Blaine_amendments

December, 2003

72 Fordham L. Rev. 493

LENGTH: 33434 words

ARTICLE: SECULARISM'S LAWS: STATE BLAINE AMENDMENTS AND RELIGIOUS PERSECUTION

Kyle Duncan*

* Associate-in-law and LL.M. Candidate, 2002-04, Columbia University School of Law; J.D., 1997, Paul M. Hebert Law Center at Louisiana State University. I am grateful to Kent Greenawalt, Ira C. Lupu, and Jay S. Bybee for their helpful suggestions. This Article is dedicated to my wife, Martha.

SUMMARY:
... “ The U.S. Supreme Court, over one dissent, declined to hear Witters' subsequent claim that Washington's constitution effectively punished him for pursuing his faith and therefore violated his right to free exercise of religion. ... No State shall make any law respecting an establishment of religion or prohibiting the free exercise thereof; and no money raised by taxation in any State for the support of public schools, or derived from any public fund therefor, nor any public lands devoted thereto, shall ever be under the control of any religious sect, nor shall any money so raised or lands so devoted be divided between religious sects or denominations. ... And Lukumi explicitly invoked that long history when it glossed “religious persecution” as laws that “discriminate[] against some or all religious beliefs or regulate[] or prohibit[] conduct because it is undertaken for religious reasons. ... First, the Court's treatment of laws targeting religious viewpoints for exclusion from limited public fora echoes the Court's approach to non-persecution in the free exercise context and to neutrality in the non-establishment context. ... But when a government funding program neutrally furthers secular interests in, for instance, education, health care, or child care, a religious person or organization seeks inclusion in the program on the basis of being a qualified education, health care, or child care provider - and not as a “religious” provider. ...

TEXT:
[*493]

State Blaine Amendments are provisions in thirty-seven state constitutions that restrict persons' and organizations' access to public benefits on religious grounds. They arose largely in the mid to late 1800s in response to bitter strife between an established Protestant majority and a growing Catholic minority that sought equal access to public funding for Catholic schools. After the failure to pass a federal constitutional amendment - the “Blaine Amendment” - that would have sealed off public school funds from “sectarian” institutions, similar provisions proliferated in state constitutions. These “State Blaines” have often been interpreted, under their plain terms, as erecting religion-sensitive barriers to the flow of public benefits that exceed the church-state separation demanded by the Establishment Clause. Today, the State Blaines are becoming increasingly relevant as the Supreme Court has progressively softened federal constitutional barriers to religious access to public funds. This Article examines the history, language, and general operation of the State Blaines. It concludes that the State Blaines generally raise explicit, religion-sensitive barriers to the allocation of otherwise available public benefits and, consequently, that the operation of the State Blaines would typically violate the religious non-persecution principle of the First Amendment.

I. Introduction

Larry Witters was a blind man who wanted to attend college. In 1979, he applied for vocational funds that Washington State provided for the visually handicapped. Witters was eligible for the funds, and he intended to use them to study to be a minister at a Christian college. But his plans met resistance. In 1984, the Washington [*494] Supreme Court ruled that the federal Establishment Clause barred Witters' use of the funds for religious training.n1 Witters sought review in the U.S. Supreme Court and won. In 1986, the Court ruled that the Establishment Clause presented no impediment to his private decision to apply the funds to religious education. n2 But Witters would never use those funds for that purpose. Three years later, the Washington Supreme Court decided on remand that Witters' plans violated a clause of the Washington State Constitution that prohibited “public money” from being “applied to any religious ... instruction.” n3 The U.S. Supreme Court, over one dissent, declined to hear Witters' subsequent claim that Washington's constitution effectively punished him for pursuing his faith and therefore violated his right to free exercise of religion. n4

Thus, at the end of a decade-long odyssey that included a unanimous victory in the Supreme Court, Witters was left with nothing. Had Witters planned to use the scholarship funds to study chemistry, American history, international law, or - interestingly - religion from a purely secular viewpoint, he would have enjoyed Washington's financial assistance. But precisely because Witters wanted to use the funds to prepare for the ministry - i.e., to lay the theological and pastoral groundwork for a career inspired by and in service of his religious faith - he was denied that assistance.

[*495] The provision that ultimately blocked Witters' claim belongs to a class of state constitutional provisions that appear in over thirty-five state constitutions and are known collectively as “State Blaine Amendments.” Most State Blaines arose in the mid to late 1800s, in response to a widespread controversy over whether Roman Catholics could obtain access to public funding for their schools.n5 At that time, American public schools were overwhelmingly and explicitly Protestant, and private schools were predominantly Catholic. Many people wanted to keep public funds as far from Catholic schools as possible, a project zealously pursued and realized in its most concrete form in the State Blaines. While collectively aimed at this object, the language of individual State Blaines takes various forms. n6 Almost all, however, can be fairly read to thwart plans like Witters' - i.e., to bar the use of generally available public benefits precisely because the recipient is a person who wants to put them to a religious use or is a religiously affiliated organization. n7 These provisions have slumbered in state constitutions for over a century, n8 but they are awakening now that the Supreme Court has relaxed federal constitutional barriers to public funding of religious activities. This Article will explore the question the Supreme Court declined to take up in Witters and has never squarely addressed: If a state interprets its Blaine Amendment to erect a religion-sensitive barrier to public funding - funding that is permissible under the Establishment Clause - does the state violate any principle in the federal Constitution? n9

[*496] The tale of the State Blaines seems unfinished, because over the last century state courts have applied them infrequently. The reason is not neglect but superfluity: States have not had to rely on State Blaines to achieve a rigorous separation between public funds and religious institutions because the Supreme Court has interpreted the federal religion clauses to achieve largely that result. As late as the 1980s, only a trickle of public funds could flow to religious students or religious schools (especially elementary and secondary schools) through the sieve of a rigidly separationist interpretation of the federal Constitution.n10 The State Blaines have simply lacked occasion for robust application. But their moment may have arrived. n11 Over the last two decades, the Supreme Court has eased constitutional restrictions on religious access to public funds, n12 and, as happened on remand in Witters III, this will force state courts to ask whether State Blaines place stricter limitations on public funding for the religious. n13 [*497] Inevitably, courts will have to say whether the nature of those limitations can withstand scrutiny under the federal Constitution.

That latter inquiry is the subject of this Article. Beyond what likely motivated the passage of the State Blaines, the more significant foundational question is what they purport to do and whether that operation is consonant with a longstanding tradition of protecting religious liberties under the Constitution. In answering these questions, it is not enough to bring an indictment of anti-Catholicism against the State Blaines. Few would doubt that many, if not most, State Blaines were driven by legislators' desires to penalize a disfavored religious group. But, for my purposes, the key question will be how those motives translated into legal form in the language and operation of the State Blaine Amendments. The history of State Blaines, consequently, provides a useful context for understanding their operation, but it is only the beginning of the constitutional inquiry.

The religious dynamics of the State Blaines are different today than in the nineteenth century. Public schools are no longer Protestant or indeed traditionally religious at all - the Supreme Court's religion jurisprudence since the mid-1960s has scoured public schools of all formal religious practice.n14 Private schools, while significantly religious, are no longer overwhelmingly Catholic. n15 Anti-Catholic bias may no longer be ascendant, n16 but our public institutions have embraced, in Justice Goldberg's memorable phrase, a “brooding and pervasive devotion to the secular” that instinctively confines serious religion to the private sphere and recoils from its intrusion into the public sphere. n17 Against this reshuffled social and religious backdrop, [*498] the non-specific textual references to “religion,” “sects,” or “sectarian” in the State Blaines will operate to restrict, not only Catholic schools or Catholic organizations, but religious schools and organizations generally.n18 Thus, the most obvious function of the State Blaines will be to separate the religious from the secular in the allocation of public funds, raising explicit barriers against the use of public assistance for a variety of, if not all, religious ends and religiously affiliated organizations. n19

If that is how the State Blaines operate, then they violate the religious freedom guarantees of the First Amendment. Laws may not [*499] attach a civil disability to lawful behavior, status, or association because, and only because, they are motivated by religious impulses or connected to religious belief or observance. On this account, State Blaines are “laws that by their terms impose disabilities on the basis of religion.”n20 The State Blaines unconstitutionally “punish” religious status, behavior, and association by selectively disqualifying them from generally available public assistance. That conclusion goes to the deepest roots of American religious freedom: as Michael McConnell has observed, “from the outset [of the United States], the prevention of persecution, penalties, or incapacities on account of religion has served as a common ground among all the various interpretations of religious liberty.” n21 The State Blaines break faith with that tradition.

This Article focuses on the Free Exercise Clause as a primary, but not exclusive, source of principles that prohibit the discriminatory operation of the State Blaine Amendments.n22 The free exercise violation reaches deeply to the historical and normative roots of that clause - as originally conceived, the clause would have applied most vigorously to federal laws aimed at religious exercise. n23 Moreover, even laboring under the inconsistency of its religion jurisprudence, the Supreme Court has consistently (and unanimously) held that laws [*500] targeting religiously motivated behavior, status, or association because of their religious content or connection are presumptively unconstitutional. Beyond free exercise, aspects of the Court's non-establishment and free speech jurisprudence reinforce the constitutional prohibition against invidious government classification of religion and the religious.

Thus, a major theme in this Article is non-discrimination. The First Amendment forbids government from selectively demoting those who act on religious conviction to second-class citizenship in the distribution of public benefits.n24 A second theme is federalism. The Free Exercise, Establishment and Free Speech Clauses apply to the states because they are “incorporated” into the Fourteenth Amendment. n25 Before incorporation of the religion clauses, the states [*501] presumably could discriminate against religion generally, or against certain faiths, as much as they liked.n26 But incorporation of the First Amendment has taken religious discrimination at any level of government off the table. n27

The effects of incorporating the religion clauses foreclose a general conceptual objection to my argument. This objection, addressed below in Part V.A., is posited on a federalism rationale that states may, through their more restrictive Blaine Amendments, legitimately “define[] [a] vision of religious freedom as one completely free of governmental interference.”n28 In the course of my argument, I will demonstrate that the settled application of the Free Exercise, Establishment, and Free Speech Clauses to the states significantly restrains states in how they pursue this elusive vision of a society where religion and government are “completely free” from one another. Specifically, states cannot further such a goal by erecting, on the basis of their Blaine Amendments, “secular” or “non-religious” as motivational, behavioral or associational requirements for access to generally available public benefits. If the origins and operation of the State Blaines are properly understood, then the principle of non-persecution embedded in the First Amendment will strictly [*502] circumscribe, if not completely nullify, their impact on the freedom of religious persons and organizations to participate equally in public benefits.

II. History

America's collective obsession with public schooling began in the early 1800s, when a fever of enthusiasm in the form of the “common school” movement swept the nation. The idea of public education was closely linked to the idea of moral education - and that in turn was linked with religious training. Unsurprisingly, American public schools had a distinctively religious flavor marked by the majority Protestant ethos of the day. This dismayed the growing number of American Catholics, who, with increasing volume and intermittent success, began asking for public money for their own private schools. But the Protestant majority was alarmed in turn, fearing its tax dollars would be siphoned off for “dark Catholic purposes,” and so cries went up for laws to prevent public money going to “sectarian” organizations.n29 The movement culminated, disappointingly for Protestants, in the narrow defeat of a federal constitutional amendment - the Blaine Amendment - in 1875. But rising from the ashes of the federal attempt, a host of like-minded state constitutional provisions flourished over the next quarter-century. Thus were the State Blaines born. n30

A. Common Schools

Before the middle third of the 1800s, there was no public education in America to speak of. Education was largely administered by churches and clergy and was intertwined with religious instruction.n31 But in the 1830s, riding the tide of a “massive evangelical resurgence,” [*503] the common-school movement took hold.n32 Its leading figure was Horace Mann, Massachusetts' secretary of education from 1837-49, who championed the infusion of common schools with explicitly religious moral instruction - a curriculum whose theological content evidenced a “pan-Protestant compromise, a vague and inclusive Protestantism” designed to tranquilize conflict among Protestant denominations. n33 Daily reading, without divisive commentary, of the King James Bible - along with recitation of the Lord's Prayer and the singing of hymns - thus became the foundation of religious instruction in the common schools. n34 So entrenched was this vague Protestant ethos that educators like Mann could claim that the common schools' religious content was not “sectarian,” insofar as the curriculum excluded doctrines “peculiar to specific denominations but not common to all.” n35 Only in this narrow liberal Protestant sense could [*504] American public schools in the mid-1800s be fairly characterized as “religious but nonsectarian.”n36 But the common consensus supporting the common schools' religious and moral foundations plainly excluded Catholics, other non-mainstream believers (Mormons, Jehovah's Witnesses, and the like), and non-believers. n37

B. Growing Catholic Population and Influence

At this time, American Catholics were increasing in numbers and political influence. Through immigration mostly from Ireland and Germany, the Catholic population in the United States increased sharply from a mere 1% of the population during the Revolution to about 3.3% in 1840, 10% in 1866, and 12.9% by 1891.n38 These Catholic immigrants, poor and unfamiliar with American society, flooded into major northern cities such as New York, Chicago, Philadelphia, Boston, and Cincinnati. n39 They were easy targets for discrimination by the “nativist” Protestant population, and such sentiments readily blended with religious hatred. As Philip Hamburger writes:

Fearful of the foreigners, many native-born Protestants self-consciously identified themselves with America and its native population and, on this basis, these “nativists' opposed foreign immigration, especially by Irish Catholics. Yet even this sort of secular ethnic and class animosity often blended into the religious prejudice that would do so much to popularize the separation of church and state.n40

Nonetheless, through sheer numbers, ethnic cohesion and religious identity, American Catholics gained increasing political influence.n41 [*505] The Protestant-dominated public school system would furnish the inevitable political battleground, pitting Catholics' desires for educational and societal equality against nativist Protestants' fears of Catholic influence.

C. Conflict over School Funding

The explicit religious practices widespread in American public schools of this period were a direct affront to Catholics' religious beliefs.n42 Not only did the Catholic Church not recognize the King James translation of the Bible - the only officially approved English translation of the Bible was the Douay version - but daily “unaccompanied Bible reading, which was the cornerstone of the Protestant consensus,” violated Catholic conviction that scripture should be read only in the context of the Church's authoritative doctrinal tradition. n43 Textbooks, moreover, often denigrated Catholics and their faith. n44 Catholics responded by exercising their growing political power to oppose Protestant religious practices in public schools and, beyond that, to request public funds for their own schools. n45 This provoked from the Protestant establishment “a display [*506] of majoritarian politics of unprecedented brutality.”n46 Catholics' request for school funds inflamed latent Protestant fears of Catholic domination. For instance, the Board of Assistants of New York City - a focal point for the school funding controversy - issued an influential report that invoked fears of “religious zeal, degenerating into fanaticism and bigotry, [that] has covered many battle-fields with its victims” as well as macabre images of “the stake, the gibbet, and the prison.” n47 Such rhetoric provoked mob violence against Catholics, as, for example, when the residence of the Catholic Bishop of New York City, John Hughes, was destroyed and the militia were enlisted to defend St. Patrick's Cathedral. n48

[*507] A more systematic reaction arose in the form of legislation forbidding “sectarian control” over public schools and blocking any diversion of public money to religious institutions.n49 Roughly by the time of the attempted federal Blaine Amendment in 1875, fourteen states had passed state laws - some in the form of constitutional amendments - to seal off public funds from sectarian control. n50 Emblematic was the 1840s New York law (a direct precursor of an 1894 provision in the New York Constitution) that prohibited public funding of any school where ““any religious sectarian doctrine or tenet shall be taught, inculcated, or practiced.'“ n51

D. The Federal Blaine Amendment

The bitter fight over school funding eventually began to have national reverberations. On September 30, 1875, President Ulysses S. Grant gave an important speech in which he capitalized on Protestant alarm at perceived Catholic incursions into American education. Delivered in Des Moines, Iowa, to a convention of the Society of the Army of the Tennessee, Grant's address palpitated with anti-Catholic implications:

If we are to have another contest in the near future of our national existence, I predict that the dividing line will not be Mason and Dixon's, but it will be between patriotism and intelligence on one side, and superstition, ambition and ignorance on the other. In this centennial year, the work of strengthening the foundation of the structure laid by our forefathers one hundred years ago, should be begun. Let us all labor for the security of free thought, free speech, and pure morals, unfettered religious sentiments, and equal rights and privileges for all men, irrespective of nationality, color or religion. Encourage free schools, and resolve that not one dollar appropriated to them shall be applied to the support of any sectarian school. Resolve that neither the State or nation, nor both combined, shall support institutions of learning other than those sufficient to afford every child in the land the opportunity of a good common school education, unmixed with atheistic, pagan, or sectarian tenets. [*508] Leave the matter of religion to the family altar, the church, and the private schools, supported entirely by private contribution. Keep the Church and State forever separate.n52

Grant's speech was an obvious partisan move to shore up his Republican party, which had been wounded by corruption and had lost significant political capital in the last national election.n53 The speech effectively allied the Republicans with mainstream Protestants and with a popular, anti-Catholic form of church-state separation. n54 Less than three months later, in his annual message to Congress on December 7, 1875, Grant proposed a constitutional amendment

making it the duty of each of the several States to establish and forever maintain free public schools ... forbidding the teaching in said schools of religious, atheistic, or pagan tenets; and prohibiting the granting of any school funds or taxes, or any part thereof, either by the legislative, municipal, or other authority, for the benefit or in aid, directly or indirectly, of any religious sect or denomination, or in aid or for the benefit of any other object of any nature or kind whatever.n55

Grant ornamented his proposal with warnings that, lacking adequate intelligence and education, “ignorant men [may] sink into acquiescence to the will of intelligence, whether directed by the demagogue or by priestcraft.”n56 Grant's proposal was hailed by the New York Times and Tribune, by Harper's Weekly, and by the Chicago Tribune. n57 But, as Philip Hamburger describes, not everyone was so sanguine about the amendment's assault on federalism: “The proposed amendment's intrusion into traditional state powers provoked astonishment among such Americans as were not utterly blinded by anti-Catholicism.” n58

Unfazed by such subtleties, Congressman James G. Blaine of Maine [*509] eagerly picked up Grant's gauntlet when, one week later on December 14, 1875, Blaine proposed a constitutional amendment embodying the most popular of Grant's ideas.n59 Having lost the House Speaker's chair in the Republican congressional reversals of 1874, Blaine had set his sights on the Republican presidential nomination for the 1876 election. n60 The substance of Blaine's proposed amendment met with widespread approval (except, of course, from Catholics), but most people saw, beneath the veneer of fashionable anti-Catholicism, a transparent attempt to garner political support. n61 Blaine himself - whose own mother was Catholic and whose daughters went to Catholic boarding schools - denied any anti-Catholic motivations and explained in an open letter that his proposal was merely designed to suppress religious conflict by definitively settling the school funding controversy. n62 Blaine was more likely engaged in rank political opportunism. Once it was clear that Blaine had lost the presidential nomination to Rutherford B. Hayes, he lost all interest in the amendment, participated in none of the congressional debates, and - strikingly, as Blaine had assumed a seat in the Senate by the time that body considered the amendment - did not even show up for the Senate vote on the proposal, which failed to pass by only four votes. n63

Blaine's proposed amendment “rewrote the First Amendment to apply it to the states and to specify a single logical consequence of separation - the one most popular with anti-Catholic voters”:n64

No State shall make any law respecting an establishment of religion or prohibiting the free exercise thereof; and no money raised by taxation in any State for the support of public schools, or derived from any public fund therefor, nor any public lands devoted thereto, shall ever be under the control of any religious sect, nor shall any money so raised or lands so devoted be divided between religious sects or denominations.n65

[*510] The proposed amendment passed the House, with an addendum specifying that it did not “vest, enlarge, or diminish legislative power in the Congress,” by a vote of 180 to 7.n66 During the more extensive Senate debate on the proposal, some senators expressed confusion about the scope and application of its language. n67 The Senate subsequently proposed a more absolutist version that would have categorically prohibited any “public property,” “public revenue” or “loan of credit” from being “appropriated to or made or used for the support of any school or other institution under the control of any religious or anti-religious sect, organization, or denomination” or where the “creed or tenets” of such groups were taught. n68 Notably, the Senate proposal provided that its language “shall not be construed to prohibit the reading of the Bible in any school or institution.” n69 The Senate version failed to garner the required two-thirds majority by a mere four votes - twenty-eight to sixteen (with twenty-seven members not present, including Blaine himself) - and failed. n70

A final political wrinkle, developed in detail in Philip Hamburger's recent work, deserves mention.n71 Whereas the 1830s-50s surge in anti-Catholicism was almost exclusively fueled by nativist Protestants, the 1860s-70s surge that culminated with the failed Blaine Amendment included a significant additional motivating force: the “secularists” or “Liberals.” This diverse group united a wide variety of atheists, theists, and spiritualists in a common resentment and mistrust of Christianity's influence on government. n72 They were best exemplified by the Free Religious Association, in its central publication, The Index, and by the founder of The Index, Francis Ellingwood Abbot. n73 The Liberals were fueled in part by the misguided efforts of some Protestants under the banner of the National Reform Association, to pass a “Christian Amendment” to the U.S. Constitution. Abbot formed the National Liberal League - devoted to “the absolute separation of church and state” - to fight the Christian Amendment [*511] with secularizing counter-proposals.n74 He distilled Liberal philosophy into the 1872 publication, The Demands of Liberalism, which presciently tracked many of the most difficult church-state issues that the Supreme Court would face in the twentieth century, including church tax-exemptions, legislative chaplains, Sunday laws, and Bible reading in public schools. n75 Significantly, Abbot included in his Demands that “all public appropriations for sectarian educational and charitable institutions shall cease,” and that in both the federal and state constitutions “no privilege or advantage shall be conceded to Christianity or any other special religion” and that “our entire political system shall be founded and administered on a purely secular basis.” n76

Liberals did not think the Blaine Amendment went nearly far enough in extirpating all vestiges of religion from government. They viewed it merely as an anti-Catholic measure that explicitly preserved a generalized, non-divisive Protestantism in public schools.n77 The competing amendment proposed by Liberals in 1876 contained more explicit and comprehensive safeguards than the Blaine Amendment (particularly the House version). For instance, the Liberal amendment would have prohibited “taxing the people of any State, either directly or indirectly, for the support of any sect of religious body or of any number of sects or religious bodies;” it would have protected a person's right not to be “required by law to contribute directly or indirectly to the support of any religious society or body of which he or she is not a voluntary member;” and, reminiscent of the absolutist language that would appear sixty years later in the seminal Everson decision, n78 it would have prevented any governmental unit from “levying any tax, or making any gift, grant or appropriation, for the support, or in aid of, any church, religious sect, or denomination,” or any religious school or charity. n79

As such proposals show, the Liberal ethos took separationism to its logical extreme. “Liberals,” writes Philip Hamburger, “viewed all Christians with the same fear and horror Protestants reserved for Catholics.”n80 All government connections to religion had to be uprooted. Significantly, Liberals asserted that religious groups should be barred from participating even in public benefits distributed on [*512] secular grounds.n81 This principle would have excluded all neutrally available public appropriations for religious education or religious charities. Interestingly, the Liberals seemed to make an exception for appropriations to individuals who were religious, but not for religious groups. n82

But the Liberals' radically secular project was a political failure.n83 It was the traditionally Protestant, anti-Catholic version of separationism that proved to be more politically viable, even if it, too, did not achieve ultimate national success in the federal Blaine Amendment. The narrower House version of the amendment in particular, as well as the Bible-reading proviso of the more rigorous Senate version, plainly departed from Liberal secularist dogma. n84 Consequently, in the wake of the federal Blaine Amendment's defeat, the nativist Protestants were more successful at securing passage of local versions in state constitutions. n85 The Liberals, who had made themselves distasteful to mainstream Americans through their rigid, fundamentalist attachment to separation and secularism, were reduced to “piecemeal lobbying and cultural agitation” to spread their cause. n86 Yet, it will be useful to keep in mind the Liberals' radical secularist agenda when considering some of the similarly absolutist approaches in many of the State Blaine Amendments.

E. The Spread of State Blaines

Charles Russell, one of James Blaine's biographers, provided this bleak summary of Blaine's accomplishments: “No man in our annals has filled so large a space and left it so empty.”n87 But from the perspective of actual laws passed, Blaine's real legacy lay in the numerous state constitutional amendments spawned after the failure of his federal amendment. n88 The nativist Protestant version of separationism had gradually become part of the Republican agenda and thus, while many states adopted Blaine-like provisions voluntarily, many others were required to incorporate some form of a [*513] “non-sectarian” provision into their state constitutions as a price for entering the union.n89

The general rise and spread of State Blaines can be charted as follows. The school funding controversy beginning in the 1830s gave rise to increasing state legislation restricting religious school funding, sometimes in the form of state constitutional amendments. The failed attempt in the 1870s to pass the federal Blaine Amendment lent momentum to this anti-funding movement, resulting in a proliferation of state constitutional amendments in the closing years of the nineteenth century. As discussed above, New York adopted a restrictive funding law in the 1840s, and, by 1876, fourteen other states had “joined New York in passing measures prohibiting the division of public school funds, often in the form of constitutional amendments.”n90

During the 1870s alone, twelve states - Illinois, Pennsylvania, Missouri, Alabama, Nebraska, Colorado, Texas, Georgia, New Hampshire, Minnesota, California and Louisiana - adopted provisions similar to the federal Blaine Amendment.n91 Following the defeat of the federal Blaine Amendment, Congress also began to require newly admitted states to adopt some form of an anti-sectarian amendment in their own constitutions. n92 For example, the 1889 Enabling Act that [*514] ushered North Dakota, South Dakota, Montana and Washington into the union required that those states' constitutional conventions ““provide, by ordinances irrevocable without the consent of the United States and the people of said States ... for the establishment and maintenance of systems of public schools, which shall be open to all the children of said States, and free from sectarian control.”n93 The same requirement was contained in the Enabling Acts authorizing the statehood of Utah, Oklahoma, New Mexico, Arizona and Wyoming. n94

By 1890, twenty-nine states in all had incorporated into their constitutions explicit prohibitions against the allocation of public funds to sectarian schools and other institutions.n95 The next section [*515] will examine the various linguistic formulas in which the State Blaine Amendments concretized those objectives, and how that language may operate today. While the State Blaines arose out of a specific historical context - as described in this section, they are the legal offspring of the Protestant-Catholic school funding crisis and the political opportunism of Grant and Blainen96 - today the State Blaines have a far more generalized operation in American public life. They are a widespread mechanism for separating public benefits from all religious institutions and religious individuals.

III. State Blaines: Language and Interpretation

The categorization of a particular state constitutional provision as a “Blaine Amendment” can be plausibly approached from various perspectives - e.g., when the provision was adopted, whether it is directly traceable to the aftermath of the failed attempt to amend the federal constitution, how state courts have interpreted it, etc. - and this probably explains why different treatments of the subject find different numbers of existing State Blaines.n97 Given the parameters of my legal analysis, I propose a straightforward method of characterizing a constitutional provision as a State Blaine Amendment, focusing principally on language. For my purposes, a State Blaine means a state constitutional provision that bars persons' and organizations' access to public benefits explicitly because they are religious persons or organizations.

This is a broad definition and, consequently, the parameters of individual State Blaines will vary. For instance, some bar equal participation in public aid only to religious schools; others bar religious organizations or institutions; yet others bar non-public institutions generally, while explicitly including religious institutions in that category. State Blaines also vary in the language used to describe the bar on equal participation. But, whatever range of disabilities or disqualifications exists in the various State Blaines, all of them turn on the religious affiliation of the disabled or disqualified person, status, [*516] or organization. The plain object of disabling religion is what unifies the State Blaines.n98

State courts' interpretation of the nuances of how a particular State Blaine applies will not be exhaustively explored, but two aspects of state court interpretation will be emphasized. First, I will point out when a state court has explicitly recognized that a State Blaine creates a greater separation between church and state than the federal Establishment Clause. Second, I will point out when a state court has done the reverse, interpreting a plainly separationist State Blaine Amendment as doing nothing more than mimicking the parameters of the federal religion clauses. In either case, focusing on these state court interpretations will highlight the federalism aspects of the State Blaine Amendments - i.e., whether they have been interpreted simply to reinforce at the state level the separation the federal clauses already achieve, or whether they have been read to further a distinctive form of church-state separation that exceeds the separation between religion and public funds imposed by the federal religion clauses.

A. Language

As discussed before, by 1876 - just after the failure of the federal Blaine Amendment - fifteen states had adopted some kind of law that explicitly prohibited public funding of religious organizations.n99 These anti-funding measures often found their way into state constitutions. As early as 1848, the Wisconsin Constitution provided: “nor shall any money be drawn from the treasury for the benefit of religious societies, or religious, or theological seminaries.” n100 In the 1850s, five states incorporated similar provisions into their constitutions. The Michigan Constitution of 1850 provided that “no money shall be appropriated or drawn from the treasury for the benefit of any religious sect or society, theological or religious seminary, nor shall property belonging to the State be appropriated for any such [*517] purposes.”n101 In 1851, Indiana added a similar prohibition to its Constitution. n102 Taking an obverse approach, the Ohio Constitution of 1851 required that “no religious or other sect, or sects, shall ever have any exclusive right to, or control of, any part of the school funds of this state.” n103 In 1855, Massachusetts provided in its constitution that funds raised for “public” or “common” schools “shall never be appropriated to any religious sect for the maintenance exclusively of its own schools.” n104 Both Kansas n105 and Oregon n106 followed suit in 1859.

The end of the 1860s and the first half of the 1870s saw similar provisions adopted by South Carolina, Illinois, Pennsylvania, Missouri, Alabama and Nebraska.n107 Illinois adopted an unusually detailed provision barring any payments “in aid of any church or sectarian purpose, or to help support or sustain any school, academy, seminary, college, university, or other literary or scientific institution, controlled by any church or sectarian denomination whatever” and also forbidding any grant of “land, money, or other personal [*518] property ... to any church or for any sectarian purpose.”n108 In the latter half of the 1870s - the period directly coinciding with the failure of the federal Blaine Amendment - Colorado, Texas, Georgia, New Hampshire, Minnesota, California, and Louisiana also adopted anti-funding provisions. n109 Georgia n110 and Minnesota's n111 1877 provisions were notably explicit about the range and character of excluded institutions.

New Hampshire was an instructive and ironic case in point. Since 1784, New Hampshire's constitution had eloquently charged its legislature with promoting the educational flourishing of New Hampshire citizens:

Knowledge and learning, generally diffused through a community, being essential to the preservation of a free government; and spreading the opportunities and advantages of education through the various parts of the country, being highly conducive to promote this end; it shall be the duty of the legislators and magistrates, in all future periods of this government, to cherish the interest of literature and the sciences, and all seminaries and public schools, to encourage private and public institutions, rewards, and immunities for the promotion of agriculture, arts, sciences, commerce, trades, [*519] manufactures, and natural history of the country; to countenance and inculcate the principles of humanity and general benevolence, public and private charity, industry and economy, honesty and punctuality, sincerity, sobriety, and all social affections, and generous sentiments, among the people ... .n112

Somewhat marring the harmony and inclusiveness of these sentiments, New Hampshire added this exception in 1877: “Provided, nevertheless, that no money raised by taxation shall ever be granted or applied for the use of the schools or institutions of any religious sect or denomination.”n113

In the 1880s and 1890s another thirteen states added their numbers to this growing trend of religiously sensitive anti-funding provisions.n114 As discussed above, during this period Congress began requiring newly admitted states to provide in their constitutions for a system of public schools “free from sectarian control.” n115 Consequently, Montana, North Dakota, South Dakota, Washington and Wyoming all placed some form of anti-funding provision in their constitutions in 1889. n116 Idaho and Mississippi added similar provisions in 1890; Kentucky, in 1891. n117 New York added its anti-funding provision in 1894 after a long and bitter fight, previously discussed, over parochial [*520] school funding.n118 Rounding out the nineteenth century, Utah and Delaware added anti-funding provisions in 1896 and 1897, respectively. n119

This era of proliferating anti-funding amendments seemed to wind down in the first decade of the twentieth century. Virginia first included an explicit anti-funding provision in article IV, section 67 of its constitution in 1902.n120 Oklahoma (1907), n121 Arizona (1910), n122 and New Mexico (1911) n123 each included anti-funding provisions in their new constitutions. With these four constitutions, a long period of lawmaking - stretching back over sixty years to the Wisconsin Constitution of 1848 - seemed to pause for breath. When it did, the American state constitutional landscape could boast of some thirty-six states that explicitly barred a wide range of religious schools and institutions from access to an impressive array of public benefits. The constitutional landscape was not significantly altered until the admission of Hawaii and Alaska into the union in 1959, each new state [*521] with anti-funding constitutional provisions.n124 That brought the total of such provisions at that time to thirty-eight.

The remaining developments in relevant state constitutional language are piecemeal but reflect a preoccupation with singling out religiously affiliated organizations. For instance, both in 1956 and in 1971, Virginia amended its anti-funding provisions to create more pointed religion-based exclusions. In 1956, Virginia amended article VIII, section 10 of its constitution to allow the expenditure of public education funds for “Virginia students in public and nonsectarian private schools and institutions of learning.”n125 In 1971, Virginia added article VIII, section 11, allowing its General Assembly to provide loans or grants to “students attending nonprofit institutions of higher education in the Commonwealth whose primary purpose is to provide collegiate or graduate education and not to provide religious training or theological education.” n126 Pennsylvania had made a similar adjustment to its constitution in 1963 when it allowed for the provision of scholarship grants or loans for higher education “except that no [such] scholarship, grants or loans ... shall be given to persons enrolled in a theological seminary or school of theology.” n127 In 1970, Michigan amended its constitution with the apparent purpose of specifically barring any kind of school voucher program. n128 Finally, in 1976, Nebraska made perhaps the most pointed adjustment in any state constitution by providing that its legislature could allow government contracts with non-public institutions to provide “educational or other services” to handicapped persons under twenty-one years old, but only “if such services are nonsectarian in nature.” n129

In this section, I have taken care to acquaint the reader with the specific linguistic formulas by which the State Blaines erect religion-sensitive barriers to the allocation of public benefits. I have done this to allow the State Blaines, in a sense, to speak for themselves. State Blaines are undeniably multi-faceted, which makes it tricky to treat [*522] them generally. I will nonetheless offer four interrelated observations about the nature of the State Blaines' common objectives, as reflected in their language.

First, the State Blaines apply their prohibitions to a wide spectrum of public benefits. Restrictions are sometimes limited to particular sources of public funds - e.g., to a “public school fund” or to “educational funds” - but more commonly they apply broadly to, for instance, “public funds” or “state property,” to “money raised by taxation” or “money drawn from the treasury,” or simply to “money,” categorically forbidding “appropriations” or “payments” from these generic public sources. Second, the State Blaines restrict the application of public benefits to religious institutions in terms that not only circumscribe the destination of the benefits but, separately, their purpose and effect. So, for instance, public funds may not be applied “in aid of,” “for the benefit of,” or to “support or sustain” any religious organization, and, additionally, these forbidden applications may not be achieved “directly or indirectly.” Another way of effecting this kind of restriction is to forbid the appropriation of funds for religious “purposes,” or to prohibit religious groups from having any “control” over public funds. Third, some State Blaines limit their prohibitions to religious “schools,” while many strike more broadly at religious “institutions,” “associations,” “establishments,” and “societies.” Others dictate the tenor of instruction offered at institutions “supported” by public funds, prohibiting “sectarian instruction” at such places.

