Fact-checked by Grok 2 weeks ago

Zelman v. Simmons-Harris

Zelman v. Simmons-Harris, 536 U.S. 639 (2002), was a landmark United States Supreme Court decision upholding the constitutionality of Ohio's school voucher program in Cleveland under the Establishment Clause of the First Amendment.^[1]^[2]
The case challenged the Pilot Project Scholarship Program, which provided tuition aid to low-income parents in Cleveland's underperforming public schools, allowing them to choose from community schools, magnet schools, private non-religious schools, or private religious schools.^[3]^[1]
In a 5–4 ruling authored by Chief Justice William Rehnquist, the Court held that the program did not violate the Establishment Clause because it was neutral with respect to religion and directed public funds to religious institutions only through the genuine and independent choices of individual parents, rather than government endorsement of faith.^[1]^[2]^[3]
This decision built on prior precedents like Agostini v. Felton and Mitchell v. Helms, emphasizing that where a government aid program defines its recipients by need or disadvantage rather than religious affiliation, and where participants have ample secular alternatives, it withstands constitutional scrutiny.^[1]^[3]
The ruling marked a significant expansion of permissible state support for parental school choice, including religious options, but drew sharp dissent from Justices Stevens, Souter, Ginsburg, and Breyer, who argued it effectively subsidized religious indoctrination and eroded the wall of separation between church and state.^[1]^[4]

Historical and Factual Background

Ohio Pilot Project Scholarship Program

The Ohio Pilot Project Scholarship Program was enacted by the Ohio General Assembly in March 1995 as part of state educational reforms targeting the Cleveland City School District, which had been placed under state control earlier that year due to chronic academic failure, fiscal mismanagement, and low student performance, including only 1% proficiency in math and 7% in reading among fourth-graders.^[2]^[5] The program, effective for the 1996–1997 school year, authorized publicly funded tuition vouchers for eligible students to attend participating private schools or public schools in adjacent districts, alongside tutorial grants and educational supplies for students remaining in Cleveland's public schools.^[2]^[1] It aimed to expand parental choice amid the district's enrollment decline from over 75,000 students in 1976 to about 46,000 by 1999, with 90 schools closed since 1976.^[2] Eligibility extended to all Cleveland residents assigned to public schools, with statutory priority for low-income families (household income at or below 200% of the federal poverty guideline) via a lottery system when applications exceeded available funds.^[2] Vouchers covered up to 90% of tuition (capped at $2,250 per pupil) for low-income recipients, requiring a family copayment of up to $250, while non-priority families received up to $1,875 (75% of tuition).^[2] Participating schools, required to accept vouchers as full payment up to the cap and admit students regardless of religion, included 56 private institutions by the 1999–2000 school year, of which 46 (82%) were religiously affiliated, primarily Catholic; no adjacent public schools enrolled voucher students.^[2] Funds were disbursed directly to parents, who endorsed payments to their chosen school, ensuring neutral allocation based on private choice rather than state endorsement of specific providers.^[2] By the 1999–2000 school year, central to the litigation, over 3,700 students participated, with 96% attending religious schools and 60% from families at or below the poverty line; initial cohorts in 1996 were smaller, serving about 2,000 primary-grade students.^[2] The program was administered by the Ohio Department of Education, with appropriations determining annual capacity, and it included safeguards like required achievement testing for voucher students and continuation of aid through grade 12 for enrolled participants if funding lapsed.^[6]^[2]

Procedural History Leading to Supreme Court

In 1995, the Ohio General Assembly enacted the Pilot Project Scholarship Program as part of the state's efforts to address failing public schools in Cleveland, providing tuition aid to low-income families for alternatives including private and religious schools.^[2] In 1996, a group of Ohio taxpayers, led by Doris Simmons-Harris, filed suit in Franklin County Court of Common Pleas challenging the program under both state and federal constitutions, alleging violations of the Establishment Clause.^[7] The trial court upheld the program, but the Ohio Court of Appeals reversed, finding it unconstitutional under the First Amendment.^[7] On May 27, 1999, the Ohio Supreme Court in Simmons-Harris v. Goff rejected the federal Establishment Clause challenge but struck down the program on state grounds, holding that its enactment violated procedural requirements of the Ohio Constitution's one-subject rule (Article II, Section 15(D)).^[7]^[2] The legislature promptly reenacted the program in summer 1999 without substantive changes, curing the identified defect.^[2] Respondents then initiated a federal action in July 1999 in the U.S. District Court for the Northern District of Ohio, seeking to enjoin the reenacted program as violative of the Establishment Clause.^[8] In August 1999, the district court issued a preliminary injunction (54 F. Supp. 2d 725), which the U.S. Supreme Court stayed pending appeal (528 U.S. 983).^[2] On December 29, 1999, the district court granted summary judgment for respondents, ruling the program unconstitutional under the Establishment Clause as it resulted in excessive government entanglement with religion and primarily benefited religious schools.^[8]^[2] The Sixth Circuit Court of Appeals affirmed on December 11, 2000, in a 2-1 decision (Simmons-Harris v. Zelman, 234 F.3d 945), applying the Lemon test and concluding the program advanced religion by directing public funds predominantly to religious institutions, despite parental choice.^[9]^[2] The state petitioned for certiorari, which the Supreme Court granted on February 26, 2001 (531 U.S. 1152), to review the federal constitutional question.^[2]

