr. Moriches Union Free Sch. Dist., 508 U.S. 384 (1993); Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819 (1995). The Court reminded us most recently of this principle in Shurtleff v. City of Boston. The case involved a government program that over time allowed hundreds of private groups to fly their flags outside of Boston’s city hall. The city, however, denied such opportunity to a Christian group. In ruling against Boston on free speech grounds, the Court explained that “[w]hen a government does not speak for itself, it may not exclude speech based on ‘religious viewpoint’; doing so ‘constitutes impermissible viewpoint discrimination.’ ”¹⁹⁰Shurtleff v. City of Boston, 142 S. Ct. 1583, 1593 (2022).

If the government speech doctrine can be said to be pro-discrimination—rooted in the understanding that a democratically accountable government must have the ability to be selective as to what policy messages it will, or will not, send—its cousin, the public forum doctrine, forbids discrimination in government-owned, funded, or controlled forums. Once a government opens property or a program up to the broader public, establishing a public forum—or to a select portion of the public for more circumscribed purposes, establishing what the Court has called a “limited” public forum—it may not discriminate on the basis of “content” (or merely “viewpoint” where the public forum is limited).¹⁹¹See, e.g., McCullen v. Coakley, 573 U.S. 464 (2014); Christian Legal Soc’y v. Martinez, 561 U.S. 661 (2010). Government speech and public fora may be conceived as two poles on opposite ends of a single continuum, with government having almost complete control over what is or is not expressed on the government speech end and minimal power to restrict or dictate expression on the other.¹⁹²See Wayne Batchis, The Government Speech-Forum Continuum: A New First Amendment Paradigm and Its Application to Academic Freedom, 75 N.Y.U. Ann. Surv. Am. L. 33 (2019). The critical point is that the public forum doctrine, unlike the Court’s new anti-religious-discrimination rule, has a clear limiting principle: a government program must fall within the definition of a public forum (or limited public forum) for religion to receive protection from discrimination. Otherwise, a policy choice, and any messages associated with it—unless, of course, it “prohibits” or “respects an establishment of religion”—would be treated as government speech.

Thus Justice Thomas, in his Espinoza concurrence, begs the question when he criticizes the “strict separation” approach to the religion clauses for the way it would ostensibly remove “the entire subject of religion from the realm of permissible governmental activity . . . operat[ing] as a type of content-based restriction . . . .”¹⁹³Espinoza v. Mont. Dep’t of Revenue, 140 S. Ct. 2246, 2266 (2020) (Thomas, J., concurring). Religion would not be banished from the public sphere under the traditional, straight-forward reading of the First Amendment advocated in this article; the impact of the Religion Clauses would simply turn on whether or not the government itself is speaking or whether its “activity” was creating a public forum. As the Court has repeatedly reaffirmed, government speech is all about content-based restrictions on speech—both the discriminating choices government makes as to what messages it will or will not devote its resources to, and the structural boundaries enshrined in the Constitution that may similarly shape, limit, or direct its expressive choices. A public forum, on the other hand, does demand that the government avoid content or viewpoint-based discrimination.

Indeed, this is where a misguided concurrence by Justice Kavanaugh in Shurtleff gets it so wrong. He seeks to supplement the majority’s opinion by emphasizing that a government does not merely violate the Establishment Clause by treating religion equally to other government beneficiaries. Equal (favorable) treatment of religion by government is permissible under certain circumstances in accordance with both the public forum doctrine and the “play in the joints” religion clause principle long accepted by the Court.¹⁹⁴See supra note 61and accompanying text. But in startlingly broad terms, Kavanaugh goes on to assert that “a government violates the Constitution when (as here) it excludes religious persons, organizations, or speech because of religion from public programs, benefits, facilities, and the like.”¹⁹⁵Shurtleff, 142 S. Ct. at 1594 (Kavanaugh, J., concurring). Confined to public fora, such a statement of the rule may be true; but outside of these confines, such a rule would impose constitutionally illimitable unfunded expressive mandates on governments, potentially violating anti-establishment principles and the core premise of the government speech doctrine along the way.  