But the most significant and overarching quality that links State Blaines is that all explicitly tailor their restrictions to religion. They target institutions that are “religious,” “sectarian,” “theological,” “ecclesiastical,” “denominational,” or affiliated with a “church.” They prohibit appropriations to places where the “doctrines,” “creeds,” or “tenets” of religion are practiced or taught, or where religious “worship,” “exercise,” or “instruction” occurs. They delimit the “purposes” for which public benefits may be applied, removing “religious” purposes from the universe of other purposes. They single out individuals who, because of their religious affiliation, cannot be included in the distribution of public benefits - people such as “priests,” “preachers,” “ministers” and “teachers” of religious doctrine.n130

Recall that the State Blaine Amendments arose largely in response to widespread Protestant fears of Catholic influence on society, politics and education. Yet, it is perhaps stating the obvious to observe that the words “Roman Catholic” appear nowhere in any of [*523] the provisions. The State Blaines survive today in thirty-seven state constitutions as broad, explicit, and generic prohibitions on public funding of all religion. Their historical antecedents can help us contextualize the amendments but they should not control their application or our assessment of their constitutionality. The social and religious contexts in which the State Blaines operate today are far different from those of their origins and, consequently, faithful applications of the language of the State Blaines no longer divides, for purposes of public funding, the Protestant public schools from the Catholic private schools. Instead, they divide the thoroughly secularized public schools and other public institutions from a growing array of private religious schools and other private religious entities. They divide persons with religious affiliations or religious purposes from persons with non-religious affiliation and purposes. This operation is fully consonant with the changing dynamic of religious conflicts in modern American society. As Ira Lupu and Robert Tuttle have observed, “the religious wars in the United States in the early twenty-first century are not Protestant vs. Catholic, or Christian vs. Jew, or even the more plausible Islam vs. all others. They are instead the wars of the deeply religious against the forces of a relentlessly secular commercial culture.”n131 One hopes that such modern conflicts are fairly described as something more benign than “wars,” but, regardless, there is little doubt what side the State Blaines are fighting for: The State Blaines are, today, a widespread legal obstacle separating the secular from the religious in the allocation of public benefits. It will be that operation that I will measure against the requirements of the First Amendment.

B. Interpretation

There is no doubt room for nuanced interpretation of the various linguistic formulas that appear in State Blaines. For instance, a court might decide that a provision banning funds “in aid of” a religious school has a broader prohibitory scope than a provision simply banning direct funding.n132 This section will take a broader approach to interpretation. I will discuss state court decisions that explicitly recognize that a State Blaine Amendment has created a greater separation between public benefits and religious organizations than the federal religion clauses require. Conversely, I will note other state court decisions that do the opposite - i.e., despite a State Blaine's restrictive language, decide that the provision imposes no greater obstacles than the federal Constitution to religious groups' access to [*524] public funds. My purpose is to demonstrate that state courts have often - but not always - interpreted the State Blaine Amendments both as going beyond the federal Establishment Clause and also as creating an explicitly religion-sensitive barrier to the allocation of public funds and other benefits.

A prime example of the first kind of interpretation - one recognizing greater state separation - was provided by the Idaho Supreme Court in 1971. In Epeldi v. Engelking, the court considered a provision that provided a neutral transportation reimbursement to public and non-public schoolchildren alike, including children attending religious schools.n133 The reimbursement would have passed muster under the federal Establishment Clause, as interpreted by the Supreme Court years before in Everson and again in Allen. n134 But the Idaho Supreme Court observed that, “unlike the provisions of the Federal Constitution, the Idaho Constitution contains provisions specifically focusing on private schools controlled by sectarian, religious authorities.” n135 Referring to Idaho's Blaine Amendment - article IX, section 5 of the Idaho Constitution - the court confessed that “one cannot help but first be impressed by the restrictive language contained therein.” n136 Based on that language, the court reasoned that “the framers of our constitution intended to more positively enunciate the separation between church and state than did [*525] the framers of the United States Constitution.”n137 The court then struck down the transportation reimbursement provision under the Idaho Blaine Amendment. n138 It remarked, logically enough, that its disposition under the state constitution rendered irrelevant the federal Establishment Clause standards used by the Supreme Court in Everson and Allen. n139

The Washington Supreme Court followed a similar rationale in Witters III, already alluded to in Part I, when in 1989 it barred a blind student's use of generally available public funds for religious training - a use which the U.S. Supreme Court had already, in the same case, allowed under the federal Establishment Clause.n140 The Washington Supreme Court relied on what it called the “sweeping and comprehensive” language of the Washington Blaine Amendment - article I, section 11 of the Washington Constitution - “which prohibits not only the appropriation of public money for religious instruction, but also the application of public funds to religious instruction.” n141 The court reasoned that in this restrictive language “lies a major difference between our state constitution and the establishment clause of the first amendment to the United States Constitution,” thereby making application of federal constitutional standards superfluous. n142 Significantly, the court referred to prior decisions construing the phrase “religious instruction” in article I, section 11, and concluded that the kind of instruction constitutionally barred from funding was “devotional in nature and designed to induce faith and belief in the student,” as opposed to the “open, free, critical, and scholarly examination of the literature, experiences, and knowledge of mankind” that would occur, for instance, in a “Bible as Literature” course. n143

[*526] Further examples of this kind of expansive (i.e., resulting in greater separation than federal constitutional standards) interpretation are easy to find. For instance, in 1963, the Oklahoma Supreme Court concluded that the Blaine Amendment in article II, section 5 of its constitution created a more rigorous funding restriction than the federal Constitution and therefore prohibited the kind of busing reimbursement allowed by Everson.n144 The court reasoned that the construction of the Establishment Clause in Everson had no bearing on the effect of state constitutional provisions. n145 The court was frank and unapologetic about the practical inequity of its decision. It flatly stated that if a parent exercises his right to “provide for the religious instruction and training of his own children” and consequently places them in religious schools, then, as a matter of law, the parent must “assume the financial burden which that choice entails.” n146 The court thus left no doubt that the Oklahoma Blaine Amendment explicitly allocated that financial burden based purely and simply on the religious nature of the parents' choice.

Moreover, when state courts interpret their own constitutions as more restrictive than the federal Establishment Clause, often they also purport to “reject” the reasoning underlying the Supreme Court's Establishment Clause decisions. For instance, the California and South Dakota Supreme Courts have both explicitly rejected the “child benefit” theory relied on by the U.S. Supreme Court in Everson and other cases.n147 Joseph Viteritti observes that “at one time or another courts in nearly half the states have issued pronouncements indicating that they do not consider the Court's [school aid] decisions to be binding in interpreting their own constitutions,” and that “several have specifically rejected the “child benefit theory.'“ n148 Finally, states sometimes reach beyond weaker or even non-existent anti-funding provisions to create rigid barriers against religious funding. For instance, in 1979 the Alaska Supreme Court interpreted its fairly narrow Blaine Amendment - prohibiting only the payment of public [*527] funds for “the direct benefit” of any religious school - to achieve a strict funding prohibition.n149 Vermont has no explicit anti-funding provision in its constitution, but in 1999, the Vermont Supreme Court decided that the provision in chapter 1, article III (protecting persons from being “compelled to ... support any place of worship”) erected a stronger barrier against a neutral voucher program than the Establishment Clause. n150

On the other hand, several state courts have interpreted the plainly restrictive language in their Blaine Amendments as creating no greater separation than the federal Establishment Clause. In one significant recent decision, Kotterman v. Killian, the Arizona Supreme Court refused to interpret Arizona's anti-funding provision in a rigidly absolutist manner, while at the same time criticizing the discriminatory motives behind the federal Blaine Amendment.n151 Other states have chosen either simply to ignore the separationist language in their own constitutions or to interpret it in a manner coextensive with the federal religion clauses. n152 For instance, in approving the loaning of free textbooks to religious schools, the Mississippi Supreme Court leniently interpreted the language in its constitution prohibiting any public funds from being “appropriated toward the support of any sectarian school,” and added that “there is no requirement that the church should be a liability to those of its citizenship who are at the same time citizens of the state, and entitled to privileges and benefits as such.” n153 Similarly, the Ohio Supreme Court has suggested that its state constitution provides greater free exercise rights than the federal Free Exercise Clause, while indicating that its religious anti-funding provision - although phrased in absolutist terms - is merely coextensive with the federal Establishment Clause. n154

[*528] This section simply highlights expansive state court decisions which are significant for two reasons. First, state courts have interpreted State Blaines in a manner that explicitly goes beyond the church-state separation mandated by the federal Establishment Clause, specifically in the area of public aid to religious schools. This expansive interpretation has been occurring for as long as the Supreme Court has been interpreting the boundaries of the Establishment Clause; indeed, such state court decisions tend to cluster around instances in which the Supreme Court has allowed some form of public benefit (as with free transportation in Everson and free textbooks in Allen) to be shared equally between public and religious schools.n155 Second, state courts have frankly recognized that, under their application of the State Blaine Amendments, religiously motivated behavior pays a special price. Those burdens on religion are not incidental but rather are targeted disabilities, the predictable and intended result of a policy of self-consciously distancing the public sphere from religious persons and institutions.

More lenient interpretations of State Blaines are possible, of course, but it is fair to say that such decisions must work hard to hurdle the plainly separationist implications of the language of State Blaines. The more expansive decisions are not aberrations, however. Rather, they faithfully cleave to what the State Blaines say and to the separationist objectives that their language plainly aims to achieve. It will be the remaining task of this Article to say whether those objectives violate the First Amendment.

IV. The Jursiprudential Roots of Non-Persecution

The foregoing cross-section of the State Blaines reveals that a preference for separating public benefits from religious persons and organizations persists in over two-thirds of our state constitutions. Broadly speaking, the State Blaines are the residue of the second great historical controversy to raise profound questions about the shape of American religious liberties - the rise of public schools and the withdrawal of public funds from private religious schools.n156 Those [*529] amendments “arguably represented a political judgment on the constitutional questions raised by such funding.”n157 But we should be skeptical about accepting the judgments of State Blaines as the last constitutional word on those questions. As we have seen, the anti-funding advocates of that era failed to amend the federal Constitution, naturally raising the question whether the State Blaines themselves conflict with federal norms of religious liberty. More importantly, as Douglas Laycock observes, “the nineteenth century movement was based in part on premises that were utterly inconsistent with the First Amendment,” given that “opposition to funding religious schools drew heavily on anti-Catholicism.” n158 Anti-Catholic motives alone may not, in the final analysis, be enough to invalidate the State Blaines under the First Amendment, but their presence should at least raise some red flags. And, raising further suspicions, the plain terms of most State Blaines go well beyond the narrower questions raised by the school funding controversy.

The movement spawning the State Blaines only lapped at the shores of the federal Constitution, but failed to alter it. Thus, the federal constitutional standards governing public aid to religion have charted their own jurisprudential course. The stark kind of strict separationism between all public benefits and religion required by most State Blaines has never been regnant in Supreme Court jurisprudence. Even the first major non-establishment decision, Everson, allowed indirect state aid to religious schools, notwithstanding Justice Black's strict separationist dicta.n159 Some of [*530] the Court's non-establishment decisions may skirt the borders of Blaine-like separationism - Charles Fried recently referred to the Court's mostly-defunct decision in Committee for Public Education and Religious Liberty v. Nyquistn160 as “a kind of Court-imposed Blaine Amendment” n161 - but the Court has generally proceeded in a non-absolutist (if sometimes counterintuitive) manner in sketching the boundaries between permissible and impermissible government aid to religious persons and entities. Furthermore, the direction the Court has been taking over the last two decades highlights the gulf between federal standards of non-establishment and the rigid barriers thrown up by the State Blaines over a century ago.

For instance, it is becoming increasingly evident that the government acts within the bounds of the federal Establishment Clause when it provides secular benefits to a broad range of public and private recipients, including religiously affiliated private recipients, based on criteria that are “neutral” - in the sense that the benefits are not distributed on the basis of any religious quality, or lack thereof, in the recipient.n162 Relatedly, when those secular benefits, neutrally distributed, end up in the hands of religious organizations because of the private choices of individuals - and not because of any deliberate government design to nudge the benefits toward religious ends - government has not impermissibly “subsidized” religion. n163 Generally, the Court has emphasized that the [*531] Establishment Clause does not require a wholesale exclusion of religious entities from participation in government programs and government funding. In other words, the argument is steadily evaporating that selective discrimination against religion finds its justification in the Establishment Clause itself. To be sure, the clause “singles out” religion for a kind of disability, as Michael McConnell explains: “The disestablishment principle prevents the government from using its power to promote, advocate, or endorse any particular religious position.”n164 But this principle stands diametrically opposed to a posture of hostility toward religion that is required, or even justified, by the Establishment Clause. Again, to quote McConnell:

The suggestion that religious organizations must categorically be barred from participation in all government-funded programs must be rejected. Although favored by the so-called “strict separationists,” this has never been the rule in establishment clause cases and has been rejected by the Supreme Court in every case in which it has been seriously advanced.n165

Indeed, McConnell argues that, in both the abortion and religion contexts, “denying federal money for activities that would otherwise be funded would amount to a substantial penalty for exercising one's constitutional rights.”n166

Doubtless, there is clarifying work left to do at the federal level, but for present purposes one may observe, uncontroversially, that federal constitutional barriers to public funding of religious institutions have demonstrably softened, that “the [Supreme] Court has become more solicitous of innovative partnerships between governments and religious institutions,”n167 and that both states and Congress will likely respond - and have already responded - by enacting laws allowing [*532] religious groups to enjoy generally available public benefits.n168 Enter the State Blaine Amendments.

If I may indulge in metaphor for a moment, the role of the State Blaines will become clearer. The federal constitutional standards for permissible aid to religion were, for many years, murky water in a lake - one illustrative example was the distinction, supposedly of constitutional magnitude, between giving secular textbooks to religious schools (constitutional) and giving them maps, globes, and film-strip projectors (unconstitutional).n169 Over the last few decades, that water has gradually been clearing until we can better see what principles govern which kinds of aid the federal Constitution allows and disallows. n170 But simultaneously, we are now beginning to discern another layer of murk representing the State Blaines. As we have seen, the State Blaines are far more stringent than the federal Constitution about the barriers raised against public funding of religious persons, schools and other organizations. The real question now is whether the State Blaines are the bottom of the lake.

If they are the bottom of the lake - if, so to speak, there is nothing “beneath” them to temper or annul what they plainly do - then the resulting legal landscape among the states is fairly predictable. [*533] Depending on how each state constitution is framed and interpreted, we will have in this country a kaleidoscope of separationism: One state will hermetically seal off all public benefits from religious schools; another might do the same for all religious organizations generally; another might focus on individuals who plan to put the benefits to faith-oriented uses; and still another might decide to erect no separationist barriers at all. My canvass of the State Blaines suggests that the balance will be tilted significantly in the direction of shutting off religion from public funds. The ability of religious persons and institutions to enjoy public benefits on an equal basis will be - quite apart from how permissively the federal Establishment Clause is interpreted - refracted through the anti-funding provisions of fifty state constitutions.

But this will only be true if there exists no principle in the federal Constitution that can restrain the process. In this part, I will demonstrate that there is. That principle consists of three conceptually related strands found in Free Exercise, Establishment, and Free Speech jurisprudence. But they combine in one overarching rule - what the Supreme Court has referred to as the “fundamental nonpersecution principle of the First Amendment.”n171 Simply stated, the non-persecution rule means, among other things, that neither state nor federal governments may, consistently with the First Amendment, restrict access to generally available public benefits based on persons' or organizations' religious status, purpose, affiliation, or identity.

A. Free Exercise and Non-Persecution

Prohibiting religious discrimination lies at the heart of the free exercise clause, but it is important to carefully define “discrimination” by reference to the Supreme Court's long history of balancing the conflicting claims of religion and government. Paradoxically, the principle condemning religious discrimination - or “fundamental nonpersecution principle,” as the Court has most recently called it - is best understood against the backdrop of another important free exercise principle, one that restricts religious freedom. That background rule is the “non-exemption” rule, which was best articulated in the 1990 Smith decision but which goes back over 125 years to the Court's earliest religion clause cases.n172 Non-exemption means that the Free Exercise Clause does not require courts to grant religion-based exemptions from the burdens of genuinely neutral laws. The mere statement of the rule suggests that it interacts significantly with the narrower rule that laws may not target religious behavior or affiliation for special disabilities.

[*534] The non-exemption rule (which has been the subject of sharp scholarly debate)n173 illuminates the parameters and continuing force of the non-persecution rule, particularly as it applies to the State Blaine Amendments. As Michael McConnell has explained, whether the Free Exercise Clause requires religious exemptions (as he argues), or whether Smith correctly decided that such exemptions lie only within the province of the legislature, it is clear that the Free Exercise Clause unambiguously forbids laws that directly target religious conduct for penalties or disabilities:

Under both conceptions, it is unconstitutional ... to inflict penalties on religious practices as such. For example, zoning ordinances disallowing churches while allowing meeting halls and other uses with comparable effects are unconstitutional, as are “anti-cult” legislation, laws barring clergy from public office, and charitable solicitation regulations crafted to disadvantage a particular religious sect.n174

The non-exemption rule has jurisprudential roots in the nineteenth century conflict between the Mormon Church and the territorial laws of the United States prohibiting polygamy. In its first significant religion clause decision, Reynolds v. United States, the Supreme Court held that the Mormons' religious tenets - which at the time commanded polygamy as a religious duty for male members - did not exempt them, under the Free Exercise Clause, from obedience to a generally applicable criminal prohibition against polygamy.n175 Twelve years later in Davis v. Beason, the Court explained (again with reference to Mormon polygamy) that the Free Exercise Clause permitted no interference with “man's relations to his Maker and the obligations he may think they impose, and the manner in which an expression shall be made by him of his belief on those subjects, ... provided always the laws of society, designed to secure its peace and [*535] prosperity, and the morals of its people, are not interfered with.”n176 Free exercise, then, “must be subordinate to the criminal laws of the country, passed with reference to actions regarded by general consent as properly the subjects of punitive legislation.” n177

Since 1940, when it recognized that the Free Exercise Clause applied to the states,n178 the Court has had more opportunities to develop the non-exemption rule. In Minersville School District v. Gobitis, the Court gave a more nuanced description of the rule's scope, even as it denied that Jehovah's Witnesses merited a religious exemption from compulsory flag-salute laws: “The religious liberty which the Constitution protects has never excluded legislation of general scope not directed against doctrinal loyalties of particular sects.” n179 Again, in Braunfeld v. Brown, the Court rejected an [*536] Orthodox Jew's claim that a generally applicable Sunday-closing law violated his free exercise rights by imposing an “indirect” burden on his religious beliefs, which honored Saturday and not Sunday as a day of rest.n180 But, in doing so, Braunfeld observed that, unlike a truly general law, “if the purpose or effect of a law is to impede the observance of one or all religions or is to discriminate invidiously between religions, that law is constitutionally invalid even though the burden may be characterized as being only indirect.” n181 Most strikingly, in the seminal Everson decision the Court stated in dicta that, as a consequence of free exercise, a state “cannot exclude individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Non-believers, Presbyterians, or the members of any other faith, because of their faith, or lack of it, from receiving the benefits of public welfare legislation.” n182

Thus, the Court's gradual refinement of the non-exemption rule seemed to reveal a corollary condemning laws that were not general but were instead targeted at particular faiths or at religion generally. So, in Cantwell v. Connecticut, the Court could affirm the state's power to regulate, by general and non-discriminatory legislation, the time, place and manner of door-to-door solicitation, while, at the same time, striking down the discriminatory application of that rule to Jehovah's Witnesses on free speech and free exercise grounds.n183 The [*537] licensing scheme struck down in Cantwell effectively empowered local officials to clamp down on religious solicitation that the officials deemed did not “conform[] to reasonable standards of efficiency and integrity.”n184 It is easy to see how such an unbounded power could be used, as it was in Cantwell, in the service of discriminating against unpopular or marginal faiths.

Similarly, in Torcaso v. Watkins, the Court took a non-discrimination approach to Maryland's requirement that state officeholders make a “declaration of belief in the existence of God” or forfeit their right to office.n185 In the Court's view, the oath requirement placed “the power and authority of the State of Maryland ... on the side of one particular sort of believers [sic] - those who are willing to say they believe in “the existence of God.'“ n186 The Court struck down the requirement under free exercise, explaining that “neither a State nor the Federal Government ... . can constitutionally pass laws or impose requirements which aid all religions as against non-believers, and neither can aid those religions based on a belief in the existence of God as against those religions founded on different beliefs.” n187 There is but a small step - no step at all, really - from this reasoning to the notion that government also cannot express raw preferences for the non-religious over the religious in marking off political categories.

It comes as no surprise, then, that non-discrimination based on [*538] religious affiliation or status was the controlling factor in McDaniel v. Paty, unanimously striking down Tennessee's practice of excluding ministers from public office.n188 The Tennessee Constitution embodied the last hold-out of that discredited practice, which dated back to the early republic. n189 The dispute in McDaniel arose when Tennessee tied eligibility to be a delegate at its 1977 constitutional convention to eligibility to be a state representative, by implication excluding ministers from the constitutional convention. n190

The Supreme Court unanimously invalidated Tennessee's clergy-disqualification provision. Chief Justice Burger's opinion, for a four-Justice plurality, struck down the provision under the Free Exercise Clause alone. Burger found that right to free exercise encompassed the right “to be a minister,” and he reasoned that the clergy-exclusion wrongly forced a minister to choose between that free exercise right and his right to hold state office recognized by the Tennessee Constitution.n191 Additionally, while Burger did not find that the [*539] exclusion targeted beliefs as such - in which case the law would have been absolutely prohibitedn192 - he did conclude that it targeted “status as a “minister' or “priest,'“ a status defined by religiously affiliated and motivated conduct. n193 Burger then explained that the disqualification, targeted as it was at a religiously defined status, could only escape invalidation if it were justified by compelling interests. n194 Significantly, Burger rejected Tennessee's asserted interest in “preventing the establishment of a state religion,” a goal Tennessee sought to shelter under the federal Establishment Clause. n195 While Tennessee's fears about the influence of clergy on politics were once “held in the 18th century by many, including enlightened statesmen of that day,” Burger reasoned that those fears had been overwhelmingly found baseless and provided no justification for continuing to burden ministers' free exercise rights today. n196

Concurring, Justice Brennan, joined by Justice Marshall, would have gone beyond the plurality opinion and found the clergy disqualification absolutely prohibited under Torcaso as a “religious classification ... governing the eligibility for office.”n197 Brennan's opinion was strongly influenced by his perception that the ministerial exclusion was essentially a religious discrimination, “imposing a unique disability upon those who exhibit a defined level of intensity of involvement in protected religious activity.” n198

[*540] McDaniel remains a vital precedent that forbids government from “imposing special disabilities on the basis of religious views or religious status.”n199 The decision is strong evidence of the non-persecution principle in that it particularly disfavors laws that impose disabilities on religious status - and more precisely on the behavior that is associated with the status - specifically because of its connection to religion. Significantly, McDaniel also treats with skepticism any justification for targeting religious affiliation based on discarded historical attitudes about religion that are incompatible with properly understood principles of religious freedom, or that are themselves of doubtful historical lineage. Finally, notice what little separated the plurality and concurring Justices - four subjected the law to strict scrutiny as “religious conduct discrimination,” while Brennan, Marshall and Stewart would have summarily invalidated the law as a “religious belief discrimination.”

The foregoing jurisprudential foundations for the non-exemption and non-persecution rules set the stage for the clearest interaction of those rules in two decisions from the 1990s. Those were Employment Division Department of Human Resources v. Smith - reaffirming and [*541] clarifying the non-exemption rule - and Church of theLukumi Babalu Aye, Inc. v. City of Hialeah - reaffirming and clarifying the non-persecution rule. Each decision reinforced the strength of the non-persecution rule and placed it in the context of the Court's overall Free Exercise jurisprudence.

1. Smith and Peyote

In Smith, the Court confronted whether Oregon could “include religiously inspired peyote use within the reach of its general criminal prohibition on use of that drug,” and could, consequently, deny unemployment benefits to persons who had been fired for using the drug sacramentally during a Native American Church ceremony.n200 In deciding that Oregon could do so without violating the Free Exercise Clause, the Court focused on the general nature of the criminal peyote prohibition, repeatedly characterizing it as a “neutral” or “generally applicable law.” n201 “Generally applicable” laws were explicitly contrasted with laws that “were specifically directed against” or that “discriminated against” religious behavior. n202 The Court recognized that the religious free exercise protected by the First Amendment often extends to physical acts - listing as examples devotional or otherwise religion-motivated actions such as “assembling with others for a worship service, participating in sacramental use of bread and wine, proselytizing, abstaining from [*542] certain foods or certain modes of transportation.”n203 Further demonstrating what a generally applicable law does not do, the Court hypothesized the following scenario:

It would be true, we think (though no case of ours has involved the point), that a State would be “prohibiting the free exercise [of religion]” if it sought to ban such acts or abstentions only when they are engaged in for religious reasons, or only because of the religious belief that they display. It would doubtless be unconstitutional, for example, to ban the casting of “statues that are to be used for worship purposes,” or to prohibit bowing down before a golden calf.n204

Finally, the Court also relied on the text of the Free Exercise Clause to develop that distinction. The Court explained that the clause could plausibly be read “to say that if prohibiting the exercise of religion ... is not the object of the [law] but merely the incidental effect of a generally applicable and otherwise valid provision, the First Amendment has not been offended.”n205

Smith thus solidified a sharp distinction between “generally applicable” or “across-the-board” laws that are not targeted at religious behavior but may incidentally burden it, and laws that are in fact “religion sensitive” - i.e., where the very operation penalizes behavior because of its connection to religious belief or practice. Smith's ruling thereby suggests that the way laws structure their burdens is constitutionally determinative: If a law structures its burdens deliberately to fall on religious conduct alone, then it is not generally applicable. Three years later in its Lukumi decision, the Court reinforced that distinction and demonstrated that laws of this variety - imposing religion-sensitive burdens - presumptively violate free exercise rights.

2. Lukumi and Animal Sacrifice

While the Court was evaluating judicial exemptions for religious peyote use in Smith, the Lukumi case was still working its way through the federal courts. Supporting the non-exemption rule, the Smith Court cited the federal district court's 1989 opinion in Lukumi. The Court did so merely to give an example of one of the many kinds of general civic obligations - in Lukumi, animal cruelty laws - that ought not to be forced by the Free Exercise Clause to exempt religious conduct that has been only incidentally burdened.n206 But in [*543] 1993, when the Court examined the animal cruelty laws at issue in Lukumi, it discovered that, on closer inspection, those laws were in fact a coordinated web of prohibitions and exceptions deliberately designed for one purpose - to criminalize the ritual sacrifices performed by adherents of the Santeria religion.n207 Thus, Lukumi allowed the Court to refine the distinction between generally applicable laws on the one hand, and, on the other, those rarer instances of laws whose “object or purpose ... is the suppression of religion or religious conduct.” n208

The exercise of Santeria - a fusion of Roman Catholicism with traditional African religious practices - involves ritual animal sacrifice. As the Santeria Church of the Lukumi Babalu Aye was preparing to begin worship in the southern Florida community of Hialeah, the Hialeah city council held an emergency session, during which it passed a number of resolutions and ordinances concerning animal cruelty and ritual sacrifice.n209 None of the ordinances passed to further the resolutions mentioned Santeria by name, n210 but, as the Court would remark in the course of its opinion invalidating them, “almost the only conduct subject to [the ordinances] is the religious exercise of Santeria church members.” n211

In essence, Lukumi announced no new rule of religious liberty. But by articulating and reinforcing the non-persecution rule implicit in the text and structure of the religion clauses, and developed throughout the Court's jurisprudence, Lukumi brings a necessary doctrinal balance to Smith. In that sense, Lukumi confirms that Smith's non-exemption rule has teeth - it may allow religious conduct to suffer incidental burdens but it draws a non-negotiable line at laws that target religion for specially tailored burdens. Reflecting this balance, at the outset of Lukumi, the Court reiterated the overarching standards from Smith:

Our cases establish the general proposition that a law that is [*544] neutral and of general applicability need not be justified by a compelling governmental interest even if the law has the incidental effect of burdening a particular religious practice. Neutrality and general applicability are interrelated, and, as becomes apparent in this case, failure to satisfy one requirement is a likely indication that the other has not been satisfied. A law failing to satisfy these requirements must be justified by a compelling governmental interest and must be narrowly tailored to advance that interest.n212

Elaborating further, the Court explained that minimal free exercise standards are violated when a law “discriminates against some or all religious beliefs or regulates or prohibits conduct because it is undertaken for religious reasons,” and that instances of such “religious persecution” lie at the historical roots of the clause.n213 A law is not neutral under the Smith standards if its object is to “infringe upon or restrict practices because of their religious motivation.” n214 A law blatantly violates neutrality when it “discriminates on its face,” by, for instance, “referring to a religious practice without a secular meaning discernable from the language or context.” n215 But a law may advance its discriminatory object more subtly - engaging in “masked” or “covert suppression of particular religious beliefs” n216 - when its operation “targets religious conduct for distinctive treatment.” n217 To illuminate what it meant by covert discrimination, the Court quoted a directive from its Establishment Clause jurisprudence: “The Court must survey meticulously the circumstances of governmental categories to eliminate, as it were, religious gerrymanders.” n218

The Court unanimously concluded that the Hialeah ordinances violated these fairly straightforward standards of non-persecution because, essentially, the ordinances prohibited a form of conduct (animal killing) only when performed in observance of the Santeria religion. The ordinances were carefully structured to exempt every other form of animal killing that could conceivably fall within their [*545] prohibitions - for instance, large-scale slaughterhouses, small-scale farm slaughter, kosher butchers, and hunting.n219 The Court characterized this as a religious “gerrymander” whose effect was “that few if any killings of animals are prohibited other than Santeria sacrifice, which is proscribed because it occurs during a ritual or ceremony and its primary purpose is to [fulfill Santeria religious requirements], not food consumption.” n220 The ordinances, therefore, were not neutral because they “had as their object the suppression of religion.” n221 Therefore the Court applied strict scrutiny to the ordinances, citing McDaniel and Smith, and candidly acknowledging that “[a] law that targets religious conduct for distinctive treatment or advances legitimate governmental interests only against conduct with a religious motivation will survive strict scrutiny only in rare cases.” n222 Unsurprisingly, given the plain object and operation of the Hialeah ordinances, Lukumi was not one of those rare cases. n223

Justice Scalia's concurrence, joined by Chief Justice Rehnquist, sheds additional light on Lukumi's analysis, particularly because Scalia wrote Smith. Scalia clarified that the “terms “neutrality' and “general applicability' are not to be found within the First Amendment itself,” but instead have been used by the Court “to describe those characteristics which cause a law that prohibits an activity a particular individual wishes to engage in for religious reasons nonetheless not to constitute a “law ... prohibiting the free exercise' of religion within the meaning of the First Amendment.”n224 In Scalia's view, laws are not neutral in that sense when “by their terms [they] impose disabilities on the basis of religion (e.g., a law excluding members of a certain sect from public benefits).” n225 By [*546] contrast, laws lack general applicability when, “though neutral in their terms, through their design, construction, or enforcement [they] target the practices of a particular religion for discriminatory treatment.”n226 Scalia allowed that his line between these two qualities of discriminatory laws was “somewhat different” from the one drawn in Justice Kennedy's majority opinion, but he judged the distinction inconsequential because the categories overlapped significantly. n227

3. Summary: Non-Persecution and Free Exercise

The consistent rejection in the Court's free exercise jurisprudence of laws that target religious conduct for special disabilities - laws that impose religion-sensitive penalties - undergirds the non-persecution principle. The Court has long recognized that the laws of a pluralist society will inevitably intrude on certain behavioral demands that religions make of their adherents. In early cases like Reynolds and Davis, Mormons' religious obligation to engage in polygamous marriages had to give way before society's different conception of marital limits. Over a century later in Smith, Native Americans' celebration of a sacrament of their religion bowed before society's need to regulate harmful substances. But there is a deeper principle at work governing the burdens society may legitimately place on religious conduct, one evident in the parameters of the non-exemption [*547] rule itself. For that rule coherently operates only in the context of laws that further legitimate governmental goals through “neutral and generally applicable” means and that, by definition, place burdens on religiously motivated conduct only incidentally. In other words, the Court has always premised the soundness of the balance struck in the non-exemption rule on the notion that the laws in question circumscribe conduct for legitimate reasons independent of its religious affiliation or motivation. Once laws begin to impose burdens based on whether a status, organization, or behavior is connected to religion, then the entire basis for the non-exemption rule crumbles.

Gerard Bradley has persuasively explained the intersection between these two complementary lessons. Commenting on the relationship between Smith and Lukumi, Bradley argues that “those cases stand for the proposition that where an action is legitimately generally prohibited, the Constitution does not require different treatment for believers who engage in the activity for religious reasons, or for the religious significance they see in or attach to it.”n228 But the necessary corollary to this rule, Bradley is careful to add, flows from what I have described as the backbone principle of non-persecution: “Where public authority generally permits an activity - say, slaughtering animals - it may not discriminate against persons who would engage in the activity for religious reasons or for the religious significance they see in or attach to it.” n229 Thus, we can broadly say that the Free Exercise Clause does not withhold from government the power to prohibit all polygamy, but does withhold power to prohibit Mormon polygamy only or polygamy engaged in “for religious purposes.” Government may forbid peyote use across-the-board for the religious and non-religious alike, but it may not prohibit the “ritual” or “sacramental” use of peyote while exempting all other uses. Eligibility for public office may be regulated based on any number of general criteria (age, citizenship, and criminal record come to mind), but eligibility may not be premised on the nature of a person's connection to religion or to a person's role in a church. Government may enact generally applicable public health rules for animal slaughter and disposal, but it may not tailor those rules to target religious animal slaughter only, while leaving the butcher, the farmer, and the hunter inexplicably unregulated.