Constitutional Context

Establishment Clause and Precedents

The Establishment Clause of the First Amendment states that "Congress shall make no law respecting an establishment of religion," prohibiting government actions that advance or inhibit religion or foster excessive entanglement between government and religious institutions. In the context of public funding for education, the Clause has been interpreted to permit certain neutral aid to religious schools under the "child benefit" theory, where benefits accrue to children rather than institutions, provided the aid does not primarily serve religious purposes.^[10] This framework evolved through decades of Supreme Court decisions balancing free exercise protections against establishment concerns. Everson v. Board of Education, 330 U.S. 1 (1947), marked the incorporation of the Establishment Clause against the states via the Fourteenth Amendment and upheld a New Jersey law reimbursing parents for bus transportation to both public and parochial schools, reasoning that the aid was neutral and comparable to secular public services like police and fire protection.^[10] Subsequent cases refined this approach; Lemon v. Kurtzman, 403 U.S. 602 (1971), introduced a three-pronged test requiring government actions to have a secular purpose, a primary effect that neither advances nor inhibits religion, and no excessive entanglement with religion, striking down direct state subsidies to religious schools for teacher salaries and instructional materials due to entanglement risks.^[11] Later precedents eroded strict applications of the Lemon test in favor of neutrality and private choice principles. In Mueller v. Allen, 463 U.S. 388 (1983), the Court upheld a Minnesota tax deduction for educational expenses, including tuition at religious schools, as it was available to all parents without regard to religious affiliation and followed genuine private choice.^[12] Similarly, Witters v. Washington Department of Services for the Blind, 474 U.S. 481 (1986), permitted state vocational aid to a blind student attending a religious college, emphasizing that the program's neutrality and the individual's choice prevented any impermissible establishment. Zobrest v. Catalina Foothills School District, 509 U.S. 1 (1993), extended this to allow a sign-language interpreter under a federal disabilities program for a deaf student at a Catholic high school, rejecting claims of indoctrination since the aid was neutral and parent-directed. Agostini v. Felton, 521 U.S. 203 (1997), overruled prior restrictions on public employees providing remedial services in religious schools, consolidating Lemon factors into a modified inquiry focused on whether aid creates political divisiveness or advances religion through actual diversion, and upholding neutral, indirect aid.^[13] These rulings collectively established that government programs providing neutral financial assistance, where religious schools benefit only through independent parental decisions, do not offend the Establishment Clause, paving the way for scrutiny of broader voucher systems.^[14]