In contrast, drawing a boundary between a public forum—where religious expression would be protected from discrimination—and government speech—where government would have the option and sometimes obligation to discriminate against religious messages––is remarkably consistent with the inherent tension built into the Religion Clauses. It brings the First Amendment full circle, connecting the Speech and Religion Clauses in a logically coherent way. Governments may establish public fora to facilitate private speech, as governments have a rich and important history of doing, whether it is a public park, the after-hours use of public facilities for associational meetings, or a public university’s student organization program. These are venues that may be owned and maintained by the state, but as public forums, the speech that occurs there is protected and not presumed to represent the government’s voice. As a result, such expression, even if overtly religious, is unlikely to raise traditional Establishment Clause concerns; it is unlikely to generate the impression of government endorsement or to have a coercive effect. Government, in other words, would be free to facilitate free exercise values in a way that is cognizant of Establishment Clause values, while at the same time acting consistently with free speech doctrine.

CONCLUSION

The Supreme Court’s decision in Carson v. Makin is the third in a trilogy of cases dramatically upending the meaning of the First Amendment’s Religion Clauses. Beginning with Trinity Lutheran in 2017, and followed by Espinoza in 2020, the Court has moved forward with an aggressive project of transforming the Religion Clauses into a broad anti-religious-discrimination clause. In this paper, I traced this doctrinal devolution and argued that the Court’s novel reinterpretation is deeply misguided.

By design, the Religion Clauses require discrimination—religion is to be treated differently from non-religion in a broad range of state action. The Establishment Clause targets religion specifically by prohibiting laws that intermingle government with religion in an impermissible manner—whereas intermingling government with other philosophies, worldviews, institutions, or sets of values is a perfectly ordinary and generally acceptable aspect of policymaking. The Free Exercise Clause likewise forbids government interference with religious practice—whereas government is certainly free to, and is indeed expected to, interfere with a vast range of non-religion related conduct deemed to violate criminal and civil law. Religion, in short, is different. The contemporary Supreme Court, however, has inverted this most basic insight.

The Court’s new Religion Clause jurisprudence is also on a collision course with its burgeoning government speech doctrine. That doctrine recognizes that in a democratic polity, every policy choice entails paths not chosen. Government must be able to select its own message, and in turn, discriminate against those messages it wishes not to communicate. While there are some exceptions to the rule—specifically, the boundaries set by the Constitution itself—the default is governmental discretion, tempered only by accountability at the ballot box. Thus, the Religion Clauses, in conjunction with the government speech doctrine, mandate that government either be free to speak with its own voice when it is acting within the “play in the joints” in between the two clauses, or treat religion distinctly—to discriminate—when required to do so under the Constitutional mandate of establishment or free exercise.

To say that discrimination is required under the Free Exercise or Establishment Clause is not to say discrimination against religion is always constitutional. Outside of the Religion Clauses, other protections against objectionable discrimination remain. The Court’s public forum doctrine, for example, protects free expression of religion from content-based discrimination when the government itself is not speaking. Adverse or favored treatment by government targeting religion generally, particular religious sects, or particular religious practices, may be impermissible. But when it comes to the Religion Clauses, these are circumstances in which the discrimination provides evidence that the government is either prohibiting free exercise or making a law respecting an establishment of religion. The discrimination itself is not the Constitutional offense. Acting as if it is, is highly misleading. The Religion Clauses provide nothing like the broad anti-discrimination mandate today’s Court imputes to them. They demand the opposite.

The heart of the Court’s recent trilogy of cases—from Trinity Lutheran v. Comer to Carson v. Makin—is a constitutional mandate that government subsidize religious speech to avoid a Religion Clause “discrimination” claim. It is a command that government express ideas it may not wish to express. The Court’s reimagining of its Religion Clauses jurisprudence is inconsistent with the First Amendment’s original meaning, anti-democratic, and in direct tension with the government speech doctrine.