What counts here is whether religion is the triggering mechanism for the burden imposed. The distinction between legitimate and illegitimate burdens on religious practice shows that the constitutional defect arises when categorizations such as “religious,” “religious affiliation,” or “religious purposes” are used as the organizing principle for imposing legal disabilities. “Incidental” burdens - those [*548] which, in a sense, accidentally occur only because general laws may conceivably burden someone's religious practice in a religiously plural society - are constitutionally permissible. But laws that reserve their burdens for religious conduct only - “religious gerrymanders,” in Justice Harlan's phrasen230 - are impermissible because, in allocating the burdens and benefits of society's laws, they force religiously motivated conduct alone to bear the burdens and forego the benefits. The Free Exercise Clause condemns such laws because, as Michael McConnell explains, “the free exercise principle “singles out' religion for special protection against governmental hostility or interference.” n231

Notice how the subtle ripening of the non-persecution principle, as seen in the long progression from Reynolds in 1878 to Lukumi in 1993, reinforces the idea that, at bottom, precisely what non-persecution prohibits is invidious religious categorization. Reynolds seemed to stingily protect only Mormons' religious opinions and leave their actions entirely open to legal prohibition, provided they were “in violation of social duties or subversive of good order.”n232 In 1890, Davis perhaps promised slightly more protection - shielding not only “man's relations to his Maker and the obligations he may think they impose,” but also “the manner in which an expression shall be made by him of his belief on those subjects.” n233 Like Reynolds, Davis also recognized the trumping power of criminal law, but added that such laws must be “passed with reference to actions regarded by general consent as properly the subjects of punitive legislation.” n234 Looking forward to Gobitis in 1940, we find the Court suggesting that “religious liberty” is offended by laws “directed against the doctrinal loyalties of particular sects” or laws “aimed at the promotion or restriction of religious beliefs.” n235 A short seven years later gives us the Court's striking dicta in Everson that free exercise prohibits states from “excluding individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Non-believers, Presbyterians, or the members of any other faith, because of their faith, or lack of it, from receiving the benefits of public welfare legislation.” n236 Braunfeld, in 1961, condemned laws imposing even indirect burdens on religious practice if their “purpose or effect” was “to impede the observance of [*549] one or all religions” or “to discriminate invidiously between religions.”n237 In 1978, McDaniel invalidated laws targeting religious status - in the sense of conduct or activity affiliated with religion - for special disabilities. n238 And, in the 1990s, Smith and Lukumi solidified the prohibition against laws that impose disabilities on a category defined in religious terms. n239

This can plausibly be viewed as a progression of free exercise principles from simply forbidding laws targeting religious beliefs, to forbidding encroachments on religious observance and practice, to forbidding exclusions based on religiously motivated conduct, status, and affiliation. Overall, the movement has been toward forbidding invidious religious categorization altogether. The elaboration of “general” versus “targeted” laws in Smith and Lukumi cannot be properly understood apart from this matrix of free exercise decisions stretching back over a century. And Lukumi explicitly invoked that long history when it glossed “religious persecution” as laws that “discriminate[] against some or all religious beliefs or regulate[] or prohibit[] conduct because it is undertaken for religious reasons.”n240 Thus, the Court does not invoke the loaded term persecution carelessly or outside the context of its own jurisprudence, and it has not suggested that the term is confined to the grossest instances of official religious discrimination. Understanding the term's proper place in free exercise jurisprudence shows that persecution is constitutionally accomplished by the more sophisticated method of an invidious classification based on religion alone. n241

[*550] In the next section, I will examine how principles from Court's non-establishment and free speech jurisprudence reinforce and round out the scope of this non-persecution rule. But it will be useful to pause at this point and assess the State Blaines in light of the basic tenets of non-persecution drawn from the Court's free exercise cases. Those tenets call the obvious textual applications of the State Blaine Amendments into serious question.n242 All State Blaines explicitly single out religious purposes, religious institutions, and religious affiliation for exclusion from otherwise generally available public benefits. The object of the amendments, which is plain on the face of all the State Blaines, is to place religion at a civil disadvantage with respect to all conduct, institutions, and persons that are “non-religious.” In doing so, the State Blaines explicitly exclude themselves from the category of “neutral and generally applicable laws” - the only kind of laws which, under the Free Exercise Clause, may place burdens on religious conduct. Like the clergy exclusion in McDaniel, the State Blaines force persons whose behavior or status affiliates them with religion to choose between adhering to that affiliation and receiving public benefits to which eligible “non-religious” persons are entitled. Like the animal sacrifice laws in Lukumi, the State Blaines tailor their burdens and exclusions to conduct that is undertaken for religious reasons - only the State Blaines add to that the additional defect of discriminating against religion openly. n243

[*551]

B. “Neutrality” and Non-Persecution

It is often stated that the religion clauses demand that laws be “neutral” toward religion.n244 The notion continues to play a major conceptual role in the Supreme Court's non-establishment jurisprudence. But “neutrality” is an incomplete and open-ended term; as Douglas Laycock observes, “those who think that neutrality is meaningless have a point. We can agree on the principle of neutrality without having agreed on anything at all.” n245 Yet Laycock rightly does not dismiss neutrality as an intelligible concept - indeed, he argues that one of the jurisprudential roots of religious non-discrimination lies in the Court's repeated assurances over the last two decades that the Constitution mandates government neutrality toward religion. n246 Neutrality, in short, has something to tell us about the non- [*552] persecution principle and, in turn, how that principle applies to the State Blaines.n247

Among scholars of American religious liberties, there are two prominent competing views of what a principle of neutrality toward religion requires. My purpose is not to choose one over the other.n248 Instead, my modest point is that either view of neutrality supports the non-persecution principle gleaned from the Court's free exercise jurisprudence. I will briefly demonstrate that the Court has often suggested as much - i.e., that religious discrimination is inconsistent with any plausible notion of government neutrality toward religion - when elaborating the requirements of neutrality in its non-establishment cases.

One account of neutrality posits that the religion clauses are co-belligerents in the cause of promoting religious freedom: free exercise forbids discrimination against particular religions and against religion generally, while non-establishment “prevents the government from using its power to promote, advocate, or endorse any particular religious position.”n249 Douglas Laycock has coined the influential term “substantive neutrality” to capture this notion - i.e., that “the religion clauses require government to minimize the extent to which it either encourages or discourages religious belief or disbelief, practice or nonpractice, observance or nonobservance.” n250 The religion clauses, so often accused of being in tension, should instead be read holistically as mutually reinforcing guarantees of positive religious liberty. n251 Seen that way, “most of the tension between them disappears. They are complementary provisions, both in the service of the same fundamental right. They bar Congress from abridging religious freedom in one specific way (by legislation “respecting an establishment of religion'), and in general (“or prohibiting the free exercise thereof').” n252 In a similar vein, Michael McConnell explains that “the Free Exercise and Establishment Clauses serve a complementary function: to reduce the power of government over [*553] religion, whether to help, hurt, or control, to the greatest extent consistent with the achievement of legitimate secular objectives.”n253

A competing notion of religious neutrality is “formal neutrality.” This view holds that “government cannot utilize religion as a standard for action or inaction,” because the unified thrust of the Free Exercise and Establishment Clauses “prohibits classification in terms of religion either to confer a benefit or to impose a burden.”n254 Formal neutrality thus draws a strikingly different inference from the complementarity of free exercise and non-establishment. Although it reads the clauses as stating a single precept, n255 that precept directs government not merely to avoid interfering with religion, but rather to adopt a mechanistic evenhandedness toward religion, “without regard to whether such evenhandedness helps or hinders religion.” n256

These two views of neutrality make a difference on some important issues. For instance, does the Establishment Clause allow legislatures [*554] to make specific exemptions from laws for religiously-motivated behavior or religious organizations? A “substantively neutral” view would hold that, generally speaking, government may (or perhaps must) do so, and this, indeed, is how the issue has been resolved historically in American legislatures and courts.n257 A “formally neutral” view would reject any special religious exemptions by courts or legislatures. Smith indicates that the Supreme Court was guided by concerns with formal neutrality when deciding whether religious behavior should receive judicial exemptions from generally applicable laws. n258 At the same time, Smith did not wholly embrace formal neutrality, as the opinion itself approves of legislative exemptions. n259 Many proponents of substantive neutrality have, nonetheless, criticized Smith. n260

For present purposes, I need not resolve the tensions between formal and substantive neutrality. Why? On either account of neutrality, laws that explicitly target religion for special disabilities are non-neutral. Such laws violate substantive neutrality because they promote not religious freedom, but hostility toward religion: their object and effect is to demote and penalize religious belief, behavior, or association.n261 Such laws violate formal neutrality for formal [*555] reasons; they use religion as a category for imposing legal burdens.n262 Either conception of neutrality, then, would forbid religious discrimination and therefore accords with the general non-persecution principles under the Court's free exercise jurisprudence. A brief look at the Court's treatment of neutrality (whether that treatment reflects a more formal or more substantive view of neutrality) in its non-establishment cases will demonstrate that idea.

Neutrality as religious non-hostility can be seen as one fixed star in the otherwise untidy constellation of the Court's non-establishment cases. The Establishment Clause is neutral toward religion in that it does not “compel the exclusion of religious groups from government benefit programs that are generally available to a broad class of participants.”n263 But the Court has often suggested that neutrality goes beyond merely “not compelling” religious exclusion; neutrality affirmatively condemns governmental hostility toward religion itself. n264 As Justice O'Connor has observed, “The Religion Clauses prohibit the government from favoring religion, but they provide no warrant for discriminating against religion.” n265 For instance, neutrality means that government may not deliberately skew how it distributes aid either in favor of or against religious recipients. n266 In other words, [*556] no particular universe of aid recipients may be defined in a way that religious groups get more aid because they are religious groups; conversely, because potential recipients are religious groups, they may not designedly get less.

This religion-friendly side of neutrality is most clearly distilled in the doctrine that laws violate the federal Establishment Clause if they deliberately “inhibit” religion.n267 The notion runs back to the seminal Establishment Clause decision, Everson itself, which declared that “state power is no more to be used so as to handicap religions than it is to favor them.” n268 Everson also closely links this aspect of non-establishment jurisprudence to the Free Exercise Clause. n269 None of this is to say, however, that the most comfortable argument against religiously-hostile laws lies in the Establishment Clause proper. The Court has rarely, if ever, applied the “inhibition” prong, and there is some doubt as to the coherence of the argument that government disapproval of religion somehow establishes religion. n270 Furthermore, [*557] four members of the current Court have recently suggested that “to require exclusion of religious schools from [a genuinely neutral aid program] would raise serious questions under the Free Exercise Clause.”n271 My narrower purpose is to point out that, like free exercise jurisprudence, non-establishment jurisprudence contains a background assumption that laws violate basic canons of legitimacy when they purposefully single out religion for disfavored treatment. This background assumption is evident in much of the Court's elaboration of the neutrality requirement, as the following examples underscore.

Even when forbidding Bible reading in public schools in School District v. Schempp - a decision regarded by some as an apogee of Court-imposed separationismn272 - the Court emphasized that the Establishment Clause did not sanction purposeful religious discrimination. Constitutional limits of legislative power were transgressed, the Court said, if the “purpose and the primary effect of the enactment” is “either the advancement or inhibition of religion.” n273 Justice Goldberg's concurrence better articulated this idea, explaining that “the fullest realization of true religious liberty requires that government neither engage in nor compel religious practices, that it effect no favoritism among sects or between religion and nonreligion, and that it work deterrence of no religious belief.” n274 [*558] Plainly absent from this conception of neutrality was any justification for governmental hostility toward religion.

That benevolent view of neutrality was prominent in Walz v. Tax Commission, a decision which validated the venerable practice of granting tax exemptions to churches.n275 Walz stated categorically that “the general principle deducible from the First Amendment and all that has been said by the Court is this: that we will not tolerate either governmentally established religion or governmental interference with religion.” n276 In a thoughtful concurrence, Justice Harlan articulated two related concepts underlying the Court's application of the religion clauses - “neutrality” and “voluntarism.” n277 By voluntarism, Harlan meant the principle that “legislation neither encourages nor discourages participation in religious life.” n278 Harlan saw in neutrality an “equal protection mode of analysis,” requiring the Court to “survey meticulously the circumstances of governmental categories to eliminate, as it were, religious gerrymanders.” n279 These concepts were, as Harlan explained, “short-form for saying that the Government must neither legislate to accord benefits that favor religion over nonreligion, nor sponsor a particular sect, nor try to encourage participation in or abnegation of religion.” n280 Here, again, we have on display a relationship between government and religion - positively sanctioned by the interplay of the religion clauses - that forbids government from acting either as God's patron or as God's persecutor.

One thus sees that neutrality, which is central to the Court's non-establishment jurisprudence, is itself bottomed on the twin commands [*559] that government neither favor nor disfavor religion. But what does neutrality add to the non-persecution principle I have already discussed? Principally, neutrality should foreclose the notion that the free exercise and establishment clauses are somehow in tension with each other on the substantive issue of government religious hostility. The proper interaction of the clauses regarding religious benefits may still be murky, but their interaction on religious hostility is clear - both categorically condemn it. Secondly, neutrality reinforces the proposition that it is invidious governmental religious categories themselves that impinge on religious freedom. It is the government categorization that must be scrutinized - i.e., how the government has chosen to structure the exclusions and inclusions in its scheme of distributing benefits. When it is apparent that government has engaged in religious gerrymandering by creating a category of beneficiaries designed to exclude “religious persons” or “religious entities,” then government has likely fallen short of the neutrality that the Establishment Clause specifically, and the religion clauses more generally, demand.

Does this mean that government is constitutionally forbidden from ever conferring a special benefit on religious persons? Or does this mean that government may allow certain narrow exemptions from general laws for religious reasons? These hard questions throw us back on the original debate discussed previously over formal versus substantive neutrality. And regardless of the resolution of that debate, one concept unites both sides: Government may not confer special disabilities on religious persons or entities through its structuring of beneficiary categories. That much should be clear from the overlap between the two competing theories of neutrality, and also from the Supreme Court's consistent condemnation of categories explicitly disfavoring religion. There is, in short, some real substance behind the Court's label of neutrality as “benevolent.” Whatever benevolence may mean regarding government's favoring of religion, the idea plainly excludes governmental categories that embody malevolence toward religion.

C. Free Speech and Non-Persecution

Over the last two decades, the Supreme Court has consistently validated the “fundamental First Amendment proposition that government may not discriminate against individuals' or groups' speech on account of its religious nature or the speaker's religious identity.”n281 Two aspects of this religious speech jurisprudence reinforce the non-persecution principle that government may not target religion for special disabilities in distributing public benefits. n282 [*560] First, the Court's treatment of laws targeting religious viewpoints for exclusion from limited public fora echoes the Court's approach to non-persecution in the free exercise context and to neutrality in the non-establishment context. Second, the Court has consistently rejected as justifications for religious viewpoint discrimination both exaggerated fears of violating the federal Establishment Clause and also states' interests in crafting greater church-state separation. Each of these points reinforces my general argument that an overarching non-persecution principle forbids most of the obvious applications of the State Blaine Amendments.

Since the early 1980s, the Court has repeatedly addressed variations on the following general theme: A governmental body creates a limited public forum for the discussion or dissemination of a broadly defined range of topics, but it explicitly excludes participants if they bring speech or ideas of an overtly religious character. Thus, in Widmar v. Vincent, the University of Missouri opened its facilities to any student discussion group, but disallowed facility access to any student group that would engage in religious worship or discussion.n283 Similarly, in Lamb's Chapel v. Center Moriches Union Free School District, a local school board made public school property available for after-school use for “social, civic and recreational meetings” and other “uses pertaining to the welfare of the community,” while excluding “meetings for religious purposes.” n284 The school board applied that policy to forbid a group from showing a film that discussed child-rearing from an explicitly Christian perspective. n285 More recently, in Good News Club v. Milford Central School, an elementary school opened its facilities for the same range of uses as in Lamb's Chapel but refused to allow a Christian organization access [*561] for after-school meetings that involved religious instruction and activities.n286 Finally, in Rosenberger v. Rector and Visitors of the University of Virginia, the University established a Student Activity Fund that provided indirect financial assistance to a wide array of student publications. A student newspaper with an explicitly Christian viewpoint qualified to participate in the Fund but was denied access because of the religious content of the newspaper. n287 In each of these cases, the governmental body claimed that it could legitimately deny equal participation in otherwise generally available benefits - here, participation in a limited public forum - because of the avowedly “religious” content or affiliation of certain groups. But, in every case, the Supreme Court invalidated the religious exclusion as viewpoint discrimination under the Free Speech Clause and, moreover, refused to justify the discrimination under any theory of non-establishment. n288

The Court's consistent invalidation of the religious speech exclusions in these cases resonates with the general non-persecution principle. In each case, the governmental unit had created a “limited public forum,” opening its facilities to a broad but defined range of speakers or topics.n289 For instance, in Lamb's Chapel and Good News Club, the school boards had opened their facilities under a New York education law that allowed after-school meetings for ““social, civic and recreational meetings and entertainments, and other uses pertaining to the welfare of the community,'“ provided that such meetings were [*562] ““non-exclusive'“ and ““open to the general public.'“n290 Similarly, in Rosenberger the Student Activity Fund guidelines authorized fund access to “student news, information, opinion, entertainment, or academic communications media groups.” n291 But, in those cases the relevant access provisions mandated explicit exclusions for groups with religious purposes or content. n292 Consequently, in each case a student organization was admittedly eligible for participation in the limited forum because it fell within the forum's defined scope, but the group was nonetheless excluded from participation specifically because of its religious affiliation or religious purposes.

The Court has consistently condemned these exclusions as impermissibly discriminating on the basis of religious viewpoint. While government may permissibly limit the speakers in a limited public forum according to subject matter and speaker identity, such exclusions must be “reasonable in light of the purpose served by the forum and [must be] viewpoint neutral.”n293 In each case, participation [*563] was denied for no reason “other than the fact that the [speech] would have been from a religious perspective,”n294 and the exclusion therefore plainly amounted to forbidden viewpoint discrimination. As explained in Rosenberger, “by the very terms of the [Student Activity Fund] prohibition, the University does not exclude religion as a subject matter but selects for disfavored treatment those student journalistic efforts with religious editorial viewpoints.” n295 The Court categorically rejected the use of concepts like “religion,” “religious purpose” and “Christian viewpoint” as legitimate organizing principles for the exclusion of groups and speech from participation in the limited public fora. n296

The parallels between the Court's reasoning in these cases and its approach to religious neutrality and non-discrimination in its religion clause jurisprudence are unmistakable. The Court itself has referred to its treatment in these cases of public fora to illustrate the proper scope of religious neutrality in the Establishment Clause area.n297 Justice O'Connor made that connection explicit when, in her Rosenberger concurrence, she observed that the Court's “insistence on government neutrality toward religion explains why we have held that schools may not discriminate against religious groups by denying them equal access to facilities that the schools make available to all,” citing [*564] Lamb's Chapel and Widmar as examples.n298 The Rosenberger majority was operating on the same premise, as evidenced by its concluding statement that “the neutrality commanded of the State by the separate clauses of the First Amendment was compromised by the University's course of action.” n299 Further clarifying the connection, the Court went on to explain that “the viewpoint discrimination inherent in the University's regulation ... was a denial of the right of free speech and would risk fostering a pervasive bias or hostility to religion, which could undermine the very neutrality the Establishment Clause requires.” n300 In sum, the overarching principle in these cases is that religious speech - just like religious conduct and status - may not be excluded from the public arena simply because it is religious. “Religious” cannot be the organizing principle or the basis for classification that results in some speech or ideas being denied entry into an otherwise accessible public forum. n301

Significantly, these cases also reject “unreasonable fears of establishment” as a justification for excluding religious speech from limited public fora. The governmental units attempted to justify their religious discrimination by raising their “interest in not violating the Establishment Clause” or their “compelling interest in maintaining strict separation of church and state.” And in every case, the Court rejected that argument by concluding that allowing the religious groups to participate in the public fora was not even a colorable violation of the Establishment Clause.n302

Moreover, in Widmar, the University of Missouri also grounded its discriminatory policy on the Missouri Blaine Amendment, which the University asserted “had gone further than the Federal Constitution in proscribing indirect state support for religion.”n303 The Court [*565] approached this claim cautiously, first observing that the Missouri courts had not determined whether “a general policy of accommodating student groups, applied equally to those wishing to gather to engage in religious and nonreligious speech, would offend the State Constitution.”n304 Declining to resolve that issue, the Court also passed over whether the Supremacy Clause would override a more restrictive state policy toward religious accommodation. n305 But, in tension with those preliminary comments, the Court concluded that:

The state interest asserted here - in achieving greater separation of church and State than is already ensured under the Establishment Clause of the Federal Constitution - is limited by the Free Exercise Clause and in this case by the Free Speech Clause as well. In this constitutional context, we are unable to recognize the State's interest as sufficiently “compelling” to justify content-based discrimination against respondents' religious speech.n306

Thus, although the Court seemed to go out of its way to avoid addressing any conflict between the Missouri Constitution and the federal Constitution, its conclusion plainly favored federal religious and free speech rights.

In sum, the Court's consistent protection of religious speech against targeted exclusion from limited public fora - including a public forum in Rosenberger defined by a neutral funding mechanism - reinforces the non-persecution principle. First, the religious speech cases underscore the basic idea that religion - whether religiously motivated conduct, religiously affiliated persons or groups, or speech from a religious viewpoint - cannot be singled out for exclusion from participation in public benefits or public fora to which it would otherwise be permitted. Second, and relatedly, the religious speech cases reinforce the point that it is the invidious religious classifications themselves that are constitutionally suspect and per se disfavored. Third, they make the important additional point that religious discrimination can neither be justified by erroneous conclusions about the scope of Establishment nor by pretensions at creating a stricter separation at the state level. Michael Stokes Paulsen has concisely summed up the lessons taught and the principles reinforced by this line of cases: “The Establishment Clause does not authorize, and the Free Speech and Free Exercise Clauses do not permit, government discrimination against religious speakers or religious speech on the basis of religious content, viewpoint, or speaker identity - ever.”n307

[*566]

V. The State Blaines and Non-Persecution

What remains is to apply the non-persecution principle described in Part IV to the State Blaines. This appears to be daunting, because, as Part III showed, the State Blaines cover a lot of ground. But, for constitutional purposes, that complexity can be misleading; what unites all State Blaines is the explicit object of separating public benefits from religious persons, institutions, and purposes. I will thus limit myself to assessing that operation of the State Blaines - i.e., whether they may block religious persons' and groups' access to generally available public benefits on the basis of their religious affiliation, status, or purpose. First, I will look at whether State Blaines may operate to prevent the flow of public aid to persons who wish to use the aid to further their religious education or training. That inquiry will take us back to the example that opened this Article - Larry Witters' plan to use public financial assistance to train for the ministry - as well as the situation presented in Davey v. Locke, a recent Ninth Circuit decision involving selective state funding of non-religious degrees that will be heard by the Supreme Court in December 2003.n308 In this first section, I take up general defenses to the operation of State Blaines grounded in federalism and in the Supreme Court's non-establishment jurisprudence itself. In the next section, I address whether a state's control over how and why it spends money can provide an additional justification for the State Blaines' religion-sensitive exclusion from equal participation in public benefits.

A. Educational Funding, Federalism, and Incorporation

I began this article with Larry Witters' dilemma and now return to it. Recall that Witters qualified for state educational aid because he was blind, and he wanted to use that aid for ministry training at a Christian college. The Supreme Court told Witters he could do so under the federal Establishment Clause, because the funds were distributed without reference to religion and because they ended up at a religious school solely as a result of Witters' private choice to use them there.n309 But on remand, the Washington Supreme Court blocked Witters' use of the funds under the Washington Blaine Amendment - forbidding public funds from being appropriated or applied to any religious worship, exercise or instruction. n310 Witters arguably fell within the plain terms of the prohibition, but the court [*567] added the case law gloss that “religious instruction” meant only instruction that was ““devotional in nature and designed to induce faith and belief in the student,'“ as opposed to instruction marked by the ““open, free, critical, and scholarly examination of the literature, experiences, and knowledge of mankind.'“n311 How does this application of a State Blaine fare under the non-persecution principle?

Notice that Witters' dilemma would arise under the plain terms of any number of other State Blaines. Utah's Blaine Amendment, for instance, enacts an identical ban on funding religious instruction.n312 Pennsylvania's and Virginia's Blaines specifically disallow grants or scholarships to students in a “theological seminary or school of theology” n313 or students in a school “whose primary purpose is ... to provide religious training or theological education.” n314 Nor does it take much hermeneutical imagination to conclude that Witters' situation implicates the use of public money to “aid,” “benefit,” “assist,” or “support” a “society,” “seminary,” “institution,” “association,” “instruction” or even a “purpose” that is “religious,” “sectarian,” “theological,” “denominational,” or “controlled by” a church or religious institution. Indeed, the more difficult task is to identify any State Blaine whose terms would clearly allow Witters' contemplated use of the funds. n315 The point is not that a court could leniently interpret any State Blaine to favor Witters - as noted above, interpretations have gone both ways - but rather that state constitutions are littered with provisions whose language invites Washington's separationist result.

That result does not fare well under the non-persecution principle. First, as applied to exclude Witters' use of the funds, a State Blaine does not operate as a generally applicable law that incidentally burdens religiously-motivated conduct. Instead, it would be a law that targets its disabilities at purpose, conduct, and affiliation because of their religious character. The funds in question were generally available funds - they were made available to Witters on a religion-neutral basis (he qualified for them because he was blind) - and nothing beyond the religion-sensitive prohibition in the State Blaine would prohibit his use of the funds for ministry training.n316 That [*568] religion-penalizing application of a State Blaine would therefore merit strict scrutiny under Smith and Lukumi. Notice, moreover, how the State Blaine's exclusionary operation fits precisely into the prohibition articulated, over forty years before those decisions, in Everson - it “excludes individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Non-believers, Presbyterians, or the members of any other faith, because of their faith ... from receiving the benefits of public welfare legislation.”n317 Notice further that the State Blaines target everyone on Everson's list except the “Non-believer,” thereby privileging the areligious and the irreligious over the religious.

Second, the State Blaine's application is patently non-neutral. Washington State has made a pool of state aid generally available to handicapped students, but the State Blaine operates to categorize the recipients of that aid according to whether they will use the aid for “religious” or “non-religious” instruction.n318 This is nothing other than a religious gerrymander. n319 A government benefit program has been structured to exclude religion because it is religious - a contemplated religious use is the sole disqualifying trigger. Aid is therefore distributed to disfavor religious persons and purposes.

Finally, the religious speech cases reinforce the analysis. In those cases, religious groups were eligible to participate in limited public fora, but they were excluded only because of their religious affiliation and viewpoint. The limited public fora in those cases are directly analogous to the neutrally-available educational funds in Witters.n320 [*569] Witters was eligible to receive the funds and the federal Establishment Clause presented no plausible impediment to his using them for religious purposes. But the State Blaine operated to disqualify him solely because his purposes were religious. It is no rejoinder that Witters involved funding and not speech. The simplest answer is that Rosenberger, too, involved a religious group's access to generally available funding. But the better answer is that Rosenberger logically applied to a discriminatory funding scheme the principles of religious non-persecution found in the earlier religious speech cases, in free exercise cases like Smith, Lukumi, and McDaniel, and in the neutrality principle consistently elaborated in the Court's non-establishment jurisprudence, going back to Everson itself.n321 Religious status, purpose, or affiliation may not be independently used to exclude persons from participation in public benefits.

Notice a further complicating factor in Witters' situation. The Washington Supreme Court suggested that its Blaine Amendment targeted only “devotional” religious purposes. That is, if Witters had wanted to use the funds to become a purely secular expert in comparative religion, the State Blaine would not have barred his use of the funds.n322 This distinction weakens the constitutional footing of the State Blaine even further. First, it arguably raises the stakes of religious discrimination from religiously-motivated conduct to religious belief itself - Witters is being excluded from using the funds not simply because of a generally religious purpose, but because he takes religion seriously enough to become a minister. n323 Second, it opens the State Blaine to an independent viewpoint discrimination challenge under the Free Speech Clause - the State Blaine is not merely excluding religion, but is excluding certain religious [*570] viewpoints.n324 Finally, it unmasks the religious bigotry lurking beneath the State Blaine: Washington will tolerate handing over its educational funds to those who engage in “open, free, critical, and scholarly examination of the literature, experiences, and knowledge of mankind,” but not to those who undertake religious instruction that is “devotional in nature and designed to induce faith and belief in the student.” n325

Witters essentially resurfaced as a statutory matter in the Ninth Circuit's recent decision in Davey v. Locke.n326 Davey is significant not only because it invalidates a fairly widespread statutory discrimination against religious education, n327 but also because the Supreme Court will hear the case in December 2003. Davey addresses Washington State's “Promise Scholarship,” an aid program begun in 1999 to help fund the first two years of college for high-achieving students from low-to middle-income families. n328 But the program specifically excludes from participation students who are “pursuing a degree in theology.” n329 Defending its program before the Ninth Circuit, Washington justified the theology exclusion by reference to the Washington State Blaine - the same provision that had frustrated Larry Witters' ability to study [*571] for the ministry over two decades ago.n330 The Ninth Circuit, in an opinion by Judge Rymer, declared the theology exclusion in the Promise Scholarship criteria unconstitutional under the Free Exercise Clause, relying on the religious non-discrimination principle derived mainly from Lukumi, McDaniel, and Rosenberger, and denying that the Washington Blaine could justify the religious discrimination. n331

It is hard to see any constitutional difference between the statutory exclusion for theology degrees in Davey, and the application of Washington's Blaine to bar Witters from using state funds for religious instruction. Both operate as laws that target religion - here, education that is affiliated with religion or has a religious purpose - for exclusion from otherwise generally available public aid. Neither imposes merely incidental burdens on religious conduct. Neither is neutral toward religion in any plausible sense, because both structure categories of public aid to remove beneficiaries who are motivated by religion or who simply direct their studies toward religious ends.n332 Both laws, then, violate the religious non-persecution principle and, under strict scrutiny, must be justified by a compelling state interest. n333

In a recent article, Ira Lupu and Robert Tuttle offer some thoughtful objections to the foregoing analysis.n334 They criticize what they call the “Free Exercise Clause approach” to attacking the State Blaines - roughly equivalent to the non-persecution principle - i.e., “that the state may not generically treat religious entities worse than secular ones.” n335 Principally, they say the argument proves too much, [*572] because “American constitutional law, federal and state, has for many years done exactly what this argument condemns.”n336 By this, they mean primarily that the federal Establishment Clause has often been interpreted to require government to “single out” religious entities for “special” treatment in many areas. For instance, government cannot directly subsidize religious indoctrination, nor can it intervene in church disputes involving matters of faith. n337 Thus, by attacking any rule drawing a “line between religious and nonreligious organizations,” the free exercise/non-persecution argument against State Blaines undermines, they say, “each and every religion-specific doctrine under the federal religion clauses.” n338

Lupu and Tuttle's second rejoinder, sounding in federalism, complains that the non-persecution argument is “hostile to notions of respect for state law, and in particular to the tradition of independent state constitutional law.”n339 They contend that, even if a narrower form of the non-persecution argument would salvage the religion-sensitive doctrines in federal constitutional law, it would still “deny the states any room whatsoever for their own church-state policy.” n340 In other words, states would be wrongly confined under a ceiling of federal non-establishment principles - they would have absolutely no room “to have a non-establishment policy broader than whatever five Justices of the U.S. Supreme Court find to be the content of federal law at any given moment.” n341 The authors' resolution of the federalism issue, by contrast, would leave “each state ... free to make its own constitutional policy of church-state relations, and to extend it beyond the federal policy, so long as the state approach serves reasonable purposes of the sort associated with the regime of Separationism.” n342

[*573] Lupu and Tuttle's objections go to the heart of the religious-liberty and federalism issues presented by the State Blaines, but ultimately they neither undermine the non-persecution principle nor save the State Blaines from constitutional invalidity. First and foremost, they largely reduce the non-persecution principle to the untenable formalist notion that laws may not “single out” religion for any purpose whatsoever. But the non-persecution principle condemns a different, narrower kind of legal categorization - it forbids singling out religion for disfavored treatment and, in the context of the State Blaines, disfavored treatment of the kind that excludes persons and organizations from participation in public benefits only because they are somehow religious. Second, it is reductionist to claim that the Supreme Court has generally “singled out” religion in its religion clause jurisprudence in order to “disfavor” religion. Furthermore, that claim is premised on the implausible notion that, whether as a textual, historical, or jurisprudential matter, the Constitution itself singles out religion for disfavor. Third, the authors' federalism-based argument undervalues the effect of incorporation of the religion clauses against the states. It is more plausible to conclude that incorporation limits rather than expands states' power to achieve greater non-establishment.

[*574] At its broadest, Lupu and Tuttle's criticism of the approach this Article suggests is that “American constitutional law, federal and state, has for many years done exactly what” the non-persecution principle “condemns.”n343 But what, exactly, does non-persecution condemn? As I have been at pains to demonstrate, it condemns (among other things) the targeted exclusion of persons and organizations from public benefits (1) for which they are otherwise eligible, (2) because of their religious affiliation or purpose. Is it fair to say that “American constitutional law” has done exactly this for many years, or indeed ever?

It is virtually impossible to reduce to specifics what the Supreme Court has done over the last century as it has worked out the constitutionally permissible relationships between religion and government. Its universally criticized jurisprudence has charted an evolutionary development of doctrines seeking to balance different theories about what the religion clauses require - and not something reducible to one purpose such as disfavoring religion by excluding it from generally available public benefits.n344 In other words, what American constitutional law has been doing since at least Reynolds n345 in 1878 is, broadly speaking, trying to figure out why the Constitution singled out religion as it did, and how the purposes behind that singling out should translate into practical relationships between the polity and religion. A long-standing generalized object of disfavoring religion is, to put it mildly, hard to reconcile with the Court's many statements (dating at least from Everson n346) that the Establishment Clause does not require government hostility toward religion n347 and that government acts permissibly and even in concert with “the best of our traditions” when it seeks to accommodate religious practices and beliefs. n348 It is impossible to reconcile with the ardently pro-religious and pro-Christian statements from earlier courts, Justices, and lawmakers. n349

[*575] Lupu and Tuttle also characterize too broadly what a plausible rule of non-persecution condemns. Non-persecution simply does not amount to a formalist (a la Philip Kurland) argument that law cannot ever use “religion” as a basis for legal categorization.n350 The non-persecution rule is narrower than that. It says law may not single out religion with the object of disfavoring or punishing it. It is clearly violated when, as State Blaines do, laws exclude religious persons and organizations from public benefits because they are religious.

The State Blaines represent a political judgment of nineteenth-century vintage, enshrined in almost forty state constitutions, about the relationship between religion and public benefits. My argument is that their collective judgment is at odds with the long-standing and consistent tradition of religious non-discrimination as seen in free exercise jurisprudence, in the neutrality concept, and in the more recent religious speech cases.n351 Is it possible that certain of the Court's non-establishment decisions (particularly in the school aid context), or indeed certain Justices' individual views, have reflected a “separationist” or “religion-hostile” cast reminiscent of the State Blaines? Roughly speaking, yes. Many commentators refer to the “strict” separationism reflected in certain decisions or periods that was possibly the result of anti-religious currents. n352 The seeds of such [*576] separationism may have been sown in absolutist language in Everson,n353 or it may have grown from more deep-seated misunderstandings about the history and purposes of the religion clauses. n354 Certain Justices have been accused, plausibly, of harboring “separationist” ideas, n355 of clinging to outdated notions of religious “divisiveness,” n356 or of simply being anti-religious. n357

[*577] But there is a difference between noticing these elements in the lengthy and complex history of the Court's religion clause jurisprudence, and raising them to the level of a normative premise of that jurisprudence. An argument that American constitutional law has targeted religion for particular disfavor asks us to make just that fundamentally implausible interpretive move. Even assuming that any anti-religious stripe of separationism ever held sway in the Court's jurisprudence, it has largely vanished - particularly concerning equal access to neutrally available public benefits, where a far more neutralist regime is firmly in place.n358 Second, as noted above, such a premise would have been flatly at odds with what the Court has consistently said about government hostility toward religion. n359 Third, it is more plausible to argue that any occasional anti-religious currents in the Court's non-establishment cases were wrong to begin with because they were out of step with a proper interpretation of how the religion clauses were supposed to interact. Certainly, when the Court has consciously altered course in its non-establishment cases, it has explicitly discarded premises that were at odds with the deeper principles of the religion clauses. n360

The major examples Lupu and Tuttle rely on to support their “singling out for disfavor” argument fail to do so. It seems strange to describe the doctrine forbidding government intervention in faith-based religious disputes as primarily disfavoring religion. Perhaps, as the authors point out, that doctrine “deprives religious factions of the opportunity for authoritative dispute resolution by the state,”n361 but it seems more plausible that the doctrine simply recognizes the delicate position religion occupies in our secular polity and seeks to protect religion from the corrosive effects of direct governmental meddling in its theological affairs - an area, moreover, in which government has no special competence. The no-subsidy or no-funding rule seems a better candidate for a doctrine that affirmatively [*578] disfavors religion - by putting a church on lesser footing than a secular recipient of some forms of government largesse - but it is a weak foundation on which to build the broad premise that American constitutional law specially disfavors religion. The parameters and the historical provenance of the no-subsidy rule continue to be disputed,n362 but assume for a moment that the Establishment Clause affirmatively requires some form of a rule that prohibits direct, unrestricted cash payments to religious groups for religious purposes. It is a long, and in my view insupportable, leap to assume from that rule alone that the Constitution sanctions a general disfavoring of religion. Even if such a rule obtains, it is more plausible to regard it as, at most, one limited disadvantaging of religion that is worked out in the Constitution itself - a specific resolution, so to speak, of the so-called “tension” between free exercise and non-establishment. And, furthermore, there are good reasons to let that stand as a unique constitutional balance that the states ought not be able to aggravate, at the risk of trampling on free exercise values, especially when the federal religion clauses apply with full force to the states themselves through incorporation. At bottom, the argument that federal non-establishment doctrine itself disfavors religion begs the more fundamental question at the heart of the State Blaines' constitutional validity - can the states legitimately go beyond whatever is legitimately demanded by federal disestablishment and heap greater disfavor upon religion as a matter of state constitutional policy? n363 As my arguments throughout this piece demonstrate, that is a notion rendered deeply implausible by constitutional text, structure, history, and jurisprudence.