Evolution of Relevant Legal Tests

The Lemon test, articulated in Lemon v. Kurtzman (1971), established a three-pronged framework for evaluating Establishment Clause challenges: a government action must have a secular legislative purpose, its principal or primary effect must neither advance nor inhibit religion, and it must not foster an excessive government entanglement with religion.^[11] This test drew from prior precedents like Everson v. Board of Education (1947), which incorporated the Establishment Clause against the states via the Fourteenth Amendment and permitted certain indirect aid to religious schools under a "child benefit" theory, and Board of Education v. Allen (1968), which upheld textbook loans to parochial school students if neutrally administered.^[10] However, early applications, such as Meek v. Pittenger (1975) and Wolman v. Walter (1977), struck down direct instructional aid to religious schools under Lemon's effects and entanglement prongs, emphasizing risks of advancing religious indoctrination. Subsequent cases began eroding strict separationist interpretations. In Mueller v. Allen (1983), the Court upheld a Minnesota tax deduction for educational expenses, including tuition at religious schools, as neutral and broadly available, applying Lemon but prioritizing genuine private choice by parents over per-capita aid restrictions.^[12] Zobrest v. Catalina Foothills School District (1993) extended this by permitting a sign-language interpreter under a neutral disabilities program in a religious school, folding entanglement concerns into the effects prong and rejecting presumptions against aid that might free up religious funds. These decisions shifted focus toward neutrality—aid distributed without regard to religious status—and away from rote categorization of aid as direct or indirect. Agostini v. Felton (1997) explicitly modified Lemon, overruling Aguilar v. Felton (1985) and holding that Title I funds for public employees providing remedial services in parochial schools did not violate the Establishment Clause.^[13] The Court consolidated the entanglement prong into the effects analysis, deeming factors like government supervision, character of aid, and potential for religious indoctrination relevant only if they demonstrated actual advancement of religion; it also dismissed political divisiveness as a standalone criterion.^[13] Building on this, Mitchell v. Helms (2000) upheld Louisiana's Chapter 2 program providing neutral educational materials to private schools, including religious ones, via per-capita allocation.^[15] A plurality opinion emphasized that true neutrality—actual on-the-ground distribution without religious criteria—and private decisionmaking sufficed to avoid Establishment Clause violations, rejecting requirements for secular use restrictions or direct oversight as incompatible with evenhanded aid.^[15] These refinements prioritized empirical neutrality and voluntariness over prophylactic barriers, presaging Zelman's validation of voucher programs where public funds reach religious entities solely through independent parental choices in a pluralistic marketplace of options.^[16]

Supreme Court Opinions

Majority Opinion by Chief Justice Rehnquist

Chief Justice William H. Rehnquist delivered the opinion of the Court on June 27, 2002, in a 5-4 decision holding that Ohio's Pilot Project Scholarship Program did not violate the Establishment Clause of the First Amendment.^[17] The program, established in 1995 for the Cleveland City School District—a system under state control due to academic failure—aided over 75,000 mostly low-income and minority students by providing tuition scholarships of up to $2,250 per child for attendance at participating public or private schools, as well as tutorial aid for those remaining in public schools.^[17] In the 1999-2000 school year, of the 3,700 participating students, 96% enrolled in religious private schools out of 56 participating private schools, 46 of which were religiously affiliated.^[2] The majority reasoned that the Establishment Clause prohibits government from making laws "respecting an establishment of religion," but permits neutral programs that neither advance nor inhibit religion through coercion.^[17] Drawing from precedents such as Mueller v. Allen (1983), which upheld Minnesota's neutral tax deduction for educational expenses including religious school tuition; Witters v. Washington Department of Services for the Blind (1986), approving vocational aid to a blind student attending a religious college; and Zobrest v. Catalina Foothills School District (1993), allowing a sign-language interpreter for a deaf student in a religious school, Rehnquist emphasized that aid channeled to religious institutions through genuine private choice does not constitute impermissible endorsement.^[2] These cases demonstrated that when government aid is neutral and benefits a broad class of citizens who independently direct it to religious entities, the incidental advancement of religion results from individual decisions, not state action.^[17] Central to the analysis was the criterion that "a government aid program is not readily subject to challenge under the Establishment Clause if it is neutral with respect to religion and provides assistance directly to a broad class of citizens who, in turn, direct government aid to religious schools wholly as a result of their own genuine and independent private choice."^[17] The Ohio program satisfied neutrality by defining eligibility based on residence in the district and family income, without regard to religion, and by permitting any willing private school to participate under secular criteria like non-discrimination and basic curriculum standards.^[2] Parents exercised true choice among alternatives including community schools, magnet schools, and traditional public options, with no financial disincentive to select non-religious providers—scholarship amounts were uniform, and transportation aid was available.^[17] The majority rejected claims that the predominance of religious schools or high voucher redemption rates at them evidenced coercion or endorsement, noting that such outcomes reflect market availability rather than program design, and that precedents like Mueller dismissed reliance on statistical diversion of funds.^[2] Rehnquist distinguished the program from invalid direct-aid cases, such as those involving loans of instructional materials to religious schools, where government targeted pervasively sectarian institutions.^[17] Here, funds flowed indirectly via parental intermediaries, ensuring no state imprimatur on religious education.^[2] The opinion concluded that the program advanced a secular purpose—improving education for disadvantaged children—without coercing participation in religious activity or inhibiting it, thus reversing the Sixth Circuit's ruling that it unconstitutionally advanced religion.^[17]