A more fundamental refutation of the notion that American constitutional law has often singled out religion for disfavored treatment lies in the text and purposes of the Constitution itself. The Constitution plainly singles out religion: for instance, it forbids religious tests for federal office and “accommodates the religious desires of those who were opposed to oaths by allowing any officeholder - of any religion, or none - to take either an oath of [*579] office or an affirmation.”n364 Religious scruples here are singled out for special solicitude, not disfavor. What of the paradigmatic singling out of religion - the Free Exercise and Establishment Clauses? The former - forbidding Congress from making any law that prohibits the free exercise of religion - hardly sounds like it imposes a disadvantage on religion. Indeed, as already noted, it was originally conceived as forbidding laws punishing religion qua religion. n365 The latter, as Akhil Amar has persuasively demonstrated, was originally designed to (1) forbid Congress from creating “The Church of the United States,” and (2) prevent Congress from disestablishing existing state religious establishments. n366 The claim to find in these materials a general charter for disabling religious persons or religious organizations vis-a-vis their secular counterparts is unconvincing. If anything, their text and purposes alone would seem to leave Congress free to promote the general flourishing of religion, as it did in the territories and in its provision of legislative and military chaplains. n367 And, as we shall see, incorporation of the religion clauses against the states only lends additional weight against the general proposition that American constitutional law recognizes disfavoring religion as a valid normative premise.

So, Lupu and Tuttle's first major objection - that the non-persecution rule condemns (and would therefore dismantle) a long-standing practice of American constitutional law - turns out to be overstated. What about their federalism objection? Does the non-persecution rule unfairly handcuff the states in balancing their own church-state policy? Perhaps in 1800, but certainly not since 1940 and probably not since 1865. In other words, the federalism objection fails to take seriously the effect of incorporating the religion clauses against the states.

It is common doctrine that both religion clauses apply against the states, through the Fourteenth Amendment, with the same force as they apply against the federal government.n368 As to free exercise, the effects of this are relatively easy to understand. Free exercise is a paradigmatic individual and associational right against government overreaching, and so its application against the states should simply disable states from legislating to prohibit free exercise, just as the clause had, before, limited only the federal Congress. n369 Thus, when [*580] the Supreme Court holds that a law trenches on someone's free exercise rights, incorporation makes that the end of the story. State legislatures cannot pass such laws any longer, and thus the Supreme Court's decision (whether by a majority of five, six, seven, eight or nine Justices) in a real sense “denies the states any room whatsoever for their own church-state policy” on that issue.n370 The converse is slightly different. If the Supreme Court holds that a law does not violate free exercise, then states have some latitude to accord their citizens greater rights under state law (provided these greater rights do not independently violate the Establishment Clause). Thus, even as Smith interpreted federal free exercise not to command religious exemptions from general laws, the Court recognized (and arguably invited) states to legislate such exemptions under state law. n371 In other words, states had more latitude to develop a distinctive church-state policy under their own laws.

As to non-establishment, the effects of incorporation are knottier. It is not at all clear that non-establishment is properly described as an individual or associational right against government - perhaps it is more accurately a “right of the public at large.”n372 This makes it more difficult to say precisely what rights state citizens themselves gain when the Establishment Clause is incorporated against their state governments. n373 Regardless, it is safe to say as a matter of the Supreme Court's jurisprudence that incorporation means this: Whatever the federal government cannot do “respecting an establishment of religion,” the states also cannot do. n374 Thus, when the Supreme Court holds that a particular government practice establishes religion, that is the end of the story. States may no longer enact such practices and, to that extent, their prerogatives to experiment with different church-state policies - which they doubtlessly had before incorporation - vanish. n375 But what about when the Court, as it recently did in Zelman, declares that an existing practice does not constitute an establishment? Surely other states are [*581] not, at that point, required to enact such a practice. But the crucial question is whether the Court's non-establishment decision sets some kind of maximum ceiling for a policy of church-state separation in the states. Or, put another way, can the citizens of a state plausibly claim more non-establishment rights under state law than the Court has identified under the federal Constitution? And, if so, can they coherently claim such rights if their claims are not somehow connected to the free exercise rights (or other personal rights) that incorporation plainly gives them?

Akhil Amar has provided a complex but persuasive analysis of this question with his model of “refined incorporation” of the Bill of Rights. According to Amar, incorporation of the Establishment Clause is an awkward matter because (1) the original clause was primarily a states'-rights provision forbidding Congress from disestablishing state establishments, and (2) consequently, it is difficult to identify what additional personal rights were guaranteed to state citizens through non-establishment incorporation.n376 Amar argues that the object of the Fourteenth Amendment - designed to protect fundamental rights of United States citizens against state encroachment - suggests that collective or structural rights like non-establishment must be subtly “refined” to apply coherently against state governments. n377 On this understanding of incorporation, state citizens could claim rights of non-establishment against state laws that coerced their “bodily liberty and property,” such as “to the extent a state created a coercive establishment, decreeing that individuals profess a state creed or attend a state service or pay money directly to a state church.” n378 Amar notices, of course, that “all these examples also seem like textbook violations of religious “free exercise,'“ thus linking the rights citizens may claim under the incorporated Establishment Clause with their less-awkwardly-incorporated free exercise rights. n379

Amar's refined-incorporation proposal would, of course, significantly alter the Supreme Court's non-establishment jurisprudence by allowing the states more latitude in legislating about [*582] religion.n380 But notice its implications for our present question - may state citizens claim greater non-establishment rights than the federal Constitution supposedly gives them? Refined incorporation suggests they could not. First, because personal non-establishment rights are an elusive notion - especially when untethered from other, clearly personal rights like free exercise, free speech, or equal protection - it would not make sense under Amar's formulation to say that incorporation has guaranteed any such phantasmal rights to state citizens against their own governments, much less greater ones. Non-establishment is best conceived as a structural and collective value, and so it is hard to explain how state citizens could coherently ask for “more of it” individually as a result of incorporation. Second, Amar suggests that state citizens' proper invocation of their incorporated non-establishment rights would occur only when the state coerces their consciences or property to support an official state church or creed, or when the state has violated basic norms of religious equality - all problems reached more comfortably by free exercise, free speech, and equal protection principles. Thus, there is a sense that incorporated non-establishment values simply duplicate other incorporated rights. n381 Finally, Amar's broader view of incorporation supports a “no” answer. If incorporation of rights was designed to increase state citizens' personal liberties against state governments (and it is hard to imagine it was not), it makes little sense to argue that, post-incorporation, state legislatures have more power to define their own visions of church-state separation vis-a-vis federal standards. In other words, incorporation of the federal Establishment Clause against states should tend to nationalize, rather than localize, a uniform policy of church-state separation. To say that incorporation tended to empower states to develop their own church-state policies runs counter to any plausible understanding of incorporation, refined or not. n382

Whether or not Amar is right, thinking broadly about incorporation suggests answers to my question. For instance, we know that state citizens have equally as many free exercise rights against state governments as against the federal government. And we know that states are bound, at the very least, by a minimum standard of non-establishment - that is, what the federal government cannot do, the [*583] states cannot do. This tells us something about the limits on states when they experiment with greater church-state separation (as Lupu and Tuttle insist they can). When states do this, they are not acting on any affirmative grant of power or prerogative from the federal Constitution - they are obviously acting in their own state interests. But they are always acting under an affirmative obligation not to violate any citizen's federal free exercise rights, which plainly apply against state governments in full force. This suggests that, whether or not state citizens can coherently ask state governments for more non-establishment, what the state does in response is always limited by its citizens' federal free exercise rights.n383 This also suggests that “more non-establishment” or “greater church-state separation” cannot be independent justifications for state policies. Those policies must always be measured against the superior limitations of federal free exercise (not to mention free speech and equal protection). n384

Lupu and Tuttle's concerns with federalism and localized church-state policies thus turn out to be question begging. Whatever distinctive church-state policies a state wants to pursue will always be limited by the demands of free exercise. Incorporation of the federal Establishment Clause against the states cannot logically be interpreted as a charter for greater state power in defining its own separationist vision. Given the logic of incorporation, the only legitimate direction a state can go in - at least in the area of individual rights - is in according its citizens greater free exercise rights than those guaranteed federally. By this logic, of course, states could plausibly pursue greater church-state separation in ways that do not encroach on free exercise. They could, for instance, decide not to employ legislative chaplains or not to use any religious language or symbolism in state speech or on state property. But an argument that a principle forbidding religious discrimination or religious persecution unfairly limits states' freedom to formulate their own church-state policies is an argument against incorporation itself. By its nature, incorporation of the religion clauses limits states and it is beyond dispute that individual free exercise rights are one such limitation. Thus, assessing the validity of State Blaine Amendments throws us, not back on incorporation and federalism, but rather onto the key question - which I have explored in this Article - of whether they violate free exercise rights.

[*584]

B. Selective Funding

State Blaine Amendments are in large measure concerned with the destination and use of government funds. So, is my non-persecution argument against State Blaines open to the basic objection that the government can, indeed must, control how it spends its own limited resources?n385 The black-letter principles supporting this rejoinder, all true in the abstract, roll off the tongue. Government is under no obligation to fund the exercise of my constitutional rights - i.e., I have a constitutional right to freely exercise my religion, but that alone does not entitle me to a government-funded Bible. n386 Government may further its own policy choices through the government speech it funds and the government programs it sponsors - effectively refusing to endorse other legitimate policy choices and programs. n387 Government may create incentives to undertake certain behaviors legitimately in the public interest through selective funding, even if, to that extent, it creates disincentives to undertake other behaviors - behaviors that may be “constitutionally protected.” n388 Are these relatively straightforward maxims the answer to the State Blaine riddle? Probing under their surface suggests these principles, better understood, actually condemn the operation of the State Blaines for largely the same reasons the non-persecution principle condemns them.

First, it should be clear that the rejoinder that government need not fund the exercise of constitutional rights adds nothing to the debate. The non-persecution argument against State Blaines is not grounded on the naked demand that, simply because religion is constitutionally protected, religious persons and organizations are entitled to government funding. Instead, the argument is that, because religion is constitutionally protected, State Blaines may not exclude persons or organizations from otherwise accessible government benefits simply because they are religious. Non-persecution, therefore, is an argument against religion-sensitive exclusion, not an argument demanding religion-based inclusion. Furthermore, couching the [*585] debate in terms of “funding religion” is misleading. Strictly speaking, non-persecution does not ask that religion qua religion be funded at all.n389 But when a government funding program neutrally furthers secular interests in, for instance, education, health care, or child care, a religious person or organization seeks inclusion in the program on the basis of being a qualified education, health care, or child care provider - and not as a “religious” provider. It merely asks not to be discriminated against because of its religious affiliation. n390

When government spends money to facilitate its own speech - instead of creating public fora for the exchange of viewpoints - logically, it should be able to make choices about the content of that speech.n391 This principle overlaps with the similar notion that, when government funds a program to convey a government message - i.e., “when it enlists private entities to convey its own message” - it may “regulate the content of what is or is not expressed” in that program. n392 But, again, do these principles have anything relevant to say about the operation of the State Blaines? First, notice that they are only relevant to the narrow question of how State Blaines might restrict a state government's own speech or a state program enlisting private entities to spread a government message. If the State Blaines would typically mean that the government itself cannot use its funds to speak in a religious voice or spread religious messages, then the State Blaines do not add anything significant to preexisting federal constitutional limitations on government speech. n393 A different [*586] situation arises, however, if a State Blaine would prevent government from including any person or organization in a government message program, simply because of their religious identity or affiliation.n394 This restriction would have nothing to do with government shaping the content of its message - with regulating “what is or is not expressed” in the context of its own program - nor with government “taking legitimate and appropriate steps to ensure that its message is neither garbled nor distorted by the grantee.” n395 Precisely because it is not plausibly related to the content of government expression, this kind of categorical exclusion savors of disabling religious persons and organizations because they are religious. It is hard to see how such a policy would find constitutional shelter under the government speech doctrine.

Finally, outside the sphere of its own messages, government may use selective funding to create incentives to undertake certain private behavior, at least indirectly creating a disincentive to undertake other behavior.n396 A contentious example is abortion: Government may constitutionally structure Medicaid payments so that they are available to pay for “childbirth” but not available to pay for nontherapeutic abortions, thus creating an arguably strong incentive in favor of childbirth, and against abortion, for Medicaid recipients. n397 Is this the answer to the State Blaine issue? Just as government may [*587] financially incentivize childbirth and thereby disincentivize the constitutionally-protected right to choose an abortion, may government also use selective funding to create financial incentives in favor of secular or non-religious behaviors and the concomitant disincentives to religious behaviors and affiliations? This reasoning has some superficial appeal,n398 but to accept it requires ignoring two basic propositions. Generally, government may not use its selective funding power to unconstitutionally penalize the exercise of constitutional rights. n399 Specifically, there is a profound difference between the constitutionally-protected right to choose an abortion and the constitutionally-protected right to free exercise of religion.

A distinction of constitutional magnitude lies between the government's mere refusal to fund the exercise of constitutional rights and its penalizing the exercise of those rights by placing conditions on access to government funds.n400 This is not the place to plumb the depths of the unconstitutional conditions doctrine, n401 but its basic tenets reveal that the State Blaines go beyond refusing to fund religion and instead penalize religious identity, affiliation, and purposes. As Michael Paulsen explains, the unconstitutional conditions doctrine holds that “government may not condition one legal right, benefit, or privilege on the abandonment of another legal right, benefit, or privilege,” provided that (1) the government could not directly command the abandonment of the right, benefit, or privilege, and (2) the condition is not “directly germane to (in the sense of being practically inseparable from) the nature of the right or benefit itself.” n402 Crucial to applying the doctrine is “defining the [*588] exact nature of the “right' which is being conditioned” in order to “provide a determinate, baseline point-of-reference against which the constitutionality of the condition may be judged.”n403 How do the State Blaines fare under these principles? Take Witters and Davey as examples.

On the strength of its Blaine Amendment alone, Washington State essentially said to Larry Witters and Joshua Davey, “You may have access to state educational aid, on the condition that you not use the money for ministry training (Witters) or for a theology degree (Davey).”n404 Apart from their religious plans, Witters and Davey were, of course, eligible for the funds. Was Washington simply refusing to fund their religious choices, or was Washington wrongly penalizing the exercise of their constitutional right to free exercise? First, we must define the exact nature of the rights being conditioned. It is not difficult to imagine, just as the Supreme Court did in McDaniel, that Witters' and Davey's free exercise rights encompassed their pursuit of religious vocations. n405 Washington asked Witters and Davey to abandon those rights in order to participate in state educational funding. Washington, of course, could not have commanded this abandonment directly. Nor, importantly, was the condition imposed on access to the funds directly germane to the nature of the funds themselves. That is, the fact that instruction was religious was not fundamentally at odds with the neutral provision of educational funds for the handicapped (Witters) or for high-achieving students in certain income brackets (Davey). n406 It is thus difficult to escape the conclusion that Washington did more than refuse to fund the exercise of Witters' and Davey's constitutional rights; instead, Washington penalized the exercise of those rights by exacting the loss of all state educational assistance. n407

[*589] But is this analysis inconsistent with the Supreme Court's decisions that allow government to fund childbirth but not abortion? Briefly, no.n408 The abortion right and the free exercise rights at issue here are not congruent. Government is not required to act in an evenhanded way as between abortion and childbirth; it must refrain from imposing an undue burden on a woman's choice to have an abortion. n409 Government, however, has a legitimate interest in the protection of fetal life throughout pregnancy. n410 Thus, short of unduly burdening abortion rights, government is free to promote childbirth. n411 In other words, encouraging childbirth is a legitimate government purpose that is legally and logically separable from objective hostility to the abortion right. n412 Government can therefore encourage childbirth in [*590] its own speech and can structure programs like Medicaid to fund family planning services that include childbirth but exclude abortion.

By contrast, government must adopt a distinctly more agnostic stance toward religion. The notion that government funds could be spent in order to incentivize “the secular” over “the religious” simply flies in the face of a century-and-a-half of religion clause jurisprudence. Non-establishment doctrine has long recognized that, just as government may not prefer religion over non-religion, it also may not prefer non-religion over religion.n413 Similarly, the Free Exercise Clause, as originally understood and as confirmed by Smith and Lukumi, forbids laws that adopt a hostile stance toward religion - where laws overtly or covertly target religion qua religion - and not where neutral laws incidentally burden religious exercise. n414 Finally, the religious speech cases, based on equal access to public fora for religious and non-religious viewpoints alike, are impossible to square with a government interest in furthering the secular over the religious. n415 None of this is contradicted by the proposition that laws must have secular objects - certainly they must, but they also cannot have “encouragement of non-religion and discouragement of religion” as an object. That is, when laws have a genuinely secular purpose, they are simply agnostic toward religion; but when a law has as its [*591] purpose the encouragement of non-religious purposes, it is hard to understand that purpose, legally or logically, apart from an objective hostility to religion itself.n416

Thus, the application of the Washington State Blaine to Witters and Davey appears to constitute an impermissible penalty on their exercise of religion under the unconstitutional conditions doctrine. This accords with Michael Paulsen's broad statement of the doctrine as applied to religious persons and groups seeking equal access to public fora or public benefits. Paulsen argues that “government may not condition a religious speaker or group's equal access to a public forum, public benefit, or any otherwise generally available privilege on the religious speaker or group's abandonment of rights of religious autonomy, identity, self-definition, self-governance, or religiously-motivated conduct.”n417 Notice how Paulsen's statement of the unconstitutional conditions doctrine interacts with the non-persecution principle. Government may not broadly and neutrally offer benefits - whether in the form of access to a public forum, to public funding, or to inclusion in government programs - but essentially exclude religious recipients by attaching religion-sensitive conditions to those benefits.

We can plausibly understand the State Blaines' targeted exclusion of religious persons, groups, and purposes from public benefits in this alternate way, as a generalized condition that these persons and groups abandon their religious identity, affiliation, or purpose in order to access public benefits. The unconstitutional conditions doctrine suggests that such a condition typically amounts to a penalty on the exercise of religion. Government generally cannot condition access to a legal benefit on the abandonment of religious purposes, identity, or affiliation. Of course, government could do so if it could command the abandonment directly - but it is hard to imagine that government could ever plausibly do that. More importantly, when would such a condition be so directly germane to the benefits offered that government would have no choice but to exclude religious persons or groups from access to them? One plausible answer, of course, is if the federal Establishment Clause affirmatively forbade religious inclusion in those benefits. But, as we have seen, non-establishment law today will rarely compel exclusion of religious persons or groups from neutrally-available government benefit programs.n418 Thus, the unconstitutional conditions doctrine suggests that when states, through their State Blaines, try to reach beyond the Establishment Clause in this way - excluding religious persons and groups from [*592] neutrally available benefits because they are religious - states unconstitutionally punish religious exercise.n419

Generally, this section addresses a rejoinder to my argument grounded in government's ability to control how and why it spends money. It suggests that the general proposition that government must selectively allocate its resources sheds no light on the debate. It also suggests that, when government itself is speaking or spreading its own message through private entities, State Blaines may plausibly operate to require state government to speak in a non-religious voice. But it is doubtful that State Blaines could legitimately require state governments to restrict the participation of religious persons or groups in government message programs simply because they are religious. Such a categorical restriction has little to do with government's ability to shape its own message. Finally, the range of legitimate government purposes suggests that, while government may legitimately (albeit, non-coercively) structure subsidies to encourage childbirth over abortion, government may not legitimately encourage non-religion over religion. Relatedly, the unconstitutional conditions doctrine suggests that government may not legitimately condition access to public benefits on recipients' abandonment of religious identity or affiliation. The State Blaines' overall exclusion of religious persons, groups, and purposes from participation in public benefits runs aground on these principles. More generally, however, the “funding” rejoinder to my non-persecution argument, much like the [*593] “federalism” rejoinder, begs the foundational question posed by non-persecution: In the allocation of otherwise available public benefits, may government constitutionally discriminate against religious persons, organizations, or purposes because they are religious? The answer provided by constitutional text, structure, history, and jurisprudence is a consistent and resounding no.

VI. Conclusion

This extended analysis of the State Blaine Amendments has focused on the historical context in which the State Blaines developed and also on the legal context in which they currently operate. The State Blaines arose during a period of divisive national upheaval over the issue of funding Catholic schools. They are a legal residue of that crisis, representing a set of judgments about the relationship between religion and the public square, and they persist to the present day in almost forty state constitutions. The State Blaines use a variety of linguistic formulas, but they are united by an overarching purpose - to exclude religious persons and groups from the equal enjoyment of public benefits. Given the sentiments motivating their birth, we should not be surprised that the general operation of the State Blaines, from today's vantage point, is out of harmony with the foundational currents of the Supreme Court's religion clause jurisprudence. One of those currents in particular calls the State Blaines into serious question - the Court's consistent condemnation of laws that target religious belief, worship, status, and affiliation for disfavored treatment.

In this Article, I have focused on the likely operation of State Blaines implicated when public benefits are made generally available to religious and non-religious persons and groups on a neutral basis. As broad and varied as the State Blaines are, they will likely operate legitimately in some limited areas.n420 But in this increasingly common context - seen in the rise of “voucher” programs and “charitable choice” movements - the operation of the State Blaines raises serious constitutional questions under the First Amendment. When the State Blaines exclude persons and groups from participation in broad-based social programs, they single out religion for disfavored treatment. That disfavor cannot be justified by states' own federalism interests, nor by their prerogative to selectively fund certain activities over others. The Supreme Court has never approved a law that singles out religious persons or groups for special burdens because of their religious character. When the Court finally takes the constitutional measure of the State Blaines - and it will have that chance this term - the State Blaines are likely to fall.

FOOTNOTES:

n1. See Witters v. Comm'n for the Blind, 689 P.2d 53, 54-56 (Wash. 1984) [hereinafter Witters I]. The religion clauses of the First Amendment - “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof” - textually restrain the federal Congress only, but have been applied against the states through the Fourteenth Amendment. U.S. Const. amend. I; see Everson v. Bd. of Educ., 330 U.S. 1 (1947) (applying the Establishment Clause to the states); Cantwell v. Connecticut, 310 U.S. 296 (1940) (incorporating the Free Exercise Clause); see also generally Akhil Reed Amar, The Bill of Rights: Creation and Reconstruction 163-214 (1998). In Witters I, the Washington Supreme Court applied the Supreme Court's Lemon test - at that time the doctrinal framework for evaluating Establishment Clause cases - and found that Witters' use of the state aid for ministry training would have the “primary effect of advancing religion” and was therefore unconstitutional. See Witters I, 689 P.2d at 56 (applying Lemon v. Kurtzman, 403 U.S. 602 (1971)).

n2. See Witters v. Wash. Dep't of Servs. for the Blind, 474 U.S. 481 (1986) [hereinafter Witters II].

n3. See Witters v. State Comm'n for the Blind, 771 P.2d 1119 (Wash. 1989) [hereinafter Witters III]. The court relied on the Washington Constitution which in pertinent part states: “No public money or property shall be appropriated for or applied to any religious worship, exercise or instruction, or the support of any religious establishment.” Wash. Const. art. I, 11 (1889); see infra note 116.

n4. See McMonagle v. N.E. Women's Ctr., Inc., 493 U.S. 901, 903-04 (1989) (White, J., dissenting from denials of petitions for certiorari in several cases). In dissent, Justice White argued that the Washington Supreme Court's interpretation of its state constitution “presents important federal questions regarding the free exercise rights of citizens who participate in state aid programs that permit recipients a private choice in using funds received and regarding the extent to which state involvement with religion that does not violate the Establishment Clause is required by the Free Exercise Clause.” Id. at 904.

n5. I discuss this controversy in detail infra, Part II.A.

n6. See generally Frank R. Kemerer, State Constitutions and School Vouchers, 120 Educ. L. Rep. 1 (1997) (canvassing the various State Blaine Amendments); Linda S. Wendtland, Note, Beyond the Establishment Clause: Enforcing Separation of Church and State Through State Constitutional Provisions, 71 Va. L. Rev. 625 (1985) (discussing state courts' interpretations of State Blaine Amendments).

n7. A representative State Blaine - this one from the 1885 Florida Declaration of Rights - reads thus: “No revenue of the state or any political subdivision or agency thereof shall ever be taken from the public treasury directly or indirectly in aid of any church, sect, or religious denomination or in aid of any sectarian institution.” Fla. Declaration of Rights 6 (1885); Fla. Const. art. I, 3.

n8. But see, e.g., Walter Gellhorn & R. Kent Greenawalt, The Sectarian College and the Public Purse (1970) (analyzing Fordham University's compliance with the New York Blaine Amendment).

n9. See Jay S. Bybee & David W. Newton, Of Orphans and Vouchers: Nevada's “Little Blaine Amendment” and the Future of Religious Participation in Public Programs, 2 Nev. L.J. 551, 574 (2002) (noting that “the U.S. Supreme Court has never had before it a challenge to the constitutionality of a Little Blaine Amendment”); see also Rebecca G. Rees, “If We Recant, Would We Qualify?”: Exclusion of Religious Providers from State Social Service Voucher Programs, 56 Wash. & Lee L. Rev. 1291, 1296 (1999) (observing that “the United States Supreme Court ... has never addressed the possibility of a conflict between First Amendment principles and a State Blaine provision that excludes a religious group or individual from a general government program or benefit”). This may change soon, however. On May 19, 2003, the Supreme Court granted certiorari in Davey v. Locke, a Ninth Circuit decision that rejected Washington's Blaine Amendment as justification for a state scholarship program that excludes students seeking theology degrees. See Davey v. Locke, 299 F.3d 748 (9th Cir. 2002), cert. granted, 123 S. Ct. 2075 (2003). The Court will hear arguments in Davey on December 2, 2003. I discuss Davey infra, Part V.A.

n10. See, e.g., Ira C. Lupu & Robert Tuttle, The Distinctive Place of Religious Entities in Our Constitutional Order, 47 Vill. L. Rev. 37, 56 (2002) (“American Separationism reached its high water mark in the early 1970s, when the United States Supreme Court laid down rules that essentially precluded any direct government assistance to the educational program of religiously affiliated elementary and secondary schools.”).

n11. See Bybee & Newton, supra note 9, at 574 (observing that “the time may have arrived when state and federal courts will have to reexamine the application and constitutionality of the Little Blaine Amendments”).

n12. See, e.g., Lupu & Tuttle, supra note 10, at 57 (“Over the past fifteen years, the prophylactic character of strict Separationism has been under siege.”); Thomas C. Berg, Anti-Catholicism and Modern Church-State Relations, 33 Loy. U. Chi. L.J. 121, 122-23 (2001) (explaining that, while “church-state separation reached its height in the 1960s and 1970s decisions forbidding public school prayers and aid to private religious schools ... in the 1980s and 1990s, this strain of separationism lost ground, particularly with respect to school aid”).

n13. The Supreme Court's recent validation of a school voucher program allowing substantial participation of religious schools should accelerate this process. See Zelman v. Simmons-Harris, 536 U.S. 639 (2002). Charles Fried has noted that, whether or not the five-Justice majority in Zelman endures, “opponents of school choice are increasingly turning to state constitutions that contain a so-called “Blaine Amendment' - a provision that insists on a more stringent and clear-cut separation between church and state than the Supreme Court requires under its First Amendment jurisprudence - to support their legal strategy.” Charles Fried, Five to Four: Reflections on the School Voucher Case, 116 Harv. L. Rev. 163, 174-75 & n.55 (2002). Consequently, the issue of the State Blaines' constitutionality has generated its share of recent attention from the media. See, e.g., Tony Mauro, Voucher War Heads to States that Ban Funding of Religious Schools, Legal Times, Aug. 5, 2002, at 1; Rob Boston, The Blaine Game, Church & St., Sept. 2002, at 4; Mark Walsh, Latest Front for Fight on Choice: Washington State, Educ. Wk., Oct. 2, 2002, at 17; George F. Will, School Choice: The Ugly Opposition, Wash. Post, Nov. 12, 2002, at A25; see also Adam Liptak, Courts Weighing Rights of States To Curb Aid for Religion Majors, N.Y. Times, Aug. 10, 2003, at A1 (discussing the Davey case). The Becket Fund - an ardent opponent of the State Blaines - catalogues much of this media attention on its website. See www.becketfund.org.

n14. See, e.g., Engel v. Vitale, 370 U.S. 421 (1962) (banning school prayer); Sch. Dist. v. Schempp, 374 U.S. 203 (1963) (prohibiting the Lord's Prayer and Bible reading in a public school); Stone v. Graham, 449 U.S. 39 (1980) (barring posting the Ten Commandments in a public school); Wallace v. Jaffree, 472 U.S. 38 (1985) (striking certain moment of silence laws); Lee v. Weisman, 505 U.S. 577 (1992) (denouncing prayers at high school graduation as unconstitutional); Santa Fe Indep. Sch. Dist. v. Doe, 530 U.S. 290 (2000) (ending prayer at high school football game).

n15. See John C. Jeffries, Jr. & James E. Ryan, A Political History of the Establishment Clause, 100 Mich. L. Rev. 279, 337-39 (2001).

n16. See generally Berg, supra note 12, at 122-23, 163-72. At the same time, Berg explains that “although negative attitudes toward Catholicism certainly remain significant, they are less widely held, are less focused on Catholic schools as such, and are only part of a broader distrust of politically active social conservatives, including evangelical Protestants.” Id. at 123. See also Lupu & Tuttle, supra note 10, at 67. Lupu and Tuttle comment that a traditional “no-aid” position on government assistance to religious schools “in practice, meant but one thing - no state assistance to Catholic elementary and secondary schools. Most happily, such sentiment is, for a variety of reasons, no longer intellectually respectable in the United States.” Id.

n17. Schempp, 374 U.S. at 306 (1963) (Goldberg, J., concurring); see, e.g., Michael W. McConnell, Religious Freedom at a Crossroads, 59 U. Chi. L. Rev. 115, 120 (1992) (criticizing the Warren and Burger Courts' “tendency to press relentlessly in the direction of a more secular society” and “to view religion as an unreasoned, aggressive, exclusionary, and divisive force that must be confined to the private sphere”); Berg, supra note 12, at 151-52 (arguing that “by invalidating officially sponsored prayers in state schools in 1962 and Bible readings the next year, the Warren Court questioned the generalized civil religion that the 1950s had affirmed” and that “the Burger Court, in a series of decisions in the 1970s ... severely limited government aid to religious elementary and secondary schools and their students” (citations omitted)); see also Richard John Neuhaus, The Naked Public Square: Religion and Democracy in America 79-82 (1984) (discussing the secularizing drift of the Supreme Court's jurisprudence).

n18. I do not, of course, mean to suggest that the State Blaines' language could ever have been correctly interpreted to apply only to Catholic schools or organizations. I know of no commentator or court having advocated that interpretation, nor - given the general references in the State Blaines to “religions,” “denominations,” and “sects” - does such an interpretation seem plausible. In any event, interpreting them that way would open the State Blaines to a charge of plain denominational discrimination under the Free Exercise Clause. See, e.g., Larson v. Valente, 456 U.S. 228 (1982). That said, I do think the history that I recount in this Article strongly suggests that there was a hope or expectation behind the enactment of State Blaines that their operation would disproportionately impact Catholic organizations. See, e.g., Mitchell v. Helms, 530 U.S. 793, 828 (2000) (plurality opinion) (linking the term “sectarian” with the anti-Catholic hostility surrounding the attempted passage of the federal Blaine Amendment, and noting that “it was an open secret that “sectarian' was code for “Catholic'“ (citing Steven K. Green, The Blaine Amendment Reconsidered, 36 Am. J. Legal Hist. 38 (1992))); Gerard V. Bradley, An Unconstitutional Stereotype: Catholic Schools as “Pervasively Sectarian,” 7 Tex. Rev. L. & Pol. 1, 5 (2002) (observing that “Justice Thomas noted in Mitchell that the term was “coined' when it “could be applied almost exclusively to Catholic parochial schools'“ (citations omitted)); see also Richard A. Baer, Jr., The Supreme Court's Discriminatory Use of the Term “Sectarian,” 6 J.L. & Pol. 449, 456-60 (1990) (discussing the provenance of term “sectarian”). In any case, as I explain throughout this Article, the question of subjective legislative motive for the State Blaines is legally distinct from the question of whether their objective operation is unconstitutional. My argument for the State Blaines' unconstitutionality does not depend on the anti-Catholic animus that brooded over their births.

n19. My observation here accords with a broader point made by Ira Lupu and Robert Tuttle (commenting on Justice Breyer's dissent in Zelman) in a recent piece. See Ira C. Lupu & Robert W. Tuttle, Zelman's Future: Vouchers, Sectarian Providers, and the Next Round of Constitutional Battles, 78 Notre Dame L. Rev. 917 (2003). Dismissing Justice Breyer's anachronistic concerns about “social strife,” Lupu and Tuttle observe that “the religious wars in the United States in the early twenty-first century are not Protestant vs. Catholic, or Christian vs. Jew, or even the more plausible Islam vs. all others. They are instead the wars of the deeply religious against the forces of a relentlessly secular commercial culture.” Id. at 954-55. For Justice Breyer's dissent, see Zelman v. Simmons-Harris, 536 U.S. 639, 717-29 (Breyer, J., dissenting).

n20. Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 557 (1993) (Scalia, J., concurring) (citing McDaniel v. Paty, 435 U.S. 618 (1978)). For a recent article reaching a similar conclusion about the operation of most State Blaines, see Mark Edward DeForrest, An Overview and Evaluation of State Blaine Amendments: Scope, and First Amendment Concerns, 26 Harv. J.L. & Pub. Pol'y 551, 556 (2003) (arguing that “many, if not most, State Blaine Amendments violate the First Amendment's provisions regarding religious liberty and free speech because they unlawfully discriminate against religious believers”).

n21. Michael W. McConnell, The Origins and Historical Understanding of Free Exercise of Religion, 103 Harv. L. Rev. 1409, 1474 (1990).

n22. There are other plausible approaches to attacking the State Blaines. See, e.g., DeForrest, supra note 20, at 617-25 (arguing that State Blaines violate freedom of speech); Toby J. Heytens, School Choice and State Constitutions, 86 Va. L. Rev. 117, 140-52 (2000) (arguing that Blaines violate equal protection); Lupu & Tuttle, supra note 19, at 962 n.204, 967-71 (suggesting certain State Blaines could be vulnerable under free speech principles, because of motivating anti-Catholic animus, or through congressional legislation under Section 5 of the Fourteenth Amendment); Rees, supra note 9, at 1313-28 (stating Blaine amendments impermissibly restrict free speech). But my approach finds that the Free Exercise Clause is the most apt locus, both historically and doctrinally, of principles condemning the State Blaines.

n23. See, e.g., McConnell, supra note 21, at 1474; see also Amar, supra note 1, at 42 (arguing that “if the phrase “Congress shall make no law' really meant that Congress simply lacked enumerated power to intrude into religious freedom in the several states, the kind of intrusion prohibited must have been a congressional law that sought to abridge religious exercise as such - a congressional law targeted at the free exercise of religion”); Kurt T. Lash, The Second Adoption of the Free Exercise Clause: Religious Exemptions Under the Fourteenth Amendment, 88 Nw. U. L. Rev. 1106, 1108, 1114 (1994) (explaining that the original Free Exercise Clause “at most ... prevented the federal government from passing laws targeting religion qua religion” and that “even if the original Free Exercise Clause could be read as an expression of individual rights, it would prohibit only those laws that directly targeted religion”).

n24. See, e.g., DeForrest, supra note 20, at 609-10 (arguing that, with reference to State Blaines, “the fundamental principle of equality of citizenship found at the heart of liberal democracy” implies “a right not to be treated as a “second-class' citizen, not only in regard to politics, but in “society's common project'“ (quoting Paul Weithman, Religious Reasons and the Duties of Membership, 36 Wake Forest L. Rev. 511, 512 (2001))).

n25. I will explore below some of the cognitive problems presented by “applying” the Establishment Clause “against” the states, and how those might impact an analysis of the State Blaines. See infra Part IV. Michael McConnell argues that application of either religion clause to the states is “somewhat anachronistic” given that the First Amendment explicitly applies only to Congress, but he allows that, “because the free exercise clause at the federal level was itself modeled on free exercise provisions in the various state constitutions, ... no structural distortions arise from assuming that, for modern purposes (after “incorporation'), the free exercise clause means the same thing for states that it has always meant for the federal government.” McConnell, supra note 21, at 1485. Not so with the Establishment Clause. Its incorporation against the states, argues McConnell, “presents far more serious interpretive difficulties, since there existed no national consensus on the question of governmental aid to religion, other than to leave the question to the states.” Id. at 1485 n.384. Akhil Amar has demonstrated what many commentators have long maintained: the Establishment Clause was originally understood only as a structural limitation on the power of the federal Congress to prevent it from meddling with, or disestablishing, state establishments. Amar, supra note 1, at 32-42; accord William K. Lietzau, Rediscovering the Establishment Clause: Federalism and the Rollback of Incorporation, 39 DePaul L. Rev. 1191 (1990). Mechanistic incorporation of the Establishment Clause against the states, consequently, is incoherent. See Amar, supra note 1, at 33-34, 41, 251-54 (criticizing mechanistic incorporation, but advocating “refined” incorporation of the Establishment Clause); see also Kurt T. Lash, The Second Adoption of the Establishment Clause: The Rise of the Nonestablishment Principle, 27 Ariz. St. L.J. 1085, 1135-36, 1151-53 (1995) (describing difficulties with incorporating the original Establishment Clause, but proposing a “reconstructed” clause more amenable to incorporation).