Concurring Opinion by Justice O'Connor

Justice Sandra Day O'Connor concurred in the judgment, agreeing that the Cleveland Pilot Project Scholarship Program did not violate the Establishment Clause due to its provision of genuine private choice among educational options.^[14] She emphasized that the program's constitutionality hinged on its specific operational features, including its focus on aiding disadvantaged children in a district with failing public schools, rather than endorsing any broadly neutral aid scheme as per se permissible.^[14] Unlike direct aid cases such as Wolman v. Walter (1977), where government funds flowed presumptively to religious institutions, the vouchers here reached religious schools only through independent parental decisions, severing any direct state-religion link.^[14] O'Connor applied a scrutiny informed by precedents like Mueller v. Allen (1983) and Agostini v. Felton (1997), evaluating the program's purpose, effects, and potential for entanglement.^[14] The program's neutral eligibility—available to all Cleveland students without religious criteria—and safeguards, such as requirements that participating schools accept vouchers as full tuition payment and adhere to antidiscrimination rules, ensured no state coercion or endorsement of religion.^[14] Voucher values were modest, capped at $2,250 for low-income families and $1,875 for others in the 1999-2000 school year, representing less than full tuition at most participating schools and comprising only about 1% of the district's total education budget.^[2] Secular alternatives existed, including 23 magnet schools within the Cleveland public system and 10 community (charter) schools, though participation rates varied; these options, while outnumbered by the 46 religious schools among 56 private participants, provided viable nonreligious paths without state-directed religious benefit.^[2] She dismissed concerns over the 96% of voucher students attending religious schools (primarily Catholic) as irrelevant to constitutionality, stating that "the constitutionality of a neutral choice program does not turn on annual tallies of private decisions made in any given year by thousands of individual aid recipients."^[14] Religious schools derived minimal funding from vouchers—$8.2 million total against $114.8 million for magnet schools—averting dependency or excessive entanglement.^[2] Responding to Justice Souter's dissent, O'Connor argued that empirical predominance of religious use did not equate to state action advancing religion, as parental choice, not government design, drove outcomes; incentives favored no particular sect, and the program's structure mitigated risks of social division.^[14] Her approval was narrowly tailored: expansive programs lacking comparable safeguards, such as statewide vouchers covering full costs without robust secular alternatives, could fail scrutiny by effectively subsidizing religious indoctrination at taxpayer expense.^[14] This reasoning preserved the modified Lemon test's focus on preventing government endorsement while accommodating aid mediated by true individual autonomy.^[14]

Concurring Opinion by Justice Thomas

Justice Thomas joined the majority opinion but filed a separate concurrence to underscore the educational benefits of the Cleveland program and to critique the application of Establishment Clause jurisprudence to state initiatives. He highlighted the program's role in addressing the systemic failures of Cleveland's public schools, particularly for low-income and minority students, noting that in the 1999-2000 school year, the district ranked last among Ohio's 295 districts in overall performance, with proficiency rates as low as 1% in math for grades 1-3 and 7% in reading for the same grades.^[18] In contrast, participating Catholic schools achieved significantly higher outcomes, with 95% of eighth graders passing reading proficiency tests compared to 57% in public schools, and 75% passing math proficiency versus 22% in public schools.^[18] Thomas argued that such data demonstrated the program's necessity as an escape mechanism from ineffective government monopolies on education, aligning with the fundamental parental right to direct their children's upbringing recognized in Pierce v. Society of Sisters, 268 U.S. 510 (1925).^[18] Invoking Frederick Douglass's observation that education served as the primary means of emancipation for freed slaves, Thomas contended that denying inner-city children access to superior religious schools perpetuated cycles of poverty and dependency, irrespective of the schools' religious character.^[18] He noted that 96% of scholarship recipients attended religious schools, with 46 of 56 participating private schools being religiously affiliated (35 Catholic), yet emphasized that this reflected genuine parental choice rather than state endorsement of religion, as funds flowed through independent decisions of families facing few viable alternatives.^[18] Thomas praised the program's neutrality, observing broad support among affected communities, including 75% of black public school parents favoring expansion.^[18] Thomas further expressed skepticism about incorporating the Establishment Clause against the states via the Fourteenth Amendment, arguing that the Clause's original text and history targeted federal overreach, not state policies, and that rigid application had constrained states' ability to experiment with reforms amid failing systems.^[18] While acknowledging the majority's advancement in upholding private choice, he critiqued lingering precedents like Aguilar v. Felton, 473 U.S. 402 (1985), for imposing undue barriers to effective aid distribution, urging greater deference to state innovations that prioritize educational outcomes over formalistic separation concerns.^[18] This approach, he maintained, better served the liberty interests of disadvantaged families without establishing religion.^[18]