Of the current Justices, only Justice Thomas has expressed a willingness to revisit the establishment-incorporation issue. See Zelman v. Simmons-Harris, 536 U.S. 639, 677-81 (2002) (Thomas, J., concurring). Thomas has suggested that the Establishment Clause, even if incorporated, should bind the states “on different terms than ... the Federal Government.” Id. at 678. Picking up on arguments made by the second Justice Harlan and more recently by Amar, Thomas suggests that states should be freer to pass laws “that include or touch on religious matters” provided they “do not impede free exercise rights or any other individual religious liberty interest.” Id. (citing, inter alia, Walz v. Tax Comm'n, 397 U.S. 664, 699 (1970) (Harlan, J., concurring); see Akhil Reed Amar, The Bill of Rights as a Constitution, 100 Yale L.J. 1131, 1159 (1991); see also Lupu & Tuttle, supra note 19, at 948 (observing that Justice Thomas has “urged that the Court limit its intervention into religious liberty issues arising under state law to those properly cognizable under the Free Exercise Clause”). These arguments will be relevant to my discussion of incorporation's impact on the State Blaines. See infra Part IV.

n26. In 1845, the Supreme Court first held explicitly that “the Constitution makes no provision for protecting the citizens of the respective states in their religious liberties; this is left to the state constitutions and laws: Nor is there any inhibition imposed by the Constitution of the United States in this respect on the states.” Permoli v. First Municipality, 44 U.S. (3 How.) 589, 609 (1845). For a general discussion of Permoli, see, for example, Jay S. Bybee, Taking Liberties With the First Amendment: Congress, Section 5, and the Religious Freedom Restoration Act, 48 Vand. L. Rev. 1539, 1571-73 (1995) [hereinafter Bybee, Liberties]; Jay S. Bybee, Substantive Due Process and Free Exercise of Religion: Meyer, Pierce and the Origins of Wisconsin v. Yoder, 25 Cap. U. L. Rev. 887, 912-13 (1996) [hereinafter Bybee, Origins]. As Bybee observes, “the Court had reaffirmed this position, both prior and subsequent to ratification of the Fourteenth Amendment.” Id. at 913 (citations omitted).

n27. See, e.g., Jay S. Bybee, Common Ground: Robert Jackson, Antonin Scalia, and a Power Theory of the First Amendment, 75 Tul. L. Rev. 251, 327 (2000).

Although the First Amendment applies, by its terms, to Congress alone, the Court's jot-for-jot incorporation has brought the First Amendment to the states on precisely the same terms. The First Amendment, applied to the states through the Due Process Clause of the Fourteenth Amendment, has become a subject matter disability to the states as well. Incorporation has blurred both the federalism and separation of powers aspects of the original First Amendment.

Id.

n28. See Davey v. Locke, 299 F.3d 748, 761 (9th Cir. 2002) (McKeown, J., dissenting), cert. granted, 123 S. Ct. 2075 (2003) (discussed infra Part V.A.).

n29. See, e.g., Berg, supra note 12, at 130 (“The Protestant majority was always particularly intense and united in opposing state aid to religious schools, which were historically primarily Catholic.”).

n30. Another recent retelling of the State Blaines' genesis can be found in DeForrest, supra note 20, at 556-76; see also Joseph P. Viteritti, Reading Zelman: The Triumph of Pluralism, and Its Effects on Liberty, Equality, and Choice, 76 S. Cal. L. Rev. 1105, 1121-22 (2003) (discussing Protestant-Catholic conflicts presaging passage of the State Blaines).

n31. See, e.g., Joseph P. Viteritti, Blaine's Wake: School Choice, the First Amendment, and State Constitutional Law, 21 Harv. J.L. & Pub. Pol'y 657, 663 (1998) (citing Bernard Bailyn, Education in the Forming of American Society (1960); Richard J. Gabel, Public Funds for Church and Private Schools (1937)). Viteritti notes Tocqueville's statement that, in America, “almost all education is entrusted to the clergy.” Id. at 663 (quoting Alexis de Tocqueville, 1 Democracy in America 320 n.4 (Phillips Bradley ed., Random House 1945) (1839)). Philip Hamburger clarifies that Tocqueville's observation was likely suggested by his American editor, John C. Spencer, and referred to Protestant clergy. See Philip Hamburger, Separation of Church and State 220 n.75 (2002).

n32. See Jeffries & Ryan, supra note 15, at 297 n.83 (citing 1 Anson Phelps Stokes, Church and State in the United States 242 (1950); David B. Tyack, Onward Christian Soldiers: Religion in the American Common School, in History and Education: The Educational Uses of the Past 212, 217 (Paul Nash ed., 1970)).

n33. See Jeffries & Ryan, supra note 15, at 299 (citing Robert Michaelsen, Piety in the Public School 78-79 (1970)). Jeffries and Ryan explain that the architects of the common school, Mann chief among them, kept religion in the schools and controversy out by “promoting least-common-denominator Protestantism and rejecting particularistic influences.” Id. at 298; see also Berg, supra note 12, at 144 (explaining that “the state-operated, or “common,' schools had been created to overcome the division between Protestant denominations during the first nineteenth-century wave of Catholic immigration - to educate those various Protestant children (and ultimately, it was hoped, their Catholic counterparts) in “common'“ (citing Joseph P. Viteritti, Choosing Equality: School Choice, The Constitution, and Civil Society 147-56 (1999))).

n34. See Jeffries & Ryan, supra note 15, at 298 (“Mann insisted on Bible reading, without commentary, as the foundation of moral education.”); id. at 298 n.86 (noting that “the first textbook used in the United States, the Hornbook” contained only the alphabet and the Lord's Prayer); see also Bybee, Origins, supra note 26, at 894 (“The public schools had long been the domain of Protestant Americans. Bible readings and prayers in school reflected Protestant beliefs. Both Protestants and Catholics regarded each other with the suspicion that their respective school systems were tools for propaganda and evangelization.”); Steven K. Green, The Blaine Amendment Reconsidered, 36 Am. J. Legal Hist. 38, 41 (1992) (noting the “obvious evangelical Protestant overtones to public education”); Hamburger, supra note 31, at 220 (describing the Protestant character of instruction in New York City public schools of this period); Viteritti, supra note 31, at 666-67 (noting that “Mann's schools required daily reading from the King James version of the Bible ... the recital of prayers and the singing of hymns”).

n35. See Jeffries & Ryan, supra note 15, at 298 (quoting David B. Tyack, Onward Christian Soldier: Religion in the American Common School, in History and Education: The Educational Uses of the Past 212, 217 (Paul Nash ed., 1970)). Mann, a theologically liberal Unitarian, clashed with more conservative Massachusetts denominations, such as orthodox Congregationalists, Baptists, and Methodists. He dismissed criticism of the common-schools' watered-down Protestant theology, and demands for more substantive religious content, as “sectarian.” Id. Viteritti highlights the essentially intolerant character of this kind of universalism: “The common-school curriculum promoted a religious orthodoxy of its own that was centered on the teachings of mainstream Protestantism and was intolerant of those who were non-believers.” Viteritti, supra note 31, at 666.

n36. Jeffries & Ryan, supra note 15, at 299 (observing that “from its inception ... American public education was religious but nonsectarian”).

n37. Id.; see also Viteritti, supra note 31, at 666 (observing that, while “the American common school was founded on the pretense that religion has no legitimate place in public education ... in reality it was a particular kind of religion that its proponents sought to isolate from public support”).

n38. See Heytens, supra note 22, at 135 & nn.101-10 (providing statistical overview of U.S. Catholic population from 1789 through 1921, relying primarily on U.S. Dep't of Commerce, Historical Statistics of the United States (1975), and James Hennesey, American Catholics: A History of the Roman Catholic Church Community in the United States (1981)); see also Bybee & Newton, supra note 9, at 555 (explaining the same statistics); Jeffries & Ryan, supra note 15, at 299-300 & nn.98-103 (using similar statistical sources).

n39. See, e.g., Hamburger, supra note 31, at 202; Viteritti, supra note 31, at 669.

n40. Hamburger, supra note 31, at 202; see also Berg, supra note 12, at 130 (discussing the “long history” of American anti-Catholicism).

n41. See, e.g., Bybee & Newton, supra note 9, at 555; Green, supra note 34, at 42-43; Viteritti, supra note 31, at 669. Bybee and Newton observe that “by 1876, it was generally assumed that the Catholic vote had “determined the results of elections since 1870.'“ Bybee & Newton, supra note 9, at 555 (quoting Marie Carolyn Klinkhamer, The Blaine Amendment of 1875: Private Motives for Political Action, 42 Cath. Hist. Rev. 15, 32 (1957)).

n42. See, e.g., Viteritti, supra note 31, at 667 (noting that although Massachusetts was the only state to mandate Bible reading in public schools by law, “between seventy-five and eighty percent of the schools in the country voluntarily followed the practice”). Viteritti discusses the 1854 decision in Donahue v. Richards, in which the highest court in Maine ruled that requiring students to read the King James Bible in public schools was “not an infringement of religious freedom,” thereby upholding the expulsion of a Catholic teenager for refusing to read the Bible in class. Id. at 667-68 (discussing Donohue v. Richards, 38 Me. 376 (1854)).

n43. See, e.g., Jeffries & Ryan, supra note 15, at 300 (observing that “the very fact of a direct and unmediated approach to God contradicted Catholic doctrine,” that the Douay Bible - aside from being the Church's approved translation - “also [provided] authoritative annotation and comment,” and that, according to Church teaching, “reading the unadorned text invited the error of private interpretation”).

n44. See, e.g., Hamburger, supra note 31, at 220 (observing that the New York City “Public School Society,” which received public funds, operated ostensibly nondenominational schools that “required children to read the King James Bible and to use textbooks in which Catholics were condemned as deceitful, bigoted, and intolerant”); id. at 223 (noting that the Public School Society later attempted to bolster the claim that its schools were nonsectarian “by offering to black out the most bigoted anti-Catholic references in its textbooks”); id. at 223 n.83 (discussing the report of a special school committee that, while generally defending the nonsectarian character of New York City public schools, nonetheless reported as “not wholly unfounded” charges that “the books used in the public schools contain passages that are calculated to prejudice the minds of children against the Catholic faith”).

n45. See, e.g., Joseph P. Viteritti, Choosing Equality: Religious Freedom and Educational Opportunity Under Constitutional Federalism, 15 Yale L. & Pol'y Rev. 113, 145 (1996) (“Church leaders in Philadelphia, Boston, Baltimore, and New York City resisted the blatant Protestantism that had dominated the public school curriculum in the form of prayers, hymns, and bible reading (the King James version, of course) and eventually began to set up their own schools.”); see also Bradley, supra note 18, at 9 (stating that “a separate Catholic school system was started in this country to protect Catholic children from the scandal of aggressive Protestantism in the public schools” (citations omitted)).

n46. See Viteritti, supra note 31, at 669.

n47. See Hamburger, supra note 31, at 222 (reproducing the New York City Board of Assistants' report rejecting the Catholics' petition for school funding); see generally id. at 219-29 (discussing the New York City school funding controversy). Partly fueling Protestant fears was the belief that Catholic doctrines were incompatible with American ideals of freedom and individual conscience: This belief was understandable in light of Papal statements of the period criticizing the separation of church and state and religious liberty. See, e.g., Stephen Macedo, Diversity and Distrust: Civic Education in a Multicultural Democracy 61 (2000) (observing that America's “core principles of individual freedom and democratic equality” were seen to be threatened by the Catholic Church's “authoritarian institutional structure, its long-standing association with feudal or monarchical governments, its insistence on close ties between church and state, its endorsement of censorship, and its rejection of individual rights to freedom of conscience and worship”); see also Jeffries & Ryan, supra note 15, at 302-03 (stating that “Rome hampered attempts by American Catholics to abandon the Church's legacy by issuing reactionary pronouncements ideally suited to confirm the rankest prejudice,” and discussing attacks by Pope Gregory XVI and Pius IX on secular education and freedom of conscience); Bybee & Newton, supra note 9, at 555 (noting that “the Vatican Decree of Papal Infallibility of 1870 added to the anti-Catholic sentiment during this time” (citing Anson Phelps Stokes & Leo Pfeffer, Church and State in the United States 329 (1964))); see also generally Hamburger, supra note 31, at 229-34 (discussing American Protestant reactions to Papal condemnation of separationism, especially Gregory XVI's 1832 encyclical Mirari Vos). Indeed, as Thomas Berg explains, as late as the 1950s, Protestants continued to be plausibly threatened by the Vatican's official position that “religious freedom was not a moral ideal in itself, but at most a prudential accommodation to the fact of diversity in religious beliefs,” and that the ideal was “a Catholic confessional state with support for the Church and at least some restrictions on the educational and evangelistic activities of other faiths.” Berg, supra note 12, at 133. With the Second Vatican Council of the 1960s, however, the Vatican clearly recognized religious freedom as a human right in its Declaration on Religious Freedom, which was strongly influenced by the work of John Courtney Murray. Id. at 135-36 (citing John Courtney Murray, Governmental Repression of Heresy (1948), and John Courtney Murray, The Problem of Religious Freedom (1965)); see also John T. Noonan, Jr., The Lustre of Our Country: The American Experience of Religious Freedom 333 (1998) (discussing Murray's conflicts with the Vatican over the question of religious freedom).

n48. See Viteritti, supra note 31, at 669; see also Hamburger, supra note 31, at 216-17 (“Aroused by religious prejudice, fears about political and mental liberty, and fantasies about sexual violation, American mobs violently attacked Catholics.”). Hamburger points to the Protestant practice in the 1830s of “burning down Catholic churches, their most notorious achievement being the destruction in 1834 of the Ursuline convent in Charlestown, Massachusetts.” Id. at 216. Thomas Berg notes that “anti-Catholicism has had a long history in America, from outbreaks of mob violence in the mid-1800s against Catholic immigrants in Philadelphia and New York to the nativist, anti-immigrant campaign in the 1920s to make private schools illegal.” Berg, supra note 12, at 130 (citing Lloyd Jorgensen, The State and the Nonpublic School, 1825-1925, at 69-110 (1987), and Viteritti, supra note 45, at 151).

n49. See Green, supra note 34, at 43; see also Viteritti, supra note 31, at 669 (describing the drafting, in the 1854 Massachusetts legislature controlled by the anti-Catholic “Know-Nothing” Party, of “the first state laws to prohibit aid to sectarian schools”).

n50. See Green, supra note 34, at 43; Berg, supra note 12, at 130.

n51. See Jeffries & Ryan, supra note 15, at 301 (citation omitted); see also Viteritti, supra note 45, at 146 n.176 (dating New York law from 1844); 1844 N.Y. Laws ch. 320, 12.

n52. See Hamburger, supra note 31, at 322 (reproducing the text of Grant's speech) (emphasis added); see also Green, supra note 34, at 47-48 (discussing Grant's speech); Viteritti, supra note 31, at 670 (discussing the same speech).

n53. See Green, supra note 34, at 48-49.

n54. Green, supra note 34, at 48; see also Hamburger, supra note 31, at 322 (observing that in the speech, Grant “made separation part of the Republicans' agenda”); Viteritti, supra note 31, at 670 (Grant's speech, followed by his proposal for a constitutional amendment, “would align the Republican party with the anti-Catholic wing of the public-school lobby”).

n55. Green, supra note 34, at 52; Viteritti, supra note 31, at 670.

n56. Bybee & Newton, supra note 9, at 551 (quoting Grant's proposal to Congress); see also 4 Cong. Rec. 175 (1876). A less remarked part of the proposal advocated the taxation of church property - Grant provided an exaggerated estimate of expected revenues - hinting darkly that “the contemplation of so vast a property as here alluded to, without taxation, may lead to sequestration without constitutional authority and through blood.” Hamburger, supra note 31, at 323-24; see also Green, supra note 34, at 53 n.95 (noting that only the Catholic World criticized the taxation proposal).

n57. Green, supra note 34, at 52-53.

n58. Hamburger, supra note 31, at 323 n.93.

n59. See id. at 324.

n60. See Green, supra note 34, at 49.

n61. Id. at 53-54 (“Few people were fooled by Blaine's motives. Blaine was running for President, and the school amendment was recognized as a means of garnering support.” (citation omitted)); see also Viteritti, supra note 31, at 671 (noting that “Blaine's transparent political gesture against the Catholic Church provoked considerable press commentary,” including denunciations from the Catholic World). Even the Nation, sympathetic to Blaine's cause, conceded that the “anti-Catholic excitement was, as everyone knows now, a mere flurry” and that “all that Mr. Blaine means to do or can do with his amendment is, not to pass it but to use it in the campaign to catch anti-Catholic votes.” Viteritti, supra note 31, at 672 (citation omitted); see also Green, supra note 34, at 54.

n62. Green, supra note 34, at 49-50, 54 & n.103.

n63. See id. at 54 & n.107, 67-68; Bybee & Newton, supra note 9, at 557 n.31.

n64. See Hamburger, supra note 31, at 297.

n65. Id. at 297-98; see 4 Cong. Rec. 205 (1875); see also Bybee & Newton, supra note 9, at 551-52, 557 & n.31 (summarizing and quoting the text of the amendment); Green, supra note 34, at 53 n.96 (quoting the text of the amendment).

n66. See 4 Cong. Rec. 5189-92 (1876); Bybee & Newton, supra note 9, at 557 & n.32; Green, supra note 34, at 58-59.

n67. See Bybee & Newton, supra note 9, at 557-58. There appeared to be confusion over whether the language prohibited only certain sources of public funds from being applied to sectarian education, and also whether public funds might still be used for other sectarian activities besides education. Id.

n68. See 4 Cong. Rec. 5453 (1876); Bybee & Newton, supra note 9, at 558 & n.37 (discussing the text of the Senate proposal); see also Jeffries & Ryan, supra note 15, at 302 (stating that “the [Senate's] final version laboriously attempted to close every possible loophole through which public money might flow to religious schools”).

n69. See Bybee & Newton, supra note 9, at 558 n.37.

n70. 4 Cong. Rec. 5595 (1876); see Bybee & Newton, supra note 9, at 558; see also Green, supra note 34, at 67; Viteritti, supra note 31, at 672 & n.72 (citing Alfred W. Meyer, The Blaine Amendment and the Bill of Rights, 64 Harv. L. Rev. 939, 942, 944 (1951)).

n71. See generally Hamburger, supra note 31, at 287-334.

n72. See id. at 288-90.

n73. Id.

n74. Id. at 290-93.

n75. See id. at 294-95 n.21.

n76. Id.

n77. Id. at 298.

n78. See Everson v. Bd. of Educ., 330 U.S. 1, 16 (1947) (claiming that “no tax in any amount, large or small, can be levied to support any religious activities or institutions, whatever they may be called, or whatever form they may adopt to teach or practice religion”).

n79. Hamburger, supra note 31, at 294 n.21.

n80. Id. at 302.

n81. Id. at 304-05 n.43 (explaining the Liberal view that “even government benefits distributed on purely secular grounds could not be given to religious organizations”).

n82. Id. at 305.

n83. See generally id. at 321-28.

n84. See supra notes 65-76 and accompanying text.

n85. See Hamburger, supra note 31, at 335, 338.

n86. Id. at 338.

n87. See Marvin Olasky, Breaking Through Blaine's Roadblock, World, Aug. 24, 2002, at 1 (quoting Charles Russell's 1933 biography of Blaine).

n88. See, e.g., Viteritti, supra note 45, at 146; see also Bybee & Newton, supra note 9, at 559 (“What Congress failed to adopt for the nation, most of the states enacted for themselves.”).

n89. See Bybee & Newton, supra note 9, at 559 (noting that “Congress began requiring new states, as a condition of their entering the union, to include some kind of Little Blaine Amendment in their constitution”); Hamburger, supra note 31, at 322 (observing that Grant's 1875 speech “made separation part of the Republicans' agenda”); Viteritti, supra note 31, at 672-73 (documenting the Republican agenda to force new states to enact Blaine Amendments focused primarily on new western states). The actual substance of the various state provisions will be discussed infra, Part III.

n90. Green, supra note 34, at 43; see supra note 51 and accompanying text.

n91. My primary source for the texts of State Blaine Amendments from 1848-1909 is the 1909 edition of the Thorpe treatise. See generally 1-7 The Federal and State Constitutions, Colonial Charters, and Other Organic Laws (Francis Newton Thorpe ed., 1909) [hereinafter Federal and State Constitutions]; see also Bybee & Newton, supra note 9, at 559 n.44; Green, supra note 34, at 43 n.33 (citing W. Blakey, American State Papers 237-66 (1890)). Other commentators have estimated that only eight or nine states enacted anti-funding provisions in the 1870s. See, e.g., Bybee & Newton, supra note 9, at 559 n.44; Viteritti, supra note 31, at 673 n.78 (citing Lloyd P. Jorgenson, The State and the Non-Public School, 1825-1925, at 114 (1987)). My count - which, as explained below, takes the view that a relevant provision is one that explicitly bars access to public funds on religious grounds - shows twelve states. I do not find that any anti-funding provision was added to the New Jersey Constitution in the 1870s, as other commentators have stated. See 7 Federal and State Constitutions, supra, at 4186-4204; Viteritti, supra note 31, at 673 n.78. Also, I would mention the Alabama provision of 1875, the Georgia and New Hampshire provisions of 1877, and the Louisiana provision of 1879, which seem to often escape notice. Finally, I do not include Nevada's anti-funding provision in the 1870s because it was not finally approved until the Nevada general election of 1880. Bybee & Newton, supra note 9, at 566.

n92. See Bybee & Newton, supra note 9, at 559 n.46; Viteritti, supra note 31, at 673 & n.76; see also Hamburger, supra note 31, at 335 (observing that “nativist Protestants ... because of the strength of anti-Catholic feeling, managed to secure local versions of the Blaine amendment in a vast majority of the states”); id. at 338 (“Not only did [nativist Protestants] renew their efforts to obtain state constitutional prohibitions on the distribution of benefits to sectarian-controlled schools, but they also demanded that Congress require such clauses in the constitutions of territories seeking admission to the Union.”).

n93. See McCollum v. Bd. of Educ., 333 U.S. 203, 220 n.9 (1948) (Frankfurter, J., concurring). One should be cautious in making too much of congressional “compulsion.” As the language of the Enabling Acts indicates, Congress did not specify that the newly-admitted states must adopt Blaine-type formulations in their constitutions. But see DeForrest, supra note 20, at 573 (stating that “Congress did compel the inclusion of Blaine Amendment language in some state constitutions,” and referring to the 1889 Enabling Act (citing Viteritti, supra note 31, at 673)). But the heightened national sensitivity to Catholic incursion into education, was, I think, evidenced by Congress' requirement that public school systems be “free from sectarian control.” See McCollum, 333 U.S. at 220 n.9 (Frankfurter, J., concurring). The states presumably could have complied with such a directive through a variety of constitutional formulations - most obviously, by providing that state public schools would be “free from sectarian control.” But, as detailed below, in response to the Enabling Acts, the states went further, adopting explicit religion-sensitive restrictions in their constitutions that either tracked or went beyond the federal Blaine Amendment. See infra Part III.

n94. See McCollum, 333 U.S. at 220 n.9 (citing 28 Stat. 107, 108 (Utah); 34 Stat. 267, 270 (Oklahoma); 36 Stat. 557, 559, 570 (New Mexico and Arizona); Wyo. Const., 1889, Ordinances, 5); see also Viteritti, supra note 31, at 673 (discussing requirement for inclusion of State Blaine Amendment in the New Mexico Constitution); cf. Bybee & Newton, supra note 9, at 560 (discussing earlier Nevada Enabling Act, which required Nevada to secure in its constitution ““perfect toleration of religious sentiment'“ and that ““no inhabitant of said state shall ever be molested in person or property on account of his or her mode of religious worship'“ (quoting 13 Stat. 31, 4 (1864))). Bybee and Newton note that “Congress placed similar restrictions in the enabling acts for the constitutions of Arizona, Idaho, New Mexico, North Dakota, South Dakota, Utah, Washington, and Wyoming.” Id. at 560 n.51 (citing Anson Phelps Stokes & Leo Pfeffer, Church and State in the United States 158 (1964)).

n95. See, e.g., Green, supra note 34, at 43; Viteritti, supra note 31, at 673; Viteritti, supra note 45, at 146-47; see also Bybee & Newton, supra note 9, at 559 & n.46 (noting some counting inconsistencies among commentators); Heytens, supra note 38, at 123 n.32 (stating that approximately thirty state constitutions currently contain some form of Blaine Amendment, but that commentators often report numbers varying from twenty-four to thirty-three). My own canvass confirms that, by 1890, twenty-nine states had incorporated Blaine provisions into their constitutions. As the following section will demonstrate, I find thirty-six State Blaine Amendments by 1911 and thirty-eight after Alaska and Hawaii entered the union in 1959. Because Louisiana deleted its Blaine Amendment in 1974, I find that the present number of State Blaine Amendments is thirty-seven.

n96. See, e.g., Bybee & Newton, supra note 9, at 560 (explaining that “although the states adopted various ... Blaine Amendments, it is at least clear that the states generally intended to forbid the use of public funds in sectarian schools; and in some cases, it appears that the amendments extended to other sectarian institutions as well”); see also DeForrest, supra note 20, at 555 (arguing that the State Blaines “were motivated by a desire to preserve an unofficial Protestant establishment in public education, and to ensure that minority religions - Catholicism, in particular - would be unable to officially challenge that unofficial establishment”).

n97. See, e.g., Heytens, supra note 38, at 123 & n.32 (discussing counting discrepancies); see also supra notes 91, 95.

n98. My treatment of the State Blaines does not foreclose an analysis that categorizes them along a “continuum” according to how restrictively or expansively a particular provision bars public funding of religion. See, e.g., DeForrest, supra note 20, at 576-601 (categorizing State Blaines generally as “less restrictive,” “moderate,” or “most restrictive”). My argument does suggest, however, that in whatever context a State Blaine operates (for instance, whether it bars “direct” funding only or also “indirect” funding, or whether it applies only to education or to a broader range of persons and institutions), State Blaines generally impose disabilities on the basis of religion and, to that extent, are unconstitutional. For instance, even though Mark DeForrest distinguishes among the State Blaines according to the severity of their funding restrictions, id., he concludes that “with some notable exceptions, State Blaine provisions specifically target religious institutions for disparate treatment from other private organizations and individuals,” id. at 607.

n99. See supra note 90 and accompanying text.

n100. Wis. Const. art. I, 18 (added 1848).

n101. Mich. Const. art. IV, 40 (1850); see Mich. Const. art. VIII, 2 (amended 1970).

n102. Ind. Const. art I, 6 (added 1851) (providing that “no money shall be drawn from the treasury, for the benefit of any religious or theological institution”).

n103. Ohio Const. art. VI, 2 (added 1851).

n104. Mass. Const. art. XVIII (1855).

n105. Kan. Const. art. VI, 8 (1859) (providing that “no religious sect or sects shall ever control any part of the common-school or University funds of the State”). This language was amended and moved to art. VI, 6 in 1966. See Kan. Const. art. VI, 6 (amended 1966).

n106. Or. Const. art. I, 5 (providing that “no money shall be drawn from the Treasury for the benefit of any religeous [sic], or theological institution” and forbidding that “any money be appropriated for the payment of any religeous [sic] services in either house of the Legislative Assembly”).

n107. See Ala. Const. art. XIII, 8 (1875) (forbidding educational funds being “appropriated to, or used for, the support of any sectarian or denominational school”); id. art. XIV, 263 (amended 1901); Ill. Const. art. VIII, 3 (1870) (forbidding, inter alia, appropriation of public funds for “anything in aid of any church or sectarian purpose”) (renumbered art. X, 3 (1970)); Mo. Const. art. XI, 11 (1875) (forbidding any payment of public funds “in aid of any religious creed, church or sectarian purpose” and to any school “controlled by any religious creed, church or sectarian denomination whatever”) (renumbered art. IX, 8); Neb. Const. art. VIII, 11 (1875) (forbidding “sectarian instruction ... in any school or institution supported in whole or in part by [public school funds]” and state acceptance of any grant of property “to be used for sectarian purposes”); id. art. VII, 11 (amended 1976); Pa. Const. art. III, 18 (1874) (forbidding appropriations “for charitable, educational or benevolent purposes ... to any denominational or sectarian institution, corporation or association”); id. art. III, 29 (1967); S.C. Const. art. X, 5 (1868) (providing that “no religious sect or sects shall have exclusive right to or control of any part of the school-funds of the State”), renumbered and amended by S.C. Const. art. XI, 4 (1973). The Pennsylvania and Nebraska Constitutions were further amended in 1963 and 1976, respectively, to impose more specific restrictions against the use of public funds for religious purposes. See Neb. Const. art. VII, 11 (amended 1976); Pa. Const. art. III, 29 (added 1963); see also infra notes 127, 129 and accompanying text.

n108. Ill. Const. art. VIII, 3 (1870) (renumbered art. X, 3 (1970)).

n109. See Cal. Const. art. IV, 30 (1879) (providing that no governmental body “shall ever ... grant anything to or in aid of any religious sect, church, creed, or sectarian purpose”); Cal. Const. art. XVI, 5, art. IX, 8 (amended 1966); Colo. Const. art. IX, 7 (adopting an anti-funding provision identical to article VIII, 3 of the 1870 Illinois Constitution, article 8, section 33 (1874)); id. art. V, 34 (1876) (prohibiting “charitable, industrial, educational or benevolent” appropriations to any “denominational or sectarian institution or association,” much like article III, section 18 of the 1874 Pennsylvania Constitution (1874)); Ga. Const. art. I, 1, P XIV (1877) (including a similar prohibition); Minn. Const. art. XIII, 2 (enacting the same provision); N.H. Const. pt. 2, art. LXXXIII (1877) (enacting the same type of provision); Tex. Const. art. I, 7 (providing that “no money shall be appropriated, or drawn from the Treasury for the benefit of any sect, or religious society, theological or religious seminary”); id. art. VII, 5(a) (barring school funds from “ever being appropriated to or used for the support of any sectarian school”); see also La. Const. art. LI (1879) (providing that “no money shall ever be taken from the public treasury, directly or indirectly in aid of any church, sect or denomination of religion, or in aid of any priest, preacher, minister or teacher thereof”); id. art. CCXXVIII (providing that no school funds “shall be appropriated to or used for the support of any sectarian schools”); cf. La. Const. art. CXL (1868) (prohibiting appropriation to “any private school or any private institution of learning whatever” but lacking any reference to “sectarian” schools). Louisiana's anti-funding provisions were deleted from its constitution in the 1974 revision. See La. Const. art. I, 8 (paralleling federal religion clauses).

n110. See Ga. Const. art. I, 1, P 14 (1877) (stating that “no money shall ever be taken from the public Treasury, directly or indirectly, in aid of any church, sect, cult, or denomination of religionists, or of any sectarian institution”); Ga. Const. art. I, 2, P 7.

n111. See Minn. Const. art. XIII, 2 (“In no case shall any public money or property be appropriated or used for the support of schools wherein the distinctive doctrines, creeds or tenets or any particular Christian or other religious sect are promulgated or taught.”) (amended and restructured in 1974).

n112. N.H. Const. pt. 2, art. LXXXIII.

n113. Id. (added 1877).

n114. For instance, in 1880 Nevada ratified the addition of article XI, section 10 to its constitution, providing that “no public funds of any kind or character whatever, State, County or Municipal, shall be used for sectarian purpose.” Nev. Const. art. XI, 10 (added 1877); see generally Bybee & Newton, supra note 9, at 565-67. In 1885, Florida provided in its Declaration of Rights that no public revenue “shall ever be taken from the public treasury directly or indirectly in aid of any church, sect or religious denomination or in aid of any sectarian institution.” Fla. Declaration of Rights 6 (1885), amended by Fla. Const. art. I, 3.

n115. See supra notes 92-94 and accompanying text; see also 25 Stat. 676, 677 (1889).

n116. See Mont. Const. art. XI, 8 (1889) (forbidding any direct or indirect appropriation from public funds “for any sectarian purpose” or “to aid” any learning institution “controlled in whole or in part by any church, sect, or denomination”) (renumbered art. X, 6); N.D. Const. art. VIII, 152 (1889) (providing that no public funds “shall be appropriated to or used for the support of any sectarian school”); S.D. Const. art. VIII, 16 (added 1889); Wash. Const. art. I, 11 (added 1889) (providing that “no public money or property shall be appropriated for or applied to any religious worship, exercise or instruction, or the support of any religious establishment”); Wyo. Const. art. VII, 8 (providing that no portion of public school funds may be used to “support or assist” any institution of learning “controlled by any church or sectarian organization or religious denomination whatsoever”); id. art. III, 36 (forbidding “charitable, industrial, educational or benevolent” appropriations to any “denominational or sectarian institution or association”).

n117. See Idaho Const. art. IX, 5 (prohibiting aid to sectarian schools via a broad anti-funding provision); Ky. Const. 189 (providing that “no portion” of any educational fund “shall be appropriated to, or used by, or in aid of, any church, sectarian or denominational school”); Miss. Const. art. VIII, 208 (added 1890) (providing that “no religious or other sect” should ever control any part of the public school funds, and that no funds should be “appropriated toward the support of any sectarian school”).

n118. Supra notes 47-48, 51 and accompanying text; see N.Y. Const. art. IX, 4 (1894) (prohibiting public funds from being used “directly or indirectly, in aid or maintenance, other than for examination or inspection, of any school or institution of learning wholly or in part under the control or direction of any religious denomination, or in which any denominational tenet or doctrine is taught”) (renumbered art. XI, 3).

n119. See Del. Const. art. X, 3 (prohibiting any part of educational funds from being “appropriated to, or used by, or in aid of any sectarian, church or denominational school”); Utah Const. art. I, 4 (providing that “no public money or property shall be appropriated for or applied to any religious worship, exercise or instruction, or for the support of any ecclesiastical establishment”).

n120. See Va. Const. art. IV, 67 (1902) (prohibiting the General Assembly from making “any appropriation” of public funds “to any church, or sectarian society, association, or institution of any kind whatever, which is entirely or partly, directly or indirectly, controlled by any church or sectarian society”). Interestingly, that same section also authorized the General Assembly to, in its discretion, “make appropriations to non-sectarian institutions for the reform of youthful criminals.” Id. Article IX, section 141 of the 1902 Virginia Constitution generally forbade appropriation of public funds to “any school or institution of learning not owned or exclusively controlled by the State or some political subdivision thereof,” but it specifically empowered counties, cities, towns and districts to “make appropriations to non-sectarian schools of manual, industrial, or technical training.” Id. art. IX, 141 (1902).

n121. Okla. Const. art. II, 5 (providing that “no public money or property shall ever be appropriated, applied, donated, or used, directly or indirectly, for the use, benefit, or support of any sect, church, denomination, or system of religion, or for the use, benefit, or support of any priest, preacher, minister, or other religious teacher or dignitary, or sectarian institution as such”).

n122. Ariz. Const. art. II, 12 (forbidding public funds from being “appropriated for or applied to any religious worship, exercise, or instruction, or to the support of any religious establishment”); id. art. IX, 10 (1910) (prohibiting taxes or appropriations “in aid of any church, or private or sectarian school”).

n123. N.M. Const. art. XII, 3 (added 1911) (barring the use of any educational funds “for the support of any sectarian, denominational or private school, college or university”).

n124. Alaska Const. art. VII, 1 (enacted 1959) (providing that “no money shall be paid from public funds for the direct benefit of any religious or other private educational institution”); Haw. Const. art. X, 1 (1959) (forbidding public funds from being “appropriated for the support or benefit of any sectarian or private educational institution”).

n125. Va. Const. art. VIII, 10 (amended 1956) (emphasis added). The former provision had been interpreted to limit the expenditure of public educational funds to public schools only, thus excluding private schools altogether.

n126. Va. Const. art. VIII, 11 (added 1971) (emphasis added).

n127. Pa. Const. art. III, 29 (added 1963).

n128. See Mich. Const. art. VIII, 2 (amended 1970) (providing that “no payment, credit, tax benefit, exemption or deductions, tuition voucher, subsidy, grant or loan of public monies or property shall be provided, directly or indirectly, to support the attendance of any student or the employment of any person at any [private, denominational or other nonpublic, pre-elementary, elementary, or secondary] school”); Kemerer, supra note 6, at 4-6 (observing that this amendment was specifically designed to bar vouchers).

n129. See Neb. Const. art. VII, 11 (amended 1976).

n130. See, e.g., DeForrest, supra note 20, at 602 (observing that “the overall effect of these Blaine-style provisions, by their express wording or through later judicial interpretations, was usually to preclude both the direct or indirect transfer of state funds to religious or sectarian schools and institutions”).

n131. See Lupu & Tuttle, supra note 19, at 954-55.

n132. See, e.g., Lenstrom v. Thone, 311 N.W.2d 884 (Neb. 1981); see also Kemerer, supra note 6, at 16 (discussing the impact of this specific language on courts' application in terms of Nebraska's State Blaine Amendment).

n133. Epeldi v. Engelking, 488 P.2d 860, 861-62 & n.1 (Idaho 1971) (discussing Idaho Code 33-1501 (Michie 1970)).

n134. See id. at 865. Everson, the seminal establishment decision, concluded that a neutral transportation reimbursement did not violate the Establishment Clause merely because it incidentally helped some children attend religious schools by paying for their bus transportation. See Everson v. Bd. of Educ., 330 U.S. 1 (1947). In Allen, the Supreme Court applied Everson to conclude that the neutral provision of free secular textbooks to public and nonpublic schools - including religious schools - also did not constitute a forbidden “establishment” of religion. See Bd. of Educ. v. Allen, 392 U.S. 236 (1968). In going beyond Everson, Epeldi was not an aberration, but was merely one example of a mode of interpretation that had prevailed in state courts for many years since Everson. Thomas Berg notes that “this stricter anti-aid position prevailed in many other forums; between 1949 and 1963, seven of the eight state supreme courts to consider bus reimbursement for Catholic students ruled it invalid under state constitutional provisions.” Berg, supra note 12, at 128. Berg cites several cases striking down bus aid, including: Bd. of Educ. v. Antone, 384 P.2d 911 (Okla. 1963); State ex rel. Reynolds v. Nusbaum, 115 N.W.2d 761 (Wis. 1962); Matthews v. Quinton, 362 P.2d 932 (Alaska 1961); McVey v. Hawkins, 258 S.W.2d 927 (Mo. 1953); Zellers v. Huff, 236 P.2d 949 (N.M. 1951); Visser v. Nooksack Valley Sch. Dist., 207 P.2d 198 (Wash. 1949); Silver Lake Consol. Sch. Dist. v. Parker, 29 N.W.2d 214 (Iowa 1947); see also Anson Phelps Stokes & Leo Pfeffer, Church and State in the United States 431 (1964). But see Snyder v. Newtown, 161 A.2d 770 (Conn. 1961) (upholding aid).

n135. Epeldi, 488 P.2d at 865.

n136. Id. Idaho's State Blaine Amendment is discussed supra note 117. It broadly prohibits appropriation of public funds, inter alia, “to help support or sustain any school ... controlled by any church, sectarian or religious denomination whatsoever.” Idaho Const. art. IX, 5.

n137. Epeldi, 488 P.2d at 865.

n138. Id. at 868.

n139. Id. at 867-68. This expansive reading of the Idaho Constitution was reiterated in 1996 by the Idaho Supreme Court, when, citing Epeldi, it remarked that “the Idaho Constitution has been held to provide greater restrictions on the state's involvement in parochial activities than the Establishment Clause of the First Amendment.” See Doolittle v. Meridian Joint Sch. Dist., 919 P.2d 334, 342 (Idaho 1996). Interestingly, in that case the court additionally held that the Idaho Constitution's anti-funding provision was preempted by the reimbursement provisions of the IDEA, a federal disability law. Id.

n140. See supra note 3 and accompanying text; Witters v. State Comm'n for the Blind, 771 P.2d 1119 (Wash. 1989) (en banc) (Witters III).

n141. Witters III, 771 P.2d at 1122 (citations omitted). The Washington Blaine Amendment, dating from 1889, is discussed in supra note 116. For a general discussion of the origins of the Washington Blaine, see DeForrest, supra note 20, at 574-76.

n142. Witters III, 771 P.2d at 1122.

n143. Id. (quoting Calvary Bible Presbyterian Church v. Bd. of Regents, 436 P.2d 189, 193 (Wash. 1967) (en banc)); see generally State v. Gunwall, 720 P.2d 808, 811-13 (Wash. 1986) (containing an extensive discussion of general analysis for determining whether the Washington Constitution provides broader civil liberties than the federal Constitution). For a detailed discussion of Washington's “uniquely developed body of Blaine Amendment jurisprudence,” see DeForrest, supra note 20, at 590-601.

n144. See Bd. of Educ. v. Antone, 384 P.2d 911, 912-13 (Okla. 1963). Oklahoma's Blaine Amendment, discussed supra note 121, provides that no public money “shall ever be appropriated ... directly or indirectly, for the use, benefit, or support of any ... sectarian institution.” Okla. Const. art. II, 5.

n145. Antone, 384 P.2d at 912-13.

n146. Id. at 913; accord Meyer v. Oklahoma City, 496 P.2d 789, 790-92 (Okla. 1972).

n147. See, e.g., Elbe v. Yankton Indep. Sch. Dist., 372 N.W.2d 113, 117 (S.D. 1985) (noting that it had “clearly rejected the child benefit doctrine” in an earlier case and deeming that doctrine irrelevant in applying the South Dakota Blaine Amendment); Cal. Teachers' Ass'n v. Riles, 632 P.2d 953, 960-64 (Cal. 1981) (criticizing and refusing to follow child benefit doctrine in applying stricter provisions of the California Blaine Amendment). For a general discussion of the child benefit doctrine, see, for example, Viteritti, supra note 30, at 1123-25.

n148. Viteritti, supra note 45, at 149, nn.194-98 (citing Chester James Antieu et al., Religion Under the State Constitutions (1965), G. Alan Tarr, Church and State in the States, 64 Wash. L. Rev. 73 (1989), and Wendtland, supra note 6).

n149. Sheldon Jackson Coll. v. State, 599 P.2d 127, 129-32 (Alaska 1979) (interpreting Alaska Const. art. VII, 1 (emphasis added)).

n150. See Chittenden Town Sch. Dist. v. Dep't of Educ., 738 A.2d 539, 562-63 (Vt. 1999); Vt. Const. Ch. I, art. III (1777).

n151. See Kotterman v. Killian, 972 P.2d 606, 623-24 (Ariz. 1999) (en banc); see also DeForrest, supra note 20, at 583 (discussing Kotterman).

n152. See, e.g., Americans United for Separation of Church & State v. State, 648 P.2d 1072 (Colo. 1982) (en banc) (interpreting art. IX, 7 and art. V, 34 of the Colorado Constitution); People ex rel. Klinger v. Howlett, 305 N.E.2d 129 (Ill. 1973) (interpreting art. X, 3 of the Illinois Constitution); Chance v. Miss. State Textbook Rating & Purch. Bd., 200 So. 706 (Miss. 1941); Simmons-Harris v. Goff, 711 N.E.2d 203 (Ohio 1999) (interpreting art. VI, 2 of the Ohio Constitution); Durham v. McLeod, 192 S.E.2d 202 (S.C. 1972) (interpreting former art. XI, 9 of the South Carolina Constitution); Jackson v. Benson, 578 N.W.2d 602 (Wis. 1998) (interpreting art. I, 18 of the Wisconsin Constitution); Soc'y of Separationists, Inc. v. Whitehead, 870 P.2d 916 (Utah. 1993) (interpreting art. I, 4 of the Utah Constitution).

n153. Chance, 200 So. at 707, 710 (interpreting Miss. Const. art. VIII, 208).

n154. See Humphrey v. Lane, 728 N.E.2d 1039, 1044-45 (Ohio 2000) (stating that the “rights of conscience” provision in art. I, 7 of the Ohio Constitution provides broader free exercise rights than the federal Constitution); Simmons-Harris, 711 N.E.2d at 212 (interpreting the Ohio Blaine Amendment in art. VI, 2 in a non-separationist manner and as generally coextensive with federal Establishment Clause).

n155. See, e.g., Viteritti, supra note 45, at 149 (observing that “federal rulings to the contrary, many state courts have, from time to time, invalidated public assistance to private or parochial school students in the form of transportation or textbooks” (footnotes omitted)).

n156. See Douglas Laycock, The Underlying Unity of Separation and Neutrality, 46 Emory L.J. 43, 48-53 (1997). The first great historical controversy, as Laycock explains, was the 1780s dispute over church financing that gave rise to Madison's Memorial and Remonstrance. Id. at 48-49; see also McConnell, supra note 17, at 183 (“One of the most important eighteenth-century abuses against which the no-establishment principle was directed was mandatory support for churches and ministers. This system was support for religion qua religion; it singled out religion as such for financial benefit.”).

n157. Laycock, supra note 156, at 50.

n158. Id. Laycock contrasts the nineteenth-century resolution of the school funding problem (i.e., the proliferation of State Blaines) with the eighteenth-century resolution of the church funding problem. He argues that Madison's solution to the latter problem was a principled one that virtually everyone today still accepts, and that itself is firmly embedded in federal religion clause jurisprudence: Government cannot directly fund religious teaching and it certainly cannot exclusively fund teachers of only one kind of religion. See id. at 49 (explaining that the General Assessment was “a tax solely for the support of clergy in the performance of their religious functions,” that only Christian teachers were subsidized, and that “the essence of the general assessment was massive discrimination in favor of religious viewpoints”). In sharp contrast, the school funding crisis “did not produce a principled resolution to a difficult problem” but “produced instead a nativist Protestant victory over Catholic immigrants” that was “only a pretense of neutrality.” Id. at 52.

n159. See Everson v. Bd. of Educ., 330 U.S. 1, 15-16 (1947) (stating that “neither a state nor the Federal Government can ... aid one religion, [or] aid all religions ... [and] no tax in any amount, large or small, can be levied to support any religious activities or institutions”). At the same time, as I discuss below, Everson contains an equally strong condemnation of discrimination against religion. See id. (“On the other hand ... [a state] cannot hamper its citizens in the free exercise of their own religion. Consequently, it cannot exclude individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Non-believers, Presbyterians, or the members of any other faith, because of their faith, or lack of it, from receiving the benefits of public welfare legislation.”). Douglas Laycock observes that “the essence of both the no-aid and the nondiscrimination theories is succinctly laid out in [these] two paragraphs.” Laycock, supra note 156, at 53.

n160. 413 U.S. 756 (1973).

n161. See Fried, supra note 13, at 196. In Nyquist, the Court invalidated a New York program that gave grants to nonpublic schools and tax credits to parents whose children attended those schools, which included religious schools. 413 U.S. 756, 798 (1973). The scope of Nyquist seems to have been sharply limited by Zelman. See infra note 360.

n162. See Zelman v. Simmons-Harris, 536 U.S. 639, 662-63 (2002) (finding a program of generally available school vouchers, neutral with respect to religion, does not violate the Establishment Clause); Mitchell v. Helms, 530 U.S. 793, 807-10 (2000) (Thomas, J., plurality opinion) (finding that a program of secular governmental aid, neutrally offered to a wide range of private groups without reference to religion, does not violate Establishment Clause).

n163. See, e.g., Zelman, 536 U.S. at 649 (distinguishing between provision of government aid “directly to religious schools” and “programs of true private choice, in which government aid reaches religious schools only as a result of the genuine and independent choices of private individuals”); Mitchell, 530 U.S. at 810 (observing that the Court has, “as a way of assuring neutrality,” considered whether government aid is channeled to religious schools only because of private choice); see also Zobrest v. Catalina Foothills Sch. Dist., 509 U.S. 1, 10 (1993) (arguing that because a government-provided sign-language interpreter was present in a religious school “only as a result of the private decision of individual parents,” the aid did not violate the Establishment Clause); Witters v. Wash. Dep't of Servs. for the Blind, 474 U.S. 481, 488-89 (1986) (determining that a blind student's private decision to use neutral, generally available scholarship funds for ministry training did not violate the Establishment Clause).

n164. See Michael W. McConnell, The Problem of Singling Out Religion, 50 DePaul L. Rev. 1, 43 (2000); see also Laycock, supra note 156, at 70-71 (explaining that sometimes even a substantively neutral view of the religion clauses “requires that religion be treated in ways that are arguably worse than the treatment available to similar secular activities,” such as prohibiting the government from “celebrating religion or leading religious exercises”).

n165. Michael W. McConnell, The Selective Funding Problem: Abortions and Religious Schools, 104 Harv. L. Rev. 989, 1027 (1991) (citing cases).

n166. Id. at 1028; see also Berg, supra note 12, at 163.

Since about 1980, we have been in a third period of modern church-state relations. The last two decades have seen the decline of strong separationism as the dominant church-state ideal - a slow, partial, but continuing decline - and the corresponding rise of the principle that religion can be an equal participant with other ideas and activities in public life, including in government benefit programs.

Id. I will say more below about “selective” funding of “non-religious” persons and entities, about whether that is a plausible way of defending some operations of State Blaines, and about the relationship of that issue to selective funding of childbirth over abortion. See infra notes 397-412.

n167. Bybee & Newton, supra note 9, at 574.

n168. Bybee and Newton discuss several federal and state initiatives that take advantage of a more flexible approach to government involvement with religious organizations. See id. at 552-53. For instance, they discuss the 1996 Charitable Choice Act, a federal law allowing states that participate in certain federally funded programs “to contract with religious organizations, or to allow religious organizations to accept certificates, vouchers, or other forms of disbursement [under these programs].” Id. at 552 (quoting Charitable Choice Act of 1996, Pub. L. No. 104-193, tit. 1, 104, 110 Stat. 2161 (1996) (codified at 42 U.S.C. 604(b) (1996)). They also point to President Bush's announced policy of “encouraging faith-based solutions in partnership with the federal government” and the extensive media coverage of that initiative. Id. at 552-53 & n.10. Finally, they mention the increasing number of states that have begun experiments with school vouchers. Id. at 552-53 & n.11; see also Lupu & Tuttle, supra note 10, at 45-47 (commenting on the increasing role of religious organizations in Charitable Choice).

n169. Compare Bd. of Educ. v. Allen, 392 U.S. 236, 238 (1968) (finding that the Establishment Clause was not violated by lending of secular textbooks to students attending religious schools), with Meek v. Pittenger, 421 U.S. 349, 362-66 (1975) (determining the Establishment Clause was violated by providing secular instructional materials to religious schools), and Wolman v. Walter, 433 U.S. 229, 248-51 (1977) (finding the same). See also Viteritti, supra note 30, at 1130 (discussing the “disarray” and “inconsistency” in the Court's non-establishment and school aid jurisprudence of the 1970s).

n170. See, e.g., Mitchell v. Helms, 530 U.S. 793, 849-52 (2000) (O'Connor, J., concurring) (explaining why distinctions made in Meek and Wolman “created an inexplicable rift within our Establishment Clause jurisprudence concerning government aid to schools”); Wallace v. Jaffree, 472 U.S. 38, 110 (1985) (Rehnquist, J., dissenting) (observing that the Court's non-establishment jurisprudence, at the time, meant that “a State may lend to parochial school children geography textbooks that contain maps of the United States, but the State may not lend maps of the United States for use in geography class”); see also Viteritti, supra note 30, at 1132-41 (charting gradual clarification of school aid aspects of the Court's non-establishment jurisprudence, from Mueller through Zelman).

n171. Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 523 (1993); see infra Part IV.A.2.

n172. Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872 (1990).

n173. Compare, e.g., Kent Greenawalt, Should the Religion Clauses of the Constitution Be Amended?, 32 Loy. L.A. L. Rev. 9 (1998), and Michael W. McConnell, Free Exercise Revisionism and the Smith Decision, 57 U. Chi. L. Rev. 1109 (1990), with Gerard V. Bradley, Beguiled: Free Exercise Exemptions and the Siren Song of Liberalism, 20 Hofstra L. Rev. 245 (1991); Jay S. Bybee, Common Ground: Robert Jackson, Antonin Scalia, and a Power Theory of the First Amendment, 75 Tul. L. Rev. 251 (2000); Frederick Mark Gedicks, An Unfirm Foundation: The Regrettable Indefensibility of Religious Exemptions, 20 U. Ark. Little Rock L. Rev. 555 (1998); Philip A. Hamburger, A Constitutional Right of Religious Exemption: An Historical Perspective, 60 Geo. Wash. L. Rev. 915 (1992).

n174. McConnell, supra note 21, at 1418 (citing cases); see also Lash, supra note 23, at 1113 (agreeing with McConnell that “even if the original Free Exercise Clause was intended to express norms of individual freedom, the scope of the Clause appears to be limited to a prohibition of laws that abridge religion qua religion”).

n175. Reynolds v. United States, 98 U.S. 145 (1878). The Court drew a basic distinction between “mere opinion,” which the Free Exercise Clause clearly protected, and “actions ... in violation of social duties or subversive of good order,” which Congress could proscribe. See id. at 164.

n176. Davis v. Beason, 133 U.S. 333, 342 (1890).

n177. Id. at 343. Provocatively, the Court glossed this statement by including examples both of sects with tenets requiring “the necessity of human sacrifices, on special occasions,” as well as of “sects which denied as a part of their religious tenets that there should be any marriage tie, and advocated promiscuous intercourse of the sexes as prompted by the passions of its members.” Id. at 343.

Another “pre-incorporation” instance of the non-exemption principle was Hamilton v. Regents of the University of California, which concluded that the University did not violate Methodist conscientious objectors' “liberty” under the Fourteenth Amendment, when it refused to exempt them from mandatory military science instruction. See 293 U.S. 245, 263-65 (1934). Concurring, Justice Cardozo assumed that the Free Exercise Clause applied to the states through the Fourteenth Amendment. Relying on Davis, 133 U.S. at 333, Cardozo concluded that the objectors' religious scruples did not entitle them to an automatic exemption from the required military instruction. See id. at 265-66 (Cardozo, J., concurring). Cardozo broadly observed that “the right of private judgment has never yet been so exalted above the powers and the compulsion of the agencies of government,” and concluded in vintage oracular style that “one who is martyr to a principle ... does not prove by his martyrdom that he has kept within the law.” Id. at 268 (Cardozo, J., concurring).

n178. See Cantwell v. Connecticut, 310 U.S. 296 (1940).

n179. Minersville Sch. Dist. v. Gobitis, 310 U.S. 586, 594 (1940) (emphasis added). The Court also explained that “conscientious scruples have not, in the course of the long struggle for religious toleration, relieved the individual from obedience to a general law not aimed at the promotion or restriction of religious beliefs.” Id. (emphasis added). Only three years later, Gobitis was overruled by West Virginia State Board of Education v. Barnette, but in a way that left intact Gobitis' reasoning about the tempered scope of the non-exemption rule. See 319 U.S. 624, 639-42 (1943). The Barnette majority opinion relied on the principle that laws may not compel speech under the First Amendment. Id.; but cf. id. at 643-44 (Black, J., concurring) (relying, by contrast, on a free exercise rationale); id. at 645 (Murphy, J., concurring) (relying on the same rationale). Much later in Smith, the Supreme Court explicitly relied on Gobitis for its discussion of the non-exemption rule. Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872, 879 (1990) (quoting Gobitis, 310 U.S. at 594-95); see infra Part IV.A.1. Jay Bybee's explanation of the dynamic between Gobitis and Barnette accords with my reading of Gobitis. Justice Jackson, the author of Barnette, “broadened the [Gobitis] inquiry to take the focus off of the religious aspects of the conflict between the Witnesses and the Board of Education. The issue was compelled speech, not infringement of religious beliefs.” Bybee, supra note 27, at 279. Indeed, as Bybee explains it, Justice Jackson's general approach to the First Amendment accords with the later non-exemption/non-persecution rationale illuminated by Smith and Lukumi: “In large measure, the First Amendment applied principally when governments attempted to regulate religion qua religion or speech qua speech, but not religion or speech qua something else.” Id. at 290 (citations omitted).

n180. Braunfeld v. Brown, 366 U.S. 599, 607 (1961).

n181. Id.

n182. Everson v. Bd. of Educ., 330 U.S. 1, 16 (1947). Everson, of course, was an Establishment Clause decision and thus did not actually resolve any dispute about the scope of the Free Exercise Clause. The Court had made an equally striking statement condemning religious discrimination - one, moreover, in the context of school funding - almost forty years before in Quick Bear v. Leupp, 210 U.S. 50 (1908). That case principally involved the construction of a treaty with the Sioux tribe regarding whether the treaty terms permitted contracts with and payments to religious schools for tribe members' education. Id. But, in dicta, the Court rejected the notion that the Constitution would forbid such payments. Id. at 81-82. The Court adopted the statement of the Court of Appeals that:

It seems inconceivable that Congress should have intended to prohibit [the Sioux] from receiving religious education at their own cost if they so desire it; such an intent would be one “to prohibit the free exercise of religion' amongst the Indians, and such would be the effect of the construction for which the complainants contend.

Id. at 82 (quoting the Court of Appeals). It should be noted, however, that the Court specifically characterized the treaty funds as the Sioux's “own money” and “the only moneys that [they] can lay claim to as matter of right; the only sums on which they are entitled to rely as theirs for education.” Id. at 81-82. It should also be noted that the Supreme Court has recently referred to Quick Bear as only “indirectly” addressing the free exercise issue. Mitchell v. Helms, 530 U.S. 793, 807 n.4 (2000) (plurality opinion).

n183. Cantwell v. Connecticut, 310 U.S. 296, 305-10 (1940); see also Watchtower Bible & Tract Soc'y of N.Y., Inc. v. Village of Stratton, 536 U.S. 150 (2002) (striking down solicitation licensing requirement as applied to Jehovah's Witnesses under the First Amendment).

n184. Cantwell, 310 U.S. at 302. Solicitation was allowed only after case-by-case review, under which the secretary of public welfare determined whether the promoted cause met the requirements quoted above. Id. The Court concluded that this licensing scheme amounted to “the exercise of a determination by state authority as to what is a religious cause ... laying a forbidden burden upon the exercise of liberty protected by the Constitution.” Id. at 307. Cantwell may be more about religious speech than about religious conduct. See, e.g., Bybee, supra note 27, at 266-67. I agree with Bybee that Cantwell “concerned religious liberty only because the Connecticut statute specifically regulated religious canvassing.” Id. at 267. But, again, I think that very point is what makes Cantwell relevant to the issue of religious non-persecution. Douglas Laycock, for instance, has observed that the “religious free speech cases from the Jehovah's Witness era” are an important aspect of the foundation of the Court's religious “nondiscrimination theory.” Laycock, supra note 156, at 63 & n.124 (citations omitted).

n185. See Torcaso v. Watkins, 367 U.S. 488, 489 (1961) (quoting Md. Const. art. XXXVII).

n186. Id. at 490.

n187. Id. at 495 (footnotes omitted). The Court quoted James Iredell, later a Supreme Court Justice, during the North Carolina Convention ratification debates. Discussing the prohibition of religious tests for federal officers in proposed Article VI, see U.S. Const. art. VI, 3, and responding to the fear that the people may consequently “choose representatives who have no religion at all, and that pagans and Mahometans may be admitted into offices,” Iredell asked: “But how is it possible to exclude any set of men, without taking away that principle of religious freedom which we ourselves so warmly contend for?” Torcaso, 367 U.S. at 495 n.10 (quoting 4 Jonathan Elliot, Debates in the Several States Conventions on the Adoption of the Federal Constitution 194, 200).

n188. See McDaniel v. Paty, 435 U.S. 618, 622-25 (1978). Article IX, section 1 of the Tennessee Constitution provided: “Whereas Ministers of the Gospel are by their profession, dedicated to God and the care of Souls, and ought not to be diverted from the great duties of their functions; therefore, no Minister of the Gospel, or priest of any denomination whatever, shall be eligible to a seat in either House of the Legislature.” Id. at 621 & n.1. The provision dated from article VIII, section 1 of the 1796 Tennessee Constitution. Id. In 1978, Tennessee remained the only state in the union that excluded ministers from some public offices. Id. at 625. Maryland's clergy-disqualification provision had been struck down by a federal district court in 1974. Id.

n189. See id. at 622-25. The Court noted Madison's condemnation of the practice, underscoring the equality notions inherent in his view of religious liberty:

Does not The exclusion of Ministers of the Gospel as such violate a fundamental principle of liberty by punishing a religious profession with the privation of a civil right? does it [not] violate another article of the plan itself which exempts religion from the cognizance of Civil power? does it not violate justice by at once taking away a right and prohibiting a compensation for it? does it not in fine violate impartiality by shutting the door [against] the Ministers of one Religion and leaving it open for those of every other.

Id. at 624 (quoting 5 Writings of James Madison 288 (G. Hunt ed., 1904)) (emphasis added). The Court remarked that Madison's view “accurately reflects the spirit and purpose of the Religion Clauses of the First Amendment.” Id. In a recent essay on Madison's Memorial and Remonstrance, Vincent Blasi underscores Madison's linkage of equality with religious liberty: “There can be no dispute that considerations of equal treatment lay at the core of Madison's conception of religious liberty, both his aversion to any form of religious establishment and his emphasis on the notion of “free exercise.'“ Vincent Blasi, School Vouchers and Religious Liberty: Seven Questions From Madison's Memorial and Remonstrance, 87 Cornell L. Rev. 783, 802 (2002); see also DeForrest, supra note 20, at 614-15 (discussing Madison's Memorial in relation to State Blaines).

n190. McDaniel, 435 U.S. at 618. The justifications for the minister's disqualification put forward by the Tennessee Supreme Court were not novel - they were the same reasons that proponents of such measures had long relied on. See Hamburger, supra note 31, at 79-83.

n191. McDaniel, 435 U.S. at 626. Burger relied on the “balancing” approach of Sherbert v. Verner in this part of his opinion. Id.; see infra note 414 (discussing Sherbert). Sherbert has been limited by Smith, but Smith independently emphasized McDaniel's continuing force. See Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872, 877 (1990).

n192. See McDaniel, 435 U.S. at 626-27 (emphasis omitted). Burger was referring principally to Torcaso v. Watkins, see supra note 185, in which Maryland conditioned access to public office on the willingness to swear to the existence of God.

n193. McDaniel, 435 U.S. at 626-27. The Court relied in part on the language of the Tennessee Constitution, which “inferentially defines the ministerial profession in terms of its “duties,' which include the “care of souls,'“ and also on its construction by the Tennessee Supreme Court, which reasoned that the exclusion reaches, e.g., “those filling a “leadership role in religion.'“ Id. at 627 n.6.

n194. See id. at 627-28. Burger relied principally on Wisconsin v. Yoder, 406 U.S. 205 (1972), a case that invalidated on free exercise grounds Wisconsin's attempt to force the parents of Amish children to attend public schools after the age of 14. Like Sherbert, Yoder has also been limited by Smith. See infra notes 199-202 and accompanying text. But, again, Smith itself confirms that McDaniel still has significant impact for analyzing laws that target religiously affiliated statuses or behavior. See supra note 179.

n195. McDaniel, 435 U.S. at 628.

n196. Id. at 629. The Court's earlier quotation of Madison, as well as its observation that even in the founding era “many clergymen vigorously opposed any established church,” both suggest that the discriminatory exclusion of ministers from public office was never justified under the Free Exercise Clause. Id. at 629 n.9 (emphasis added).

n197. Id. at 632 (Brennan, J., concurring). Brennan would have also invalidated the exclusion under the Establishment Clause. Id. at 636-42.

n198. Id. at 632 (Brennan, J., concurring). That this was Brennan's perception of the law is reinforced by his citation to the language in Everson condemning laws that disabled various denominations “because of their faith, or lack of it, from receiving the benefits of public welfare legislation.” Id. at 633 n.7 (emphasis omitted) (quoting Everson v. Bd. of Educ., 330 U.S. 1, 16 (1947)); see also McDaniel, 435 U.S. at 635 n.8 (Brennan, J., concurring) (arguing that, because the clergy disqualification “by its terms ... operates against McDaniel because of his status as a “minister' or “priest,' it runs afoul of the Free Exercise Clause simply as establishing a religious classification as a basis for qualification for a political office” (citation omitted)).

In a separate concurrence, Justice Stewart agreed with Brennan that the clergy exclusion implicated the absolute prohibition against laws targeting beliefs, a principle supported by “the judgment that ... government has no business prying into people's minds or dispensing benefits according to people's religious beliefs.” Id. at 643 (Stewart, J., concurring). Justice White's concurrence stated that he would have invalidated the exclusion under the Equal Protection Clause. Id. at 643-46 (White, J., concurring).

n199. See Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872, 877 (1990); accord DeForrest, supra note 20, at 615-16 (relying on McDaniel, in part, to condemn State Blaines as generally unconstitutional). It is an error to read McDaniel narrowly to forbid only religious disqualification from “participation in the political process” or as presenting a unique conflict between state and federal rights. See, e.g., Lupu & Tuttle, supra note 19, at 965 n.218 (characterizing clergy disqualification in McDaniel as “coercively excluding clergy from one aspect of the right of self-government”); Davey v. Locke, 299 F.3d 748, 763 (9th Cir. 2002) (McKeown, J., dissenting) (arguing McDaniel merely involved the “juxtaposition [of] two fundamental rights,” one of which was the right “to directly engage in the political process”). The precedential value of the decision is better described by the Supreme Court itself - in Smith, the Court described McDaniel as forbidding government to “impose special disabilities on the basis of religious views or religious status.” See 494 U.S. at 877; see also Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 533 (1993) (reiterating Smith's interpretation of McDaniel); id. at 557 (Scalia, J., concurring) (reinforcing the same). The fact that Tennessee had imposed a religious disability on “the right to self-government” likely made the case that much easier to decide, but the controlling factor was the religious disability itself, as Smith and Lukumi make clear. See Lukumi, 508 U.S. at 520; Smith, 494 U.S. at 872. It is implausible to suggest that McDaniel would have come out differently if Tennessee had instead, for instance, generally forbidden clergy from participating in an otherwise accessible government charity program, simply because of their identity as clergy.

n200. Smith, 494 U.S. at 874. Oregon classified peyote, “a hallucinogen derived from the plant Lophophora williamsii Lemaire,” as a Schedule I controlled substance, the possession of which was punishable as a felony. See id. (citations omitted).

n201. See id. at 874 (“general criminal prohibition” on peyote use), 878 (“generally applicable law”), 879 (““valid and neutral law of general applicability'“), 880 (“a neutral, generally applicable regulatory law”), 881 (a “neutral, generally applicable law”), 884 (“a generally applicable criminal law” and “an across-the-board criminal prohibition”), 885 (“generally applicable prohibitions of socially harmful conduct”). The Court was careful to distinguish the general criminal prohibition at issue in Smith from the individualized denials of unemployment compensation the Court had invalidated in Sherbert, Thomas and Hobbie. See Smith, 494 U.S. at 876, 882-84; see also Sherbert v. Verner, 374 U.S. 398 (1963); Thomas v. Review Bd., 450 U.S. 707 (1981); Hobbie v. Unemployment Appeals Comm'n, 480 U.S. 136 (1987). Smith left these decisions intact, while limiting their applicability outside the context of “individualized” denials of religious exemptions. See 494 U.S. at 884.

n202. See Smith, 494 U.S. at 877 (observing that “the government may not ... impose special disabilities on the basis of religious views or religious status” (citing McDaniel v. Paty, 435 U.S. 618 (1978), and Fowler v. Rhode Island, 345 U.S. 67, 69 (1953))); see also id. (explaining that government would be prohibiting free exercise if it “sought to ban [religious] acts or abstentions only when they are engaged in for religious reasons, or only because of the religious belief that they display”); id. at 878 (characterizing respondents' argument that their religious motivation “places them beyond the reach of a criminal law that is not specifically directed at their religious practice”); id. at 886 n.3 (explaining that the Court “strictly scrutinizes governmental classifications based on religion” (citing McDaniel, 435 U.S. at 618, and Torcaso v. Watkins, 367 U.S. 488 (1961))).

n203. Smith, 494 U.S. at 877.

n204. Id. at 877-78.

n205. Id. at 878 (emphasis added); accord Amar, supra note 1, at 42-43 (referring to the “unreconstructed” free exercise clause). But see id. at 254-56 (discussing the “reconstructed” clause).

n206. See Smith, 494 U.S. at 889 (citing Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 723 F. Supp. 1467 (S.D. Fla. 1989)).

n207. See Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520 (1993).

n208. Id. at 533.

n209. See generally id. at 524-28. Various resolutions expressed, for example, ““concern'“ that “certain religions may propose to engage in practices which are inconsistent with public morals, peace or safety,” and aimed to oppose ““the ritual sacrifices of animals.'“ Id. at 526-27 (quoting Resolutions 87-66 and 87-90).

n210. Id. at 527-28 (quoting Ordinances 87-52, 87-71, and 87-72). For instance, the ordinances (1) prohibited animal “sacrifice,” defined as “to unnecessarily kill ... an animal in a public or private ritual or ceremony not for the primary purpose of food consumption;” (2) restricted that prohibition to any individual or group that “kills, slaughters or sacrifices animals for any type of ritual;” (3) exempted certain “licensed establishments” from the slaughtering prohibition for animals “specifically raised for food purposes” and set zoning areas for slaughterhouse use; and (4) further exempted from regulation the slaughter or processing for sale of “small numbers of hogs and/or cattle per week” in accordance with other state law. Id.

n211. Id. at 535.

n212. Id. at 531-32 (discussing Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872 (1990)).

n213. Id. at 532 (citations omitted).

n214. Id. at 533. Using largely the same expression, the Court also remarked that neutrality is violated when “the object or purpose of a law is the suppression of religion or religious conduct.” Id.

n215. Id.

n216. Bowen v. Roy, 476 U.S. 693, 703 (1986) (Burger, C.J.)

n217. Lukumi, 508 U.S. at 534.

n218. Id. (quoting Walz v. Tax. Comm'n, 397 U.S. 664, 696 (1970) (Harlan, J., concurring)); see infra notes 275-80. As to “general applicability,” the Court explained that this inquiry focused on equality-of-treatment concerns and was guided by “the principle that government, in pursuit of legitimate interests, cannot in a selective manner impose burdens only on conduct motivated by religious belief.” Id. at 542-43. The Court admitted that the “general applicability” and “neutrality” inquiries are “interrelated.” Id. at 531. Concurring, Justice Scalia “frankly acknowledged that the terms are not only “interrelated,' ... but substantially overlap.” Id. at 557 (Scalia, J., concurring).