Dissenting Opinion by Justice Stevens

Justice Stevens authored a separate dissenting opinion, maintaining that the Ohio Pilot Scholarship Program impermissibly advanced religion in violation of the Establishment Clause by subsidizing indoctrination in sectarian schools with public funds. He contended that the program's structure, which directed the overwhelming majority of tuition aid to religious institutions, constituted direct governmental support for religious education rather than neutral parental choice.^[3] Stevens opened his dissent with a pointed hypothetical, questioning whether a statute authorizing public expenditure to pay for the religious indoctrination of thousands of grammar school children in the tenets of the Roman Catholic faith—or any other faith—could be deemed constitutional merely because parents voluntarily selected that option. He asserted that such a law plainly offends the First Amendment's prohibition on laws "respecting an establishment of religion," as it represents governmental endorsement of specific religious doctrines through taxpayer dollars. This framing underscored his view that the scale and directness of the funding distinguished the Ohio program from permissible minor or incidental aid, emphasizing that the Establishment Clause forbids the state from financing "the religious training and indoctrination of the child" even if routed through parental decisions.^[3] Central to Stevens' critique was the empirical reality of the program's operation: in the 1999–2000 school year, approximately 96 percent of the 1,400 scholarship recipients attending private schools enrolled in religiously affiliated institutions, with nearly all of Cleveland's private schools being sectarian. He argued this outcome was not coincidental but inherent to the program's design in a district where secular private options were scarce and public alternatives, though flawed, existed but were ineligible for direct aid. Unlike the majority's reliance on a "true private choice" framework, Stevens maintained that the disproportionate flow of funds to religious schools demonstrated the program's primary effect of advancing religion, contravening precedents such as Committee for Public Ed. & Religious Liberty v. Nyquist (1973), which invalidated tuition reimbursements that effectively subsidized sectarian education. He distinguished cases like Mueller v. Allen (1983), noting that the Ohio aid was narrowly targeted to a failing urban district dominated by religious private schools, creating incentives that steered low-income families toward faith-based indoctrination rather than providing a genuine array of neutral alternatives.^[3] Stevens further criticized the majority for diluting the Establishment Clause's protections by prioritizing formal neutrality over substantive effects, warning that the decision eroded the wall of separation between church and state essential to preventing governmental favoritism toward religion. He rejected the notion that parental choice neutralized the constitutional infirmity, insisting that the state's role in funding religious instruction—regardless of intermediary parents—breached the principle that public money must not support "the propagation of a religious creed." In his view, the program's permissibility turned not on the voluntary nature of selections but on the inevitable result: state-sponsored religious education for a substantial number of children, which risked entangling government with sectarian interests and undermining religious liberty by fostering divisions along faith lines.^[3] Ultimately, Stevens joined Justice Souter's dissent but filed separately to highlight these concerns, concluding that the ruling marked a regrettable departure from foundational Establishment Clause jurisprudence, potentially opening the door to broader public financing of religious activities under the guise of choice.^[3]

Dissenting Opinion by Justice Souter

Justice David Souter, joined by Justices Stevens, Ginsburg, and Breyer, dissented in Zelman v. Simmons-Harris, arguing that the Ohio Pilot Project Scholarship Program violated the Establishment Clause of the First Amendment by effectively subsidizing religious education on an unprecedented scale.^[3] Souter contended that the program's structure funneled public funds predominantly to religious institutions, creating a direct financial aid mechanism that the Clause prohibits, as it compelled taxpayers to support religious indoctrination regardless of any nominal parental choice.^[2] He emphasized that the aid's scale—projected at $8 million for the 2001-2002 school year alone, part of $33 million expended since 1996—distinguished it from prior cases involving incidental or de minimis benefits to religion.^[2] Souter critiqued the majority's "true private choice" framework as insufficient to neutralize the program's establishment of religion, asserting that choice was illusory in Cleveland's context.^[3] Empirical data showed that 96.6% of voucher recipients in the 1999-2000 school year attended religious schools, with 82% of participating private schools being religiously affiliated and only one nonreligious private school initially involved, which had limited capacity.^[2] Approximately 60% of voucher students came from families at or below the poverty line, heightening coercion concerns, as low-income parents faced few viable secular alternatives beyond the failing public schools the program implicitly critiqued.^[2] Souter argued this predominant religious redirection of funds indicated not neutral parental decision-making but a system designed to benefit religious providers, undermining the neutrality required under precedents like Mueller v. Allen.^[3] Drawing on historical foundations, Souter invoked James Madison's Memorial and Remonstrance Against Religious Assessments and Everson v. Board of Education's "no tax in any amount" principle to warn that public funding of religious teaching risked reviving the sectarian strife that prompted the Clause's adoption.^[3] He distinguished the Ohio program from narrower aids in