n219. Id. at 535-37. The Court observed that, under Florida case law, even “the use of live rabbits to train greyhounds” would not violate the Florida animal cruelty laws, which the Hialeah ordinances had incorporated. Id. at 537 (citing Kiper v. State, 310 So. 2d 42 (Fla. App. 1975), cert. denied, 328 So. 2d 845 (Fla. 1975)).

n220. Id. at 536.

n221. Id. at 542. For largely the same reason, the ordinances were also not “generally applicable” - while they pursued legitimate governmental interests, at least broadly speaking, in seeking to prevent animal cruelty and to protect public health, they did so “only against conduct motivated by religious belief.” Id. at 542, 545. The Court reasoned that the ordinances were blatantly “underinclusive” in furthering the asserted legislative goals - failing to encompass many non-religious kinds of animal cruelty and public health hazards. Id. at 543-45. For no legitimate reason, the ordinances forced religiously motivated conduct alone to “bear the burden” of their prohibitions and they therefore had “every appearance of a prohibition that society is prepared to impose upon [a Santeria worshiper] but not upon itself.” Id. at 544, 545 (quoting Fla. Star v. B.J.F., 491 U.S. 524, 542 (1989) (Scalia, J., concurring in part and concurring in the judgment)).

n222. Id. at 546 (citing McDaniel v. Paty, 435 U.S. 618, 628 (1978), and Employment Division v. Smith, 494 U.S. 872, 888 (1990)).

n223. Id. at 546-47.

n224. Id. at 557 (Scalia, J., concurring).

n225. Id. Illustrating that proposition, Scalia cited McDaniel and also Chief Justice Burger's opinion in Bowen v. Roy, in which Burger stated that “denial of [governmental] benefits by a uniformly applicable statute neutral on its face is of a wholly different, less intrusive nature than affirmative compulsion or prohibition, by threat of penal sanctions, for conduct that has religious implications.” Id.; see Bowen v. Roy, 476 U.S. 693, 704 (1986) (Burger, C.J.).

n226. Lukumi, 508 U.S. at 557 (Scalia, J., concurring) (citation omitted). Jay Bybee provides an accurate synthesis of Scalia's opinions in Lukumi and Smith. As Bybee explains, the law upheld by Scalia's majority opinion in Smith “prohibited the use of peyote generally, ... [and] necessarily prohibited the religious use of peyote.” Bybee, supra note 27, at 313. The impact on religiously motivated conduct was incidental, not deliberate. The prohibition was not religion-sensitive. By contrast, in Lukumi, Scalia concurred in invalidating “a city ordinance barring the ritual slaughter of animals,” a law in which “ritual use was an element of the crime.” Id. The Lukumi law's prohibition was tied to religious motivation; its burden on the Santeria practitioners was unique and deliberate. The law was religion-sensitive. See also Michael Stokes Paulsen, A Funny Thing Happened on the Way to the Limited Public Forum: Unconstitutional Conditions on “Equal Access” for Religious Speakers and Groups, 29 U.C. Davis L. Rev. 653, 698 (1996) (observing that Smith's “less well known” holding, which was confirmed in Lukumi, “reiterated that laws “imposing special disabilities on the basis of religious views or religious status' are presumptively unconstitutional, and subject to strict scrutiny”).

n227. Lukumi, 508 U.S. at 557, 558 (Scalia, J., concurring). Any difference seems slight and immaterial. Scalia and the majority agree on the qualities of a law that render it discriminatory for purposes of free exercise analysis, but they merely group those qualities differently under the rubrics of “neutrality” and “general applicability.” Id. at 557. It appears that Scalia would treat “neutrality” more narrowly than the majority - focusing more on the actual terms of the law - but would treat “general applicability” more broadly - including the “design [and] construction” of the law. See id. at 557-58 (Scalia, J., concurring).

n228. See Bradley, supra note 18, at 15 (citations omitted).

n229. Id.

n230. See Walz v. Tax Comm'n, 397 U.S. 680, 696 (1970) (Harlan, J., concurring).

n231. McConnell, supra note 164, at 43.

n232. Reynolds v. United States, 98 U.S. 145, 164 (1878).

n233. Davis v. Beason, 133 U.S. 333, 342-43 (1890).

n234. Id. at 343.

n235. Minersville Sch. Dist. v. Gobitis, 310 U.S. 586, 594 (1940).

n236. Everson v. Bd. of Educ., 330 U.S. 1, 16 (1947); see also Braunfeld v. Brown, 366 U.S. 599, 607 (1961) (stating that, unlike a “general” law, “if the purpose or effect of a law is to impede the observance of one or all religions or is to discriminate invidiously between religions, the law is constitutionally invalid even though the burden may be characterized as being only indirect”).

n237. Braunfeld, 366 U.S. at 607.

n238. McDaniel v. Paty, 435 U.S. 618, 626-28 (1978).

n239. See Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520 (1993); Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872 (1990).

n240. See Lukumi, 508 U.S. at 532. Lukumi specifically says that “it was “historical instances of religious persecution and intolerance that gave concern to those who drafted the Free Exercise Clause.'“ Id. (quoting Bowen v. Roy, 476 U.S. 693, 703 (1986) (Burger, C.J.)).

n241. Just as it is wrong to read McDaniel narrowly, see supra notes 188-199, it is wrong to restrict Lukumi to its facts. See, e.g., Lupu & Tuttle, supra note 19, at 963 n.211 (distinguishing Lukumi because it involved “coercive, animal protection legislation upon a particular religious sect, rather than the limitation of a government benefit to secular organizations”); Davey v. Locke, 299 F.3d 748, 762 (9th Cir. 2002) (McKeown, J., dissenting) (declining to find “any guidance in Lukumi beyond the criminal ordinances at issue there”). Not only does this ignore the Court's language in Lukumi - which broadly teaches that, “at 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” - it more fundamentally ignores Lukumi's place in the larger development of the Court's religious non-discrimination jurisprudence - again, which the Court made clear in its opinion. 508 U.S. at 532-33 (emphasis added). The laws at issue in Lukumi doubtlessly presented egregious violations of free exercise, because they were designed to stamp out a central religious practice of a minority religious group. But neither the opinion itself, nor the Court's non-discrimination jurisprudence generally, gives any reason to think that Lukumi represents a minimum level of “religious persecution” which must be reached before the Free Exercise Clause is triggered.

n242. In a recent article, Mark DeForrest reaches a similar conclusion about the State Blaines. See DeForrest, supra note 20, at 607. More generally, DeForrest also argues that the State Blaines violate a “principle of nondiscrimination” inherent in liberal democracy itself and in principles of distributive justice. See generally id. at 607-13 (relying principally on Paul Weithman, Religious Reasons and the Duties of Membership, 36 Wake Forest L. Rev. 511 (2001); Ashley Woodiwiss, Ecclesial Profiling, 36 Wake Forest L. Rev. 557 (2001); John Courtney Murray, We Hold These Truths: Catholic Reflections on the American Proposition (1960)).

n243. My application of the non-persecution rule to the State Blaines does not rely on the subjective motivations legislators may have had, individually or collectively, in promulgating them. It is not clear whether such “legislative purposes” - those hopes or fears which may lurk in lawmakers' breasts but find no objective expression in the language, structure, or operation of the laws they pass - should figure in analyzing the validity of laws under the Establishment or Free Exercise Clauses. See, e.g., Mark Tushnet, Vouchers After Zelman, 2002 Sup. Ct. Rev. 1, 17-18 (2002) (suggesting that the “bad motivations” behind many State Blaines should be irrelevant to assessing their constitutionality). Some of what the Court has said in non-establishment cases suggests that legislators' subjective motivations could be relevant. See, e.g., Agostini v. Felton, 521 U.S. 203, 222-23 (1997) (stating that “we continue to ask whether the government acted with the purpose of advancing or inhibiting religion”); Wallace v. Jaffree, 472 U.S. 38, 56-61 (1985) (considering legislators' subjective motivations for “moment of silence” law in determining ““whether government's actual purpose is to endorse or disapprove of religion'“ (quoting Lynch v. Donnelly, 465 U.S. 668, 690 (1984) (O'Connor, J., concurring))); see also id. at 73-78 (O'Connor, J., concurring) (generally discussing use of legislative history, including some limited use of legislators' statements, in assessing secular purpose of the law). But see Tushnet, supra, at 17 & n.55 (relying on Sunday Closing Cases to argue that “bad motivation at the outset can become irrelevant over time, if eventually a law can be justified by identifying some permissible goals the legislature might be pursuing (today) in keeping it on the books”). As to free exercise cases, the evidence is shakier. In Lukumi, only two Justices relied on statements of individual council members' subjective motivations for the animal cruelty ordinances. See 508 U.S. at 540-42 (Kennedy, J., joined by Stevens, J.). That reliance was strongly rejected in Justice Scalia's concurrence. See id. at 558-59 (Scalia, J., concurring, joined by Rehnquist, C.J.). In any event, the parameters of the non-persecution rule I have analyzed in this article suggest that an objective notion of legislative purpose is the relevant one for free exercise purposes. Non-persecution asks how a law operates objectively with respect to religious persons, organizations, and purposes. It would not seem to regard as a necessary or a sufficient condition for a law's invalidity that the lawmakers who passed it subjectively wished to persecute religion - provided those subjective wishes found no objective expression in the language, structure, or operation of the law. A view that such subjective wishes are alone enough to invalidate a law seems inconsistent with the distinction clarified in Smith and Lukumi between “religion neutral” and “religion targeted” laws.

This issue impacts an analysis of the State Blaines. If invalidation of a particular State Blaine required a specific showing that the legislators passing it subjectively intended to persecute Catholics, the task would be difficult indeed. See Lupu & Tuttle, supra note 19, at 967-70 (describing difficulties in mounting a purely “animus-based” attack on State Blaines). Further, it would raise the hard question of whether lawmakers' subjective purposes in the late nineteenth century should even matter today. But my analysis of the State Blaines regards such subjective motivation as irrelevant. The State Blaines, on their face, objectively structure categories of public beneficiaries to exclude the religious. Understanding the State Blaines' historical provenance, of course, helps explain why such laws exist. But if we had no knowledge about why the State Blaines came into being, they would still operate unconstitutionally against religion. On this point, I disagree with Ira Lupu and Robert Tuttle that a subjective “animus-based” attack is the only way to invalidate the State Blaines. See id.

n244. See, e.g., Everson v. Bd. of Educ., 330 U.S. 1, 18 (1947) (remarking that the First Amendment “requires the state to be a neutral in its relations with groups of religious believers and non-believers”); see also Douglas Laycock, Formal, Substantive, and Disaggregated Neutrality Toward Religion, 39 DePaul L. Rev. 993, 993 (1990) (observing that “[a] wide range of courts and commentators commonly say that government must be neutral toward religion” and assuming that “neutrality is an important part of the meaning of the religion clauses” (citations omitted)).

n245. See Laycock, supra note 244, at 994.

n246. See Laycock, supra note 156, at 63.

n247. See, e.g., DeForrest, supra note 20, at 608 n.468 (linking State Blaines' discriminatory operation to the Court's use of neutrality in its religion jurisprudence).

n248. A substantive conception of neutrality does seem, however, more congruent with the religion-promoting text and purposes of the religion clauses.

n249. McConnell, supra note 164, at 43.

n250. Laycock, supra note 244, at 1001; see also Laycock, supra note 156, at 45 (reiterating argument for substantive neutrality that “an underlying purpose of religious liberty is to minimize government influence on religious choices”); Berg, supra note 12, at 122 n.5 (agreeing with Laycock's view of “substantive neutrality”); Lupu & Tuttle, supra note 10, at 66 n.96 (contrasting Laycock's “substantive neutrality” with a more formalist view of neutrality).

n251. See Mary Ann Glendon & Raul F. Yanes, Structural Free Exercise, 90 Mich. L. Rev. 477, 541 (1991).

n252. Id.

n253. McConnell, supra note 164, at 11. In an earlier article, McConnell proposed a similar view of what he called a “pluralistic” approach to interpreting the Establishment Clause. According to him, “a pluralistic approach would not ask whether the purpose or effect of the challenged action is to “advance religion,' but whether it is to foster religious uniformity or otherwise distort the process of preaching and practicing religious convictions. A governmental policy that gives free rein to individual decisions (secular and religious) does not offend the Establishment Clause, even if the effect is to increase the number of religious choices. The concern of the Establishment Clause is with governmental actions that constrain individual decisionmaking with respect to religion, by favoring one religion over others, or by favoring religion over nonreligion.” McConnell, supra note 17, at 175.

n254. See Philip Kurland, Of Church and State and the Supreme Court, 29 U. Chi. L. Rev. 1, 96 (1961); see also Mark Tushnet, Of Church and State and the Supreme Court: Kurland Revisited, 1989 Sup. Ct. Rev. 373; Laycock, supra note 244, at 999-1000 & nn.22-23; Lupu & Tuttle, supra note 10, at 66 & n.96.

n255. Kurland, supra note 254, at 96.

n256. Lupu & Tuttle, supra note 10, at 66 & n.96; cf. Glendon & Yanes, supra note 251, at 541 (arguing that the First Amendment should be read holistically as a straightforward declaration that “forbids Congress to interfere with a group of important freedoms,” first among which is simply “religious freedom”). Purely formal neutrality has been widely criticized. For instance, Laycock claims that “formal neutrality has been almost universally rejected,” that “no major commentator [has] endorsed it for a generation” (he excepts Tushnet, supra note 243), and that “hardly anyone else has been willing to apply it universally, because it produces surprising results that are inconsistent with strong intuitions.” Laycock, supra note 244, at 1000. McConnell rejects what he calls religion-blindness as an across-the-board standard for interpreting the religion clauses, and he points out that Kurland's formulation itself illogically uses “religion” as a legal categorization. See McConnell, supra note 164, at 11. I would add that it is difficult to derive a rule of formal neutrality from the text and purposes of the religion clauses themselves. If the religion clauses, as Akhil Amar has persuasively demonstrated, see Amar, supra note 1, at 33-34, 41, simply withdrew two objects of legislative power from Congress (i.e., the power to forbid the free exercise of religion and to meddle with state establishments of religion), then why should we read them as impliedly making the additional and vastly broader withdrawal of any power to legislate on religious matters altogether? Indeed, based on text and purposes alone, it would seem more plausible to reason, by negative implication, that the religion clauses empower Congress to promote the flourishing of religion generally.

n257. See, e.g., McConnell, supra note 164, at 5-6 (arguing that “the Supreme Court has repeatedly held that religious accommodations are constitutionally permissible, even if not constitutionally required” (citations omitted)). McConnell also states:

Not one historian or constitutional scholar has [in recent years] claimed that the founding generation deemed religious accommodations illegitimate. Accommodations of religion during the years leading up to the framing of the First Amendment were common (the most frequent examples were exemption from military conscription or jury duty, exemption from oath requirements, and exemption from tithes).

Id. at 14; see also, e.g., Bd. of Educ. v. Grumet, 512 U.S. 687, 705 (1994) (stating that “we do not deny that the Constitution allows the State to accommodate religious needs by alleviating special burdens” and that “our cases leave no doubt that in commanding neutrality the Religion Clauses do not require the government to be oblivious to impositions that legitimate exercises of state power may place on religious belief and practice”).

n258. See supra Part IV.A.1.

n259. See, e.g., McConnell, supra note 17, at 166-67.

The formal neutrality position would make unconstitutional all legislation that explicitly exempts religious institutions or individuals from generally applicable burdens or obligations. Yet the theory of Smith is that exemptions are a form of beneficent legislation, left to the discretion of the political branches ... . Smith thus rejects the formal neutrality position under the Establishment Clause.

Id.

n260. See, e.g., Laycock, supra note 244, at 1000 (strongly criticizing Smith); Lupu & Tuttle, supra note 10, at 71-72 & nn.113-15 (discussing criticism and defense of Smith); see also supra Part IV.A.1.

n261. See, e.g., McConnell, supra note 17, at 184-87 (arguing that selective exclusion of religious institutions from generally available public benefits would violate neutrality insofar as it “uses the government's coercive power to disadvantage religion” (citing Michael W. McConnell, Unconstitutional Conditions: Unrecognized Implications for the Establishment Clause, 26 San Diego L. Rev. 255 (1989))).

n262. See Kurland, supra note 254, at 96 (religious clauses “prohibit classification in terms of religion either to confer a benefit or to impose a burden” (emphasis added)); see also Lupu & Tuttle, supra note 10, at 66 n.96 (stating that the “Neutralist believes that religious entities and causes are to be treated exactly like their secular counterparts - no worse but no better,” and is one “who equates neutrality with nondiscrimination between religious institutions and their secular counterparts” (emphasis added)).

n263. Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819, 861 (1995) (Thomas, J., concurring) (citations omitted); see also Bowen v. Kendrick, 487 U.S. 589, 609 (1988) (observing that the Court has never held, under the Establishment Clause, “that religious institutions are disabled ... from participating in publicly sponsored social welfare programs”); Zorach v. Clauson, 343 U.S. 306, 314 (1952) (refusing to find under the Establishment Clause any “constitutional requirement which makes it necessary for government to be hostile to religion”).

n264. See, e.g., Rosenberger, 515 U.S. at 839 (“More than once have we rejected the position that the Establishment Clause even justifies, much less requires, a refusal to extend free speech rights to religious speakers who participate in broad-reaching government programs neutral in design.”); id. at 846 (O'Connor, J., concurring) (stating that “insistence on government neutrality toward religion explains why we have held that schools may not discriminate against religious groups by denying them equal access to facilities that the schools make available to all” (emphasis added)); Everson v. Bd. of Educ., 330 U.S. 1, 18 (1947) (declaring that “State power is no more to be used so as to handicap religions than it is to favor them” (emphasis added)).

n265. Bd. of Educ. v. Grumet, 512 U.S. 687, 717 (1994) (O'Connor, J., concurring) (emphasis in original).

n266. See, e.g., Zelman v. Simmons-Harris, 536 U.S. 639, 653 (2002) (finding Ohio voucher program “neutral in all respects toward religion” in that the aid is “allocated on the basis of neutral, secular criteria that neither favor nor disfavor religion” (emphasis added) (quoting Agostini v. Felton, 521 U.S. 203, 231 (1997)). The Court in Agostini found that government aid does not advance religion by creating religious incentives “where the aid is allocated on the basis of neutral, secular criteria that neither favor nor disfavor religion and is made available to both religious and secular beneficiaries on a nondiscriminatory basis.” Agostini, 521 U.S. at 231 (emphasis added).

n267. See, e.g., Lemon v. Kurtzman, 403 U.S. 602, 612 (1971) (law is an “establishment” of religion if its “primary effect ... advances [or] inhibits religion”); see also Agostini, 521 U.S. at 222-23 (confirming that “we continue to ask whether the government acted with the purpose of advancing or inhibiting religion”).

n268. See Everson v. Bd. of Educ., 330 U.S. 1, 18 (1947). Further linking neutrality to non-hostility, Everson also stated that the First Amendment “requires the state to be a neutral in its relations with groups of religious believers and non-believers; it does not require the state to be their adversary.” Id.

n269. See id. at 16 (“[The Free Exercise Clause] commands that New Jersey cannot hamper its citizens in the free exercise of their own religion. Consequently, it cannot exclude individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Non-believers, Presbyterians, or the members of any other faith, because of their faith, or lack of it, from receiving the benefits of public welfare legislation.” (emphasis in original)).

n270. As to the State Blaines, the argument would be that they themselves “establish” religion, because their purpose and effect is to “inhibit” religion by disqualifying it from generally available public benefits. See Lemon, 403 U.S. at 612 (a law's “principal or primary effect must be one that neither advances nor inhibits religion”); see also McDaniel v. Paty, 435 U.S. 618, 636-42 (1978) (Brennan, J., concurring) (arguing that Tennessee clergy exclusion also violated the Establishment Clause since the clause, “properly understood, is a shield against any attempt by government to inhibit religion as it has done here” (citations omitted)); Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819, 856-57 n.2 (1995) (Thomas, J., concurring) (suggesting that legal categorization that explicitly discriminates against religion is unconstitutional because it wrongly takes “cognizance” of religion) (citing Madison's Memorial and Remonstrance). Cutting against this line of argument, Michael McConnell has argued that the “apparent symmetry” of the Lemon “inhibition” prong is “spurious,” pointing out that “in actual practice, actions “inhibiting' religion are dealt with under the Free Exercise Clause” and that the only case in which the Supreme Court has applied “inhibition” as a matter of establishment law is Larson v. Valente, 452 U.S. 904 (1981), a case involving denominational discrimination. See McConnell, supra note 17, at 118 n.9, 152. In a similar vein, Douglas Laycock has argued that “the Court never took the “inhibiting' prong of Lemon seriously in the context of school finance.” Laycock, supra note 156, at 56.

n271. See Mitchell v. Helms, 530 U.S. 793, 835 n.19 (2002) (plurality opinion) (citing Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 532 (1990); Everson v. Bd. of Educ., 330 U.S. 1, 16 (1947); Rosenberger, 515 U.S. at 819).

n272. See Berg, supra note 12, at 151-52.

n273. Sch. Dist. of Abbington Township v. Schempp, 374 U.S. 203, 222 (1963) (emphasis added). The Schempp majority underscored the religious neutrality and non-hostility guaranteed by both religion clauses, noting that “the two clauses may overlap.” Id. As a general matter, the Court remarked that “the ideal of our people as to religious freedom ... [is] one of “absolute equality before the law, of all religious opinions and sects'“ and that ““the government is neutral, and, while protecting all, it prefers none, and it disparages none.'“ Id. at 214-15 (quoting Minor v. Bd. of Educ., 23 Ohio St. 211, 253 (1872) (Taft, J., dissenting)) (emphasis added). The Court described the religion clauses' overarching approach as “wholesome “neutrality.'“ Schempp, 374 U.S. at 222. The Court added that “we agree of course that the State may not establish a “religion of secularism' in the sense of affirmatively opposing or showing hostility to religion, thus “preferring those who believe in no religion over those who do believe.'“ Id. at 225 (quoting Zorach v. Clauson, 343 U.S. 306, 314 (1952)).

n274. Schempp, 374 U.S. at 305 (Goldberg, J., concurring). Justice Goldberg's elaboration of neutrality seems to have more of a “substantive” flavor than the majority's articulation, insofar as Goldberg emphasized that non-establishment disabled the government from engaging in or compelling religious practices, from showing “favoritism” to particular sects or to religion generally, and from “deterring” religious belief. Id. The majority, by contrast, reasoned that laws may not have the “effect” of either advancing or inhibiting religion. Id. at 222. As Douglas Laycock points out, the first two prongs of the Lemon test (in particular, the “neither advances nor inhibits” language) “are taken almost verbatim from the Court's elaboration of “benevolent neutrality' in [Schempp].” Laycock, supra note 156, at 56.

n275. See Walz v. Tax Comm'n, 397 U.S. 664, 676 (1970).

n276. Id. at 669. In the same passage, the Court also disclaimed undue rigidity in adhering to “the course of constitutional neutrality,” warning that “rigidity could well defeat the basic purpose of these provisions, which is to insure that no religion be sponsored or favored, none commanded, and none inhibited.” Id.

n277. See id. at 694-700 (Harlan, J., concurring); see also Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 534 (1993).

n278. Walz, 397 U.S. at 696. Harlan cited examples such as school-sponsored prayer or Bible reading or “released-time” programs that were structured to encourage participation in religious instruction. Id. As Harlan described it, “voluntarism” still factors significantly into the Court's approach to “neutrality,” as seen in the Court's recent discussions of when “religious indoctrination” can be ascribed to the government. See, e.g., Mitchell v. Helms, 530 U.S. 793, 809 (2000) (plurality opinion) (discussing governmental indoctrination).

n279. Walz, 397 U.S. at 696. As already discussed, in Lukumi the Court drew on Harlan's idea of “religious gerrymanders” to describe a significant impermissible aspect of the Hialeah ordinances - i.e., that they pursued otherwise legitimate governmental objectives only against religious conduct. See supra Part IV.A.1.

n280. Walz, 397 U.S. at 694. Supporting this statement, Harlan quoted the passage from Justice Goldberg's Schempp concurrence discussed earlier in this section, and also cited the Court's free exercise discussion in Torcaso, discussed supra in Part IV.A., which condemned government discrimination in favor of some or all religions. Id. at 695 (discussing Schempp, 374 U.S. at 305 (Goldberg, J., concurring), and Torcaso v. Watkins, 367 U.S. 488, 495 (1961)).

n281. Paulsen, supra note 226, at 655.

n282. In this Article, I do not address at length the argument that certain applications of State Blaines independently violate the Free Speech Clause. There are undoubtedly applications of State Blaines that would squarely abridge free speech - e.g., if State Blaines are used to justify excluding religious viewpoints from public or limited public fora. But the more difficult question, which I do not explore here, is whether the concept of a speech forum is sufficiently expansive to cover the wider array of situations where religious persons and institutions seek equal access to public benefits. See, e.g., Rees, supra note 9, at 1313-28 (arguing that excluding religious providers from neutral voucher programs would abridge free speech); see also DeForrest, supra note 20, at 618-25 (applying free speech principles to State Blaines); Lupu & Tuttle, supra note 19, at 962 n.204 (advocating a narrower viewpoint-discrimination ground for result in Davey v. Locke, discussed infra notes 326-32). Again, however, this Article focuses on free exercise principles as a primary source for attacking the vast majority of the State Blaines' conceivable applications, and so I discuss the Court's religious speech cases insofar as they support my general non-persecution argument.

n283. See Widmar v. Vincent, 454 U.S. 263 (1981). The student group in Widmar was called “Cornerstone,” an evangelical Christian organization whose meetings “included prayer, hymns, Bible commentary, and discussion of religious views and experiences.” Id. at 265 & n.2.

n284. See Lamb's Chapel v. Ctr. Moriches Union Free Sch. Dist., 508 U.S. 384, 386 (1993).

n285. See id. at 386-89.

n286. See Good News Club v. Milford Cent. Sch., 533 U.S. 98, 102-04 (2001).

n287. See Rosenberger v. Rectors & Visitors of the Univ. of Va., 515 U.S. 819, 820-27 (1995). Board of Education v. Mergens is another case that addresses these issues, although Mergens does so in the context of the Equal Access Act and not the First Amendment. See Bd. of Educ. v. Mergens, 496 U.S. 226 (1990).

n288. See Good News Club, 533 U.S. at 119; Rosenberger, 515 U.S. at 845-46; Lamb's Chapel, 508 U.S. at 394-97; Widmar, 454 U.S. at 276-77. For free speech purposes, the Court has said “it is axiomatic that the government may not regulate speech based on its substantive content or the message it conveys.” Rosenberger, 515 U.S. at 828 (citing Police Dep't of Chicago v. Mosley, 408 U.S. 92, 96 (1972)). Discrimination against speech because of the message conveyed is presumptively unconstitutional and, furthermore, “when the government targets not subject matter, but particular views taken by speakers on a subject, the violation of the First Amendment is all the more blatant.” Id. at 829 (citing R.A.V. v. St. Paul, 505 U.S. 377, 391 (1992)). The Court therefore characterizes viewpoint discrimination as “an egregious form of content discrimination.” Id. at 829.

n289. For example, in Widmar the Court explained that, “through its policy of accommodating their meetings, the University has created a forum generally open for use by student groups. Having done so, the University has assumed an obligation to justify its discriminations and exclusions under applicable constitutional norms.” Widmar, 454 U.S. at 267; see also Rosenberger, 515 U.S. at 829 (“Once it has opened a limited forum, however, the State must respect the lawful boundaries it has itself set.”). The speech forum thereby created should be distinguished from a “public forum” which by its nature or design is “open for indiscriminate public use for communicative purposes.” Lamb's Chapel, 508 U.S. at 392.

n290. See Lamb's Chapel, 508 U.S. at 386 (quoting N.Y. Educ. Law 414 (McKinney 1988 & Supp. 1993) (explaining that New York Education Law section 414 “authorizes local school boards to adopt reasonable regulations for the [after-school] use of school property for 10 specified purposes”); Good News Club, 533 U.S. at 102-03 (same); see also N.Y. Educ. Law 414 (McKinney 2000). In Widmar, the Court explained that the stated policy of the University of Missouri was “to encourage the activities of student organizations,” that it “officially recognized over 100 student groups,” and that it “routinely provided University facilities for the meetings of registered organizations.” Widmar, 454 U.S. at 265. The Christian group at issue in Widmar had “regularly sought and received permission to conduct its meetings in University facilities” until the University adopted its policy of religious exclusion. Id.

n291. Rosenberger, 515 U.S. at 824 (describing University guidelines relating to Student Activity Fund access). Notice that the forum created in Rosenberger involved more than equal access to facilities - it involved equal access to funding. See Paulsen, supra note 226, at 654 (“Equal access, according to the Court in Rosenberger, means no discrimination in eligibility for a right, benefit, or privilege - including funding - on the basis of religious viewpoint.” (emphasis omitted)). Paulsen calls Rosenberger's recognition of a free-speech right to equal access to a “funding” forum “a major doctrinal breakthrough in First Amendment law.” Id. at 710. He also points out that the same issue (equal access of religious persons to neutral sources of public funding) was presented on remand in Witters. Id. at 711 n.140. Paulsen's analysis of Rosenberger thus underscores the obvious connections between religious free speech and free exercise jurisprudence.

n292. In Lamb's Chapel and Good News Club, the school boards had promulgated rules stating that “school premises shall not be used by any group for religious purposes” and that otherwise forbade use “by any individual or organization for religious purposes.” Lamb's Chapel, 508 U.S. at 387; Good News Club, 533 U.S. at 103. Similarly, in Widmar, the University adopted a regulation that prohibited use of University buildings or grounds ““for purposes of religious worship or religious teaching.'“ Widmar, 454 U.S. at 265 & n.3. The exclusion in Rosenberger, as befitted a University setting, was more philosophically-nuanced. Among certain student activities excluded from the Student Activity Fund were “religious activities,” defined as any activity that “primarily promotes or manifests a particular belief in or about a deity or an ultimate reality.” Rosenberger, 515 U.S. at 825.

n293. Lamb's Chapel, 508 U.S. at 392-93 (quoting Cornelius v. NAACP Legal Defense & Educ. Fund, Inc., 473 U.S. 788, 806 (1985)).

n294. Lamb's Chapel, 508 at 393-94; see also Good News Club, 533 U.S. at 112 (reaffirming the consistent view that “speech discussing otherwise permissible subjects cannot be excluded from a limited public forum on the ground that the subject is discussed from a religious viewpoint”).

n295. Rosenberger, 515 U.S. at 831.

n296. See, e.g., Good News Club, 533 U.S. at 110-12. In Good News Club, the Court made its most pointed rejection of the argument that the “religious nature” of speech somehow makes it fair game for exclusion. The school had claimed that the explicit Christian content of the Good News Club's teaching activities distinguished them from “pure” moral teaching and character development. In the school's view, the Club's “Christian viewpoint” was “quintessentially religious” and therefore added an “additional layer” to otherwise neutral moral teaching. The Court rejected the school's argument, stating that “we can see no logical difference in kind between the invocation of Christianity by the Club and the invocation of teamwork, loyalty, or patriotism by other associations to provide a foundation for their lessons.” Id. at 111.

n297. For instance, in Mueller v. Allen, the Court approved under the Establishment Clause a general education tax deduction - one that included deductions for religious education expenses - for the primary reason that the allowable expenses were incurred by all parents, regardless of whether their children attended public, private non-religious, or private religious schools. 463 U.S. 388 (1983). The Court explicitly relied on the “state's provision of a forum neutrally “available to a broad class of nonreligious as well as religious speakers'“ in Widmar to support its conclusion that the tax deduction at issue was also “neutral” for non-establishment purposes. See id. at 397 (1983) (quoting Widmar, 454 U.S. at 274). Given Mueller's reliance on Widmar, it is easier to see the logic of Rosenberger, which “extended” the notion of a speech forum to a forum defined by a neutral funding mechanism. See, e.g., Paulsen, supra note 226, at 711 n.139 (stating that “arguably, Rosenberger is a step beyond Mueller and Zobrest in that it upholds direct state funding of specifically religious activities”).

n298. Rosenberger, 515 U.S. at 846 (O'Connor, J., concurring).

n299. Id. at 845.

n300. Id. at 845-46.

n301. See Paulsen, supra note 226, at 662 (“There is no “religion exception' to the Free Speech Clause or the Free Press Clause; religious speakers and groups are entitled to the same equal access to public fora, public facilities, and public funds as other private speakers and groups receive.”).

n302. E.g., Good News Club, 533 U.S. at 112-19; Widmar, 454 U.S. at 270-76; see also Rosenberger, 515 U.S. at 837-45; Lamb's Chapel, 508 U.S. at 394-97.

n303. Widmar, 454 U.S. at 275. The University relied in part on the general anti-religious-funding provision in article IX, section 8 of the Missouri Constitution, the only possibly relevant part of which provides that no “grant or donation of personal property or real estate [shall] ever be made by [any governmental unit] for any religious creed, church, or sectarian purpose whatever.” See supra note 107 and accompanying text (discussing Missouri Blaine Amendment). The University also relied on article I, section 6 (addressing the “seminary fund”) and article I, section 7 (addressing “county and township school funds”), neither of which seem applicable to the access issue nor to fall within the general parameters of State Blaine Amendments as I have described them. Nonetheless, the Supreme Court deferred to statements of the Missouri Supreme Court that the “Missouri Constitution requires stricter separation of church and State than does [the] Federal Constitution.” Widmar, 454 U.S. at 275 n.16 (citing Americans United v. Rogers, 538 S.W.2d 711, 720 (Mo. 1976) (en banc)).

n304. Widmar, 454 U.S. at 275.

n305. Id. at 276.

n306. Id.

n307. Paulsen, supra note 226, at 653 (emphasis omitted).

n308. Davey v. Locke, 299 F.3d 748 (9th Cir. 2002), cert. granted, 123 S. Ct. 2075 (2003).

n309. See Witters v. Wash. Dep't of Servs. for the Blind, 474 U.S. 481 (1986) (Witters II); see also supra Part I.

n310. Witters v. State Comm'n for the Blind, 771 P.2d 1119, 1122 (Wash. 1989) (Witters III); see also supra note 116 (discussing Wash. Const. art. I, 11).

n311. Witters III, 771 P.2d at 1122 (citation omitted).

n312. See supra note 119.

n313. See supra note 107.

n314. See supra note 120.

n315. Some candidates might be those State Blaines whose prohibitions appear limited to specific “funds” (such as “educational” or “public school” funds), because Witters' aid apparently came from a vocational rehabilitation fund. See, e.g., supra notes 105, 103, 107 (discussing the Kansas, Ohio, and Nebraska Constitutions).

n316. This would be different, of course, if the federal Establishment Clause independently prohibited Witters' use of the funds. In that case, construction of the State Blaine would not logically be implicated.

n317. Everson v. Bd. of Educ., 330 U.S. 1, 16 (1947); see supra note 182.

n318. Again, notice that the federal Establishment Clause does not prohibit the religious use of the aid contemplated by Witters. Thus, the pool of aid is genuinely “generally available” to Witters. Washington State is thus penalizing Witters' religious choice because it is religious, and not because its hands are tied by the Establishment Clause.

n319. See Church of the Lukumi Babalu Aye, Inc. v. City of Hialeah, 508 U.S. 520, 534 (1993); Walz v. Tax Comm'n, 397 U.S. 664, 696 (1970) (Harlan, J., concurring); see also supra notes 218-23, 230, 279 and accompanying text.

n320. See Paulsen, supra note 226, at 711-12 & nn.139-40 (explaining Rosenberger's precedential implications for neutral governmental funding programs and observing that the same principles were involved in Witters on remand). Indeed, as I have explained, the Court itself has drawn the analogy between the limited speech fora in the religious speech cases, and the notion of a “neutral” distribution of public funds based on non-religious criteria. See, e.g., Mueller v. Allen, 463 U.S. 388, 397 (1983) (quoting Widmar v. Vincent, 454 U.S. 263, 274 (1981)); supra notes 1-4. Both the majority opinion and Justice O'Connor's concurrence seemed to flinch from embracing the logical application of Rosenberger's holding to neutral disbursements from “general tax revenue.” See Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819, 840-41 (1995) (attempting to distinguish the student fees disbursements from an expenditure from a general tax fund); id. at 851-52 (O'Connor, J., concurring) (claiming that the student fund “simply belongs to the students” and is not “tax revenue”). The distinction is unpersuasive. It is difficult to understand how the student fee program - which exacts fees from all students and makes them neutrally available for student groups' private uses - is constitutionally different from the same kind of program involving “general tax revenues.” See, e.g., Paulsen, supra note 226, at 712 (criticizing as unpersuasive the majority's and Justice O'Connor's qualifications of the holding in Rosenberger as applied to a “general tax fund”); Laycock, supra note 156, at 66-67 & n.144 (arguing that the Rosenberger “majority hedged the opinion with unpersuasive distinctions and reservations” about general tax revenues and directness of funding); see also Mueller, 463 U.S. at 397 (comparing limited speech forum in Widmar to generally available tax deduction for educational expenses). Since Rosenberger, the Court has relied on the limited forum cases for “instruction” in assessing the constitutionality of a government subsidy programs derived from general tax funds. See Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 544 (2001) (observing that “limited forum” cases like Lamb's Chapel and Rosenberger “do provide some instruction” for cases in which “government establishes a subsidy for specified ends”).

n321. See, e.g., Paulsen, supra note 226, at 657 (arguing that “Rosenberger's equal access to funding follows naturally from Widmar, Mergens, and Lamb's Chapel, each of which involved a claim of some type on public resources by a religious group”).

n322. See Witters v. State Comm'n for the Blind, 771 P.2d 1119, 1122 (Wash. 1989) (Witters III) (citations omitted).

n323. See McDaniel v. Paty, 435 U.S. 618, 626-27 (1978); id. at 632 (Brennan, J., concurring); Torcaso v. Watkins, 367 U.S. 488, 489-90 (1961).

n324. See Lupu & Tuttle, supra note 19, at 962 n.204 (offering viewpoint discrimination as a narrower alternative ground for result in Davey).

n325. Witters III, 771 P.2d at 1122 (citations omitted). This “motivational” parsing of a State Blaine merely deepens its unconstitutional application as to Witters. But a “categorical” reading would amount to unconstitutional religious discrimination as well. That is, if the Washington Supreme Court had simply declared that all religious studies were ineligible for funding - whether or not they were “devotional” - it would still have singled out “religious” as a category excluded from public benefits. Nothing in the Court's development of the non-persecution principle would limit persecution to discrimination against devotional religious motivation only. But the Court has suggested that religious discrimination targeted at particular qualities of belief is especially disfavored. See, e.g., Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872, 876-77 (1990) (citing Sherbert v. Verner, 374 U.S. 398, 402 (1963), and Torcaso, 367 U.S. at 488).

n326. See Davey v. Locke, 299 F.3d 748 (9th Cir. 2002), cert. granted, 123 S. Ct. 2075 (2003).

n327. Washington's certiorari petition lists thirteen other states with similar statutory funding restrictions on financial aid to theology or divinity students. Petition for Certiorari at 21 & n.4, Davey (No. 02-1315) (citing laws from Alabama, Florida, Kentucky, Louisiana, Maryland, Michigan, Missouri, New York, Ohio, Oregon, South Carolina, South Dakota, and Wisconsin).

n328. The scholarship paid $ 1,125 during the 1999-2000 year and $ 1,542 for 2000-01 and could be spent on any educational expense, including room and board. Davey, 299 F.3d at 750-51. The general eligibility criteria require that students (1) be in the top 10% of their 1999 high school graduating class; (2) have a family income no greater than 135% of the state median income; and (3) attend an accredited public or private university in Washington. Id. at 751.

n329. Washington defines an “eligible student” as “a person who ... is not pursuing a degree in theology.” Id. at 751 n.3. The eligibility criteria are codified in Wash. Admin. Code 250-80-020(12)(a)-(f). Id. The court also noted that Wash. Rev. Code 28B.10.814 provides that “no aid shall be awarded to any student who is pursuing a degree in theology.” Id. at 750 n.1. The court did not say whether “theology” is defined by Washington state law. See generally id. at 748.

n330. Id. at 758; see supra note 3 and accompanying text. The plaintiff, Joshua Davey, was in virtually the same situation as Witters. Already selected as a Promise Scholar, Davey enrolled in an accredited private Christian school, Northwest, intending to enter the ministry, and declared a double major in Pastoral Ministries and Business. The Pastoral Ministries major was “designed to prepare students for a career as a Christian minister.” Davey, 299 F.3d at 751. Northwest's theology offerings were grounded on the assertion that “the Bible represents truth and is foundational,” whereas theology curricula at Washington public universities were generally “taught from an historical and scholarly point of view.” Id. Washington determined that Davey's major in Pastoral Ministries constituted a “theology” degree and therefore disqualified him for scholarship eligibility. Id. Davey chose to forego the scholarship and continued to pursue his major. Id. at 751.

n331. Id. at 752-58. Judge McKeown dissented, relying primarily on the federalism and funding objections that I address in this and the next section. Id. at 760-68 (McKeown, J., dissenting); see infra notes 342, 398.

n332. It was unclear from the Ninth Circuit's opinion whether the statutory exclusion in Davey has the additional vice, as Witters did, of excluding only “devotional” theology courses. See, e.g., Davey, 299 F.3d at 755-56, 760; Lupu & Tuttle, supra note 19, at 962 n.203.

n333. For a thoughtful defense of Davey, see Kent Greenawalt, Is It Davey's Locker for the No-Funding Principle?, 45 J. of Church & St. (forthcoming Dec. 2003).

n334. See Lupu & Tuttle, supra note 19, at 957-72 (2003). Their objections are not directed specifically toward the application of State Blaines in Witters and Davey, but instead are more general. That said, the authors do suggest that Davey would have been better resolved as a case of viewpoint discrimination. See id. at 962 n.204.

n335. Id. at 963-64. My approach, although normatively similar to the approach Lupu and Tuttle criticize, draws on jurisprudence not only from the Free Exercise Clause but also from the Establishment and Free Speech Clauses. That said, I think the Free Exercise Clause is the most apt constitutional locus for the State Blaines' unconstitutional operation.

n336. Lupu & Tuttle, supra note 19 at 964.

n337. Id. The authors cite, inter alia, Mitchell v. Helms, 530 U.S. 793, 809 (2000) (reaffirming that the government cannot subsidize religion by using aid that “results in governmental indoctrination”); Agostini v. Felton, 521 U.S. 203, 228-29 (1997) (holding that the government may not directly subsidize religion); Jones v. Wolf, 443 U.S. 595, 602-03 (1979); Serbian East Orthodox Diocese v. Milivojevich, 426 U.S. 696, 708-12 (1976) (holding that the government may not intervene in property or personnel disputes that are internal to religious communities and organizations and involve religious matters).

n338. Lupu & Tuttle, supra note 19, at 964. The authors also point to the doctrine excepting clergy-congregation relationships from federal anti-discrimination law, id. at 964 n.216, as well as various religious freedom restoration acts enacted by the federal government and many states in response to Smith, id. at 965 n.217 (citations omitted).

n339. Id. at 965-66.

n340. Id. at 965.

n341. Id.

n342. Id. at 966. The authors are cautious, however, about saying what such “reasonable purposes” might be. They admit that the purposes supporting a “regime of Separationism” are in need of “restatement and reinvigoration,” especially because current defenders of separationism - the Zelman dissenters, for instance - “have tended to rely excessively on justifications now viewed by many as outmoded.” Id. The authors conclude by stating that “whether states can defend a Separationist policy broader than the federal constitution requires will thus depend on the efforts of judges and academics to provide precisely this sort of rehabilitation of the Separationist ethos.” Id. The authors point to two of their articles as laying some possible groundwork. Id. at 966 n.222 (citing Lupu & Tuttle, supra note 10; Ira C. Lupu & Robert W. Tuttle, Sites of Redemption: A Wide-Angle Look at Government Vouchers and Sectarian Service Providers, 18 J.L. & Pol. 537 (2002)). Along those lines, the dissenter in Davey, Judge McKeown, herself articulated some “reasonable purposes” for Washington's Blaine Amendment. Washington, she said, could justify its State Blaine in order to “define[] its vision of religious freedom as one completely free of governmental interference,” to “reflect[] its strong desire ... to insulate itself from the appearance of endorsing religion,” and to evince “the state's strong prophylactic interest in steering clear of endorsing or supporting religion through direct funding of religious pursuits.” Davey v. Locke, 299 F.3d 748, 761-62, 766 (McKeown, J., dissenting).

Lupu's and Tuttle's suggestions are intriguing, but they leave unanswered a fundamental question. Even if judges or academics succeed in “reinvigorating” the purposes of the “Separationist ethos” - an ethos the authors admit is currently founded on a tissue of anachronism and anti-religious hostility - why should their “rehabilitated” purposes suffice as legitimate, not to mention compelling, justifications for states' targeted exclusion of religious persons and groups from public benefits? Regardless of what rejuvenated brew of “Separationism” might be concocted, the legal operation of that “ethos” will still be measured against the free exercise rights of religiously motivated state citizens who, needless to say, will continue to object to their religion-based second-class citizenship. In short, it is implausible that new reasons for religious discrimination will prove any more legitimate or compelling than the old reasons.

n343. Lupu & Tuttle, supra note 19, at 964.

n344. See, e.g., Kathleen M. Sullivan, The New Religion and the Constitution, 116 Harv. L. Rev. 1397, 1403 (2003) (observing that “the constitutional jurisprudence of the Religion Clauses navigates among competing tacit accounts of the role of religious organizations in a democratic society”).

n345. Reynolds v. United States, 98 U.S. 145 (1878).

n346. Everson v. Bd. of Educ., 330 U.S. 1 (1947).

n347. See supra notes 263-72 (discussing the non-hostility thread in Everson, Bowen, Rosenberger, Grumet, Agostini, Mitchell, and Zobrest).

n348. See, e.g., Zorach v. Clauson, 343 U.S. 306, 314 (1952) (stating that when the legislature acts to accommodate religious belief or practice, it “follows the best of our traditions”); see also Bd. of Educ. v. Grumet, 512 U.S. 687, 705 (1994); id. at 714 (O'Connor, J., concurring); id. at 723 (Kennedy, J., concurring); id. at 743-45 (Scalia, J., dissenting). Each of the Justices acknowledged the consistent American legal tradition of accommodating religious belief and practice.

n349. See, e.g., Holy Trinity Church v. United States, 143 U.S. 457, 465-72 (1892) (explaining that “no purpose of action against religion can be imputed to any legislation, state or national, because this is a religious people”); Davis v. Beason, 133 U.S. 333, 341 (1890) (remarking that “bigamy and polygamy are crimes by the laws of all civilized and Christian countries”); Vidal v. Girard's Ex'rs, 43 U.S. 127, 198-99 (1844) (stating it is unnecessary “to consider what would be the legal effect of a devise in Pennsylvania for the establishment of a school or college, for the propagation of Judaism, or Deism, or any other form of infidelity [because] such a case is not to be presumed to exist in a Christian country; and therefore it must be made out by clear and indisputable proof”); People v. Ruggles, 8 Johns. 290, 294, 296 (N.Y. Sup. Ct. 1811) (Kent, J.) (stating that “the people of this state, in common with the people of this country, profess the general doctrines of christianity, as the rules of their faith and practice” and that “though the constitution has discarded religious establishments, it does not forbid judicial cognisance of those offences against religion and morality which have no reference to any such establishment”); see also 3 Joseph Story, Commentaries on the Constitution of the United States 1871, at 728 (Fred B. Rothman & Co. 1991) (1833) (“The real object of the [Establishment Clause] was, not to countenance, much less to advance Mahometanism, or Judaism, or infidelity, by prostrating Christianity; but to exclude all rivalry among Christian sects.”), discussed in Amar, supra note 1, at 252 n; see also Amar, supra note 1, at 247 (discussing the First Congress's “extending the Confederate Congress's Northwest Ordinance of 1787, a regime that one leading scholar has described as “suffused with aid, encouragement, and support for religion'“ (quoting Gerard V. Bradley, Church-State Relationships in America 98 (1987))).

n350. See supra Part IV.B. (discussing Kurland and formal neutrality).

n351. See, e.g., Ira C. Lupu, The Increasingly Anachronistic Case against School Vouchers, 13 Notre Dame J.L. Ethics & Pub. Pol'y 375, 386 (1999) (stating that “the Protestant paranoia fueled by waves of Catholic immigration to the United States, beginning in the mid-nineteenth century, cannot form the basis of a stable constitutional principle, and the stability of the principle has been undermined by the amelioration of the concerns” (citing Hamburger, supra note 173)).

n352. See, e.g., Berg, supra note 12, at 122-23, 151-52, 161-62 (commenting on flux of “strict separationism” in religion jurisprudence and that “a distrust of Catholic power and Catholic education was still a factor in the stricter “no-aid' separationism of the 1960s and 1970s,” although less so than in the 1940s and 50s); Laycock, supra note 156, at 53-54 (discussing tension between the “no-aid” and “non-discrimination” strands in the Court's religion jurisprudence, beginning with Everson); Lupu, supra note 351, at 388 (asking “if the line of decisions from Everson to Lemon was driven substantially by the then-demographics of public and private education, coupled with anti-Catholic animus, what remains to justify principles forbidding direct aid to sectarian elementary and secondary schools?”); McConnell, supra note 17, at 120, 127 (commenting on the tendency of the Warren and Burger Courts “to press relentlessly in the direction of a more secular society” and “to view religion as an unreasoned, aggressive, exclusionary, and divisive force that must be confined to the private sphere”); id. at 127 (arguing that the Warren and Burger Courts' “legal doctrines ... reinforced their lack of sympathy for religion”).

n353. See Everson v. Bd. of Educ., 330 U.S. 1, 15-16 (1947) (stating that “neither a state nor the Federal government can ... aid one religion, [or] aid all religions ... . No tax in any amount, large or small, can be levied to support any religious activities or institutions, whatever they may be called, or whatever form they may adopt to teach or practice religion”).

n354. See, e.g., Hamburger, supra note 31, at 454-63 (discussing misapprehension of the Everson parties and Justices about the nature of Establishment Clause); see also Wallace v. Jaffree, 472 U.S. 38, 92 (1985) (Rehnquist, J., dissenting) (generally criticizing Court's non-establishment jurisprudence and observing that “it is impossible to build sound constitutional doctrine upon a mistaken understanding of constitutional history”).

n355. See, e.g., Lupu & Tuttle, supra note 19, at 949-52 (criticizing Justice Souter's no-aid separationism); Fried, supra note 13, at 188 (criticizing Souter's Zelman dissent because it treated “twenty years of jurisprudence” from Mueller to Zobrest “as a mistake,” and because Souter's no-aid separationism was actually reflected in the Court's jurisprudence for a “relatively brief” period from 1971-83).

n356. See, e.g., Lupu & Tuttle, supra note 19, at 952-55 (criticizing Justice Breyer's concerns with religious divisiveness). Lupu and Tuttle argue that Breyer's Zelman dissent “shows deep insensitivity to the history, limits, and failings of the concerns for “political divisiveness,'“ and relies on “a history of Protestant-Catholic tension in the United States that, if anything, should embarrass a Court that spawned the regime of no-aid Separationism out of deeply anti-Catholic premises.” Id. at 954.

n357. See, e.g., id. at 952 n.162 (noting Justice Stevens' “long and unbroken record of opposing the cause of religion no matter what the issues presented”); Bd. of Educ. v. Grumet, 512 U.S. 687, 749 (1994) (Scalia, J., dissenting) (claiming that Justice Stevens' concurrence was “less a legal analysis than a manifesto of secularism” that “announced a positive hostility to religion”); Santa Fe Indep. Sch. Dist. v. Doe, 530 U.S. 290, 318 (2000) (Rehnquist, C.J., dissenting) (contending that Stevens' majority opinion “bristles with hostility to all things religious in public life”); see also Berg, supra note 12, at 129 (commenting on anti-Catholic rhetoric in the opinions of Justices Black, Douglas, and Rutledge); Laycock, supra note 156, at 57 (discussing historical scholarship documenting that the “intellectual anti-Catholic movement [of the mid-1900s] attracted the favorable attention of Justices Black, Frankfurter, Rutledge, and Burton” (citation omitted)); Lupu, supra note 351, at 385 (commenting that Justice Jackson's Everson dissent and Chief Justice Burger's Lemon opinion were “open and conspicuous tracts about the pervasive religious indoctrination thought to accompany the system of Catholic education”); McConnell, supra note 17, at 121-22 (commenting on Black's anti-Catholic bias in his Allen dissent).

n358. See, e.g., Lupu & Tuttle, supra note 19, at 918 (commenting that, on the eve of Zelman, “only the most ostrich-like Separationist could have denied the flux in the law of the Establishment Clause,” explaining that “in the context of access of private parties to public fora for purposes of religious expression, and direct government transfer of material resources to religious institutions, norms of non-Establishment have been tending sharply toward the paradigm of Neutrality and away from the metaphorical wall of church-state separation” (citations omitted)).

n359. See supra notes 343-44 and accompanying text.

n360. See, e.g., Zelman v. Simmons-Harris, 536 U.S. 639, 661-62 (2002) (sharply limiting Board of Educ. v. Nyquist, 413 U.S. 756 (1973)); Mitchell v. Helms, 530 U.S. 793, 808 (2000) (plurality opinion) (overruling Meek and Wolman); id. at 837 (O'Connor, J., concurring) (agreeing with the plurality); Agostini v. Felton, 521 U.S. 203, 222-35 (1997) (overruling Aguilar v. Felton, 473 U.S. 402 (1985)); see also Paulsen, supra note 226, at 711 n.138 (noting that Nyquist has not been formally overruled but it “must be regarded as moribund in light of ... subsequent decisions”).

n361. Lupu & Tuttle, supra note 19, at 964.

n362. See, e.g., Lupu, supra note 351, at 376 (questioning “the force of the constitutional case against direct state aid to sectarian elementary and secondary schools” and claiming that “the arguments against direct aid rest on precedents and policies whose contemporary relevance has dwindled dramatically”); id. at 377-80 (criticizing the jurisprudential foundation for the “direct/indirect” distinction); id. at 388-93 (questioning reliance on General Assessment controversy and Madison's Memorial as the basis for “direct funding” prohibition). But see Greenawalt, supra note 333 (defending the continued vitality of a no-funding principle in non-establishment law).

n363. As explained infra, this question is bound up with the issue of how incorporation of the religion clauses against the states affects the states' power to craft a church-state separation greater than the federal Establishment Clause requires. See infra 369-85.

n364. Bd. of Educ. v. Grumet, 512 U.S. 687, 714 (1994) (O'Connor, J., concurring) (citing U.S. Const. art. II, 1, cl. 8; art. VI, cl. 3).

n365. See supra notes 23, 174, 179.

n366. Amar, supra note 1 at 33-34, 41, 246.

n367. Id. at 248 (citing Jed Rubenfeld, Antidisestablishmentarianism: Why RFRA Really Was Unconstitutional, 95 Mich. L. Rev. 2347 (1997)).

n368. See supra note 25.

n369. Akhil Amar and Kurt Lash have suggested that the “reconstructed” Free Exercise Clause can plausibly be interpreted to protect religious exercise more broadly than the original clause, requiring for instance religious exemptions from non-discriminatory general laws. Amar, supra note 1, at 254-56; Lash, supra note 23, at 1149-56.

n370. Lupu & Tuttle, supra note 19, at 965. Notice that the result would be no different if the invalidated policy had “been federal constitutional law a few short years ago” - i.e., if the Supreme Court had held previously that the policy did not violate free exercise, but reversed itself. Id.

n371. See Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872, 890 (1989).

n372. Amar, supra note 1 at 252.

n373. See id. at 33-34, 41, 251-54; McConnell, supra note 21, at 1485 n.384.

n374. Even this statement becomes tangled when we notice, as Akhil Amar explains, that “what the Establishment Clause prohibited the federal Congress from doing” was, in large part, “meddling with state establishments.” See Amar, supra note 1, at 33-34, 41.

n375. Lupu and Tuttle do not address why this inevitable effect of incorporation is not equally “hostile to notions of respect for state law, and in particular to the tradition of independent state constitutional law.” See Lupu & Tuttle, supra note 19, at 965-66.

n376. Amar, supra note 1, at 32-34, 41, 246-56.

n377. Id. at 251-56; see generally id. at 215-30 (explaining “refined incorporation”).

n378. Id. at 252.

n379. Id. Amar also suggests that state citizens might also claim certain refined non-establishment rights that are not strictly grounded in principles of “coercion,” but that sound rather in the “basic touchstones” of Fourteenth Amendment “ideals of liberty and equality.” Id. at 253-54. By this, he seems to mean that state citizens might be able to object to state laws on the basis of religious equality, such as if a state favored one religious denomination or declared itself “The Baptist State.” Id. At the same time, Amar admits that non-establishment incorporation “may not matter all that much” in such cases since “principles of religious liberty and equality could be vindicated via the free-exercise clause (whose text, history, and logic make it a paradigmatic case for incorporation) and the equal-protection clause.” Id. at 254.

n380. Justice Thomas has picked up on Amar's suggestion. See Zelman v. Simmons-Harris, 536 U.S. 639, 676-81 & n.4 (2002) (Thomas, J., concurring) (citing Akhil Amar, supra note 25, at 1159, and Lietzau, supra note 25, at 1206-07); see also Lupu & Tuttle, supra note 19, at 947-49.

n381. Amar, supra note 1, at 254.

n382. For an illuminating discussion of the irresolvable contradictions raised by the notion that state and federal governments can legitimately pursue different church-state policies in this area, see Viteritti, supra note 30, at 1154-55 (arguing that, in this area, “the differences between federal and state standards are so basic that they cannot coexist within a single constitutional framework”).

n383. See, e.g., id. at 1154 (discussing the Washington Supreme Court's decision on remand in Witters III and arguing that “while secularists in Washington [State] were confident that the state court was exercising legitimate authority to prevent indirect aid to a religious school, the action by the state court also served to encumber the constitutional right of the seminary student to choose a school that reflected his own values and aspirations”).

n384. See, e.g., DeForrest, supra note 20, at 605-06 (generally discussing federal constitutional limitations on State Blaines that arise inevitably from incorporation).

n385. See, e.g., Rosenberger v. Rector & Visitors of the Univ. of Va., 515 U.S. 819, 832 (1995) (noting the “unremarkable proposition that the State must have substantial discretion in determining how to allocate scarce resources to accomplish its educational mission”); McConnell, supra note 165, at 989 (“The government cannot spend money on everything. It must be selective.”).

n386. See, e.g., McConnell, supra note 165, at 1001 & n.35 (stating that it is “surely correct that there is no ... general obligation” for government to “provide the material resources necessary for the exercise of a constitutional right” (citing DeShaney v. Winnebago County Dep't of Soc. Servs., 489 U.S. 189, 198-200 (1989))).

n387. See, e.g., Legal Servs. Corp. v. Velazquez, 531 U.S. 533, 541 (2001); Bd. of Regents of the Univ. of Wis. Sys. v. Southworth, 529 U.S. 217, 229, 235 (2000); Rosenberger, 515 U.S. at 833; Rust v. Sullivan, 500 U.S. 173, 193-95 (1991).

n388. See, e.g., Harris v. McRae, 448 U.S. 297, 312-18 (1980); Maher v. Roe, 432 U.S. 464, 474-76 (1977).

n389. That request itself would run aground on the legitimate historical concerns behind the religion-funding controversies of the early republic. See, e.g., Laycock, supra note 156, at 48-49.

n390. See, e.g., McConnell, supra note 17, at 184. McConnell argues that:

When the government provides financial support to the entire nonprofit sector, religious and nonreligious institutions alike, on the basis of objective criteria, it does not aid religion. It aids higher education, health care, or child care; it is neutral to religion. Indeed, to deny equal support to a college, hospital, or orphanage on the ground that it conveys religious ideas is to penalize it for being religious.

Id. (emphasis omitted).

n391. See, e.g., Velazquez, 531 U.S. at 541 (observing that “we have said that viewpoint-based funding decisions can be sustained in instances in which the government is itself the speaker” (citing Southworth, 529 U.S. at 229, 235)); Rosenberger, 515 U.S. at 833 (recognizing “the principle that when the State is the speaker, it may make content-based choices” such as when a public university “determines the content of the education it provides”).

n392. Rosenberger, 515 U.S. at 833 (citing Rust, 500 U.S. at 194, and Widmar v. Vincent, 454 U.S. 263, 276 (1981)); see also Velazquez, 531 U.S. at 541.

n393. It is doubtful, for instance, that government could craft funding programs to further its own “religious” speech. This would cut against the dominant non-establishment principle that government must have secular purposes for its laws. As for the use of religious speech by government itself - e.g., religious language in a presidential speech, or the employment of legislative chaplains by Congress - those instances are either non-justiciable (presidential speech) or are permissible under the Establishment Clause (chaplains). See, e.g., Marsh v. Chambers, 463 U.S. 783 (1983). Perhaps a Blaine Amendment could be interpreted by a state government to forbid the funding of state legislative chaplains or prayers, or to prohibit public officials from using any religious language in public speeches, or to prohibit any religious symbolism whatsoever on public property. As I explained supra, however, those applications of a State Blaine to create a greater church-state separation than the federal Constitution demands would probably not run afoul of the non-persecution principle, because they do not plausibly limit anyone's federal free exercise rights. See supra notes 386-88 and accompanying text.

n394. For example, one might claim that the inclusion of a religiously-affiliated organization in a government message program would - even if the organization fully complied with the speech requirements of the program - nonetheless run afoul of a State Blaine that forbade public funds from being spent “for the benefit of,” “in aid of,” or “in support of” any “church,” “religious society,” or “religious institution.” Similarly, one might claim such inclusion would constitute an “appropriation” of public funds “in aid of” or “for the benevolent purposes of” a religious group.

n395. Rosenberger, 515 U.S. at 833 (citing Rust, 500 U.S. at 196-200). Nor would it be any less illegitimate if the same “anti-religious-participant” notion were expressed in the government's definition of the program itself - i.e., if the government program were described as a “non-religious child care program.” See, e.g., Paulsen, supra note 226, at 666 n.32 (rejecting “definitional manipulation” of a limited public forum to incorporate “the precise condition that is substantively unconstitutional”).

n396. See, e.g., McConnell, supra note 164, at 39-40 (commenting on government's “power to create incentives for individuals to alter their conduct by providing financial support to one choice and not to a substitute”).

n397. See Harris v. McRae, 448 U.S. 297, 314 (1980) (constitutional protection afforded a woman's choice to have abortion “did not prevent [the state] from making “a value judgment favoring childbirth over abortion and ... implementing that judgment by the allocation of public funds'“ (quoting Maher v. Roe, 432 U.S. 464, 474 (1977)); see generally McConnell, supra note 165, at 989-92, 1000-01 (discussing abortion funding decisions).

n398. It was, for example, the rhetorical centerpiece of Judge McKeown's dissent in Davey. See Davey v. Locke, 299 F.3d 748, 764-66 (9th Cir. 2002) (McKeown, J., dissenting).

n399. See, e.g., Sullivan, supra note 344, at 1415 (“Government use of funding leverage can exert coercion, as a long line of constitutional conditions decisions suggests.”); McConnell, supra note 165, at 1015 (noting that “[a] common understanding of constitutional law is that although the government has no obligation (absent exceptional circumstances) to subsidize the exercise of constitutional rights, it is forbidden to penalize the exercise of those rights”).

n400. See McConnell, supra note 165, at 989 (asking “when is the government's refusal to fund a constitutionally protected choice an impermissible “burden' on the exercise of the right?”); see also Davey, 299 F.3d at 754-55 (stating that government “may selectively sponsor or pay for programs that it believes to be in the public interest” but “government may not deny a benefit to a person because he exercises a constitutional right” (citing Regan v. Taxation With Representation, 461 U.S. 540, 545 (1983))).

n401. See generally Richard A. Epstein, Unconstitutional Conditions, State Power, and the Limits of Consent, 102 Harv. L. Rev. 5 (1988); Michael W. McConnell, Unconstitutional Conditions: Unrecognized Implications for the Establishment Clause, 26 San Diego L. Rev. 255 (1989); Kathleen M. Sullivan, Unconstitutional Conditions, 102 Harv. L. Rev. 1413 (1989); see, e.g., Paulsen, supra note 226, at 665 n.30 (noting proliferation of scholarly refinements of unconstitutional conditions doctrine).

n402. Paulsen, supra note 226, at 664-65. The “directly germane” proviso is necessarily narrow, referring to “conditions that are directly “germane,' in the strong sense of being inextricably intertwined with the nature of the right or benefit itself.” Id. at 666 n.32. The exception is narrow, explains Paulsen, to prevent government from “circumventing the general rule against unconstitutional conditions by the expedient of simply defining its “limited' public forum in terms of the precise condition that is substantively unconstitutional.” Id.

n403. Id. at 665. Similarly, Michael McConnell explains that, in assessing selective funding problems, one must first engage in “careful consideration of the nature of the constitutional right implicated by the funding decision, including the nature of the countervailing interests of the government.” McConnell, supra note 165, at 992.

n404. See supra notes 1-4, 326-36.

n405. See McDaniel v. Paty, 435 U.S. 618, 626 (1978); see also id. at 632, 635 (Brennan, J., concurring) (arguing that ministerial exclusion penalizes both religious belief and status).

n406. Imagine, by contrast, that Witters' or Davey's religious use of the funds would have independently violated the Establishment Clause. Perhaps only in that sense would a “no religious use” condition on the funds have been “directly germane” to the funding program. Of course, in that instance, the condition would merely duplicate the federal non-establishment constraints on Washington.

n407. The loss of all scholarship funds underscores the penalizing nature of Washington's condition. This was not a case where someone is merely forced to “bear the costs” of exercising constitutional rights, but rather a case in which someone is “made worse off than he would have been had he not exercised” those rights. See McConnell, supra note 165, at 1015 (emphasis added). Because of their religious choices, Witters and Davey lost the entire scholarship, not merely the amount of money that might have gone toward “religious” instruction or training. Compared to a scholarship student enrolled, say, in biochemistry or philosophy, Witters and Davey are not merely “poorer,” proportionally speaking; instead, they have been excluded from the funds altogether. A wholesale exclusion from benefits, as opposed to a reduction in benefits only “to the extent of the cost of exercising the constitutional right,” is more in the nature of a penalty. See generally id. at 1015-19.

n408. Michael McConnell exhaustively explores various answers to this question in his Selective Funding article. See McConnell, supra note 165.

n409. See Planned Parenthood v. Casey, 505 U.S. 833, 874 (1992) (state regulation violates constitutional guarantee of liberty only if it “imposes an undue burden” on woman's choice to abort); see also Maher v. Roe, 432 U.S. 464, 473-74 (1977) (explaining that Roe v. Wade, 410 U.S. 113 (1973), did not declare an “unqualified “constitutional right to an abortion'“ but rather protected a woman from “unduly burdensome interference with her freedom to decide whether to terminate her pregnancy”). Casey explained that an undue burden is “a shorthand for the conclusion that a state regulation has the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus.” 505 U.S. at 877.

n410. See Casey, 505 U.S. at 876 (referring to “the recognition that there is a substantial state interest in potential life throughout pregnancy”); see also id. at 875 (observing that “in practice” Roe's trimester framework “undervalues the State's interest in the potential life within the woman”).

n411. See id. at 878

To promote the State's profound interest in potential life, throughout pregnancy the State may take measures to ensure that the woman's choice is informed, and measures designed to advance this interest will not be invalidated as long as their purpose is to persuade the woman to choose childbirth over abortion. These measures must not be an undue burden on the right.

Id.; see also McConnell, supra note 165, at 1034-38 (describing, pre-Casey, an alternative to a pure “privacy” rationale for abortion rights, one recognizing that “the government's interest in protecting unborn life is legitimate, but limited to non-coercive means”).

n412. See McConnell, supra note 165, at 1006 & n.49 (explaining the difference between reasons for selective funding that are “hostile” to rights - i.e., reasons that “depend for their persuasive power upon antipathy to the exercise of the rights in question” - and “non-hostile” reasons that “could be accepted even by proponents of the affected rights,” even if they were not persuaded by them) (emphasis omitted).

n413. See, e.g., Everson v. Bd. of Educ., 330 U.S. 1, 18 (1947) (stating that “State power is no more to be used so as to handicap religions than it is to favor them” and that the First Amendment “requires the state to be a neutral in its relations with groups of religious believers and non-believers; it does not require the state to be their adversary”); Bd. of Educ. v. Grumet, 512 U.S. 687, 717 (1994) (O'Connor, J., concurring) (“The Religion Clauses prohibit the government from favoring religion, but they provide no warrant for discriminating against religion.”).

n414. See supra Part IV.A. This forecloses the suggestion that there persists in free exercise jurisprudence a general form of balancing test analogous to the abortion-rights inquiry. Admittedly, the Sherbert line of unemployment compensation cases engaged in such balancing. See Employment Div. Dep't of Human Res. v. Smith, 494 U.S. 872, 883 (1990) (discussing the Sherbert balancing test); see also Sherbert v. Verner, 374 U.S. 398, 402-03 (1963). And, relying on Sherbert, Judge McKeown claimed in her Davey dissent that a “substantial burden” test was still the controlling standard for free exercise violations. See Davey v. Locke, 299 F.3d 748, 763-64 (9th Cir. 2002) (McKeown, J., dissenting). It is difficult to square that view with Smith, however. See Smith, 494 U.S. at 883-85 (confining applicability of Sherbert to cases, like the unemployment compensation context, where a benefit program invites “individualized governmental assessment of the reasons for the relevant conduct,” essentially empowering government to determine whether religious reasons justify compensation). Smith explicitly excludes any form of Sherbert balancing from cases involving “across-the-board criminal prohibition on a particular form of conduct.” Id. at 884. In my view, the best reading of these passages from Smith is that Sherbert is essentially dead, insofar as it advocates a “balancing” approach to free exercise challenges to general laws. See id. at 885 (stating that “the government's ability to enforce generally applicable prohibitions of socially harmful conduct, like its ability to carry out other aspects of public policy, “cannot depend on measuring the effects of a governmental action on a religious objector's spiritual development'“ (citation omitted) (emphasis added)).

n415. See supra notes 301-10, 320-25.

n416. See McConnell, supra note 165, at 1006 & n.49.

n417. Paulsen, supra note 226, at 667.

n418. See supra note 264.

n419. Much of the current debate over unconstitutional conditions on religious participation in public benefits addresses more subtle conditions on religious providers. The debate centers on whether religious providers' access to public benefits can be conditioned on their abandonment of principles or practices connected to their religious identity. For instance, may religious schools' participation in a neutral voucher program be conditioned on their not discriminating in selecting students on the basis of religion? On their not discriminating in hiring teachers on the basis of religion? On their agreement not to require voucher students to participate in religious observance or instruction? On their agreement not to impart religious teaching that may run afoul of anti-discrimination laws? See, e.g., Paulsen, supra note 226, at 662-63; Lupu & Tuttle, supra note 19, at 972-82; see generally Symposium, Public Values in an Era of Privatization, 116 Harv. L. Rev. 1212 (2003). This important inquiry is beyond the scope of this Article. But my assessment of the unconstitutional conditions doctrine, as applied to State Blaines, does suggest some general answers. It would seem, generally speaking, that such conditions cannot have the object or effect of circumventing the foundational principles of religious non-discrimination. That is, if the general principle is that government may not exclude religious providers from otherwise available benefits, government cannot then condition participation in a way that essentially accomplishes the same thing. Such conditions would not be genuinely neutral. So, for instance, a public university cannot condition religious groups' access to generally available funds or fora on the groups' not “discriminating” on the basis of religion in selecting its officers. See Paulsen, supra note 226, at 691. Similarly, government cannot condition religious schools' participation in a voucher program on the schools' not teaching religious tenets that “discriminate” against other religions or against behavior objectionable from their religious standpoint. The issues here quickly become far more complex, but this is not the occasion to explore them more fully.

n420. See supra Part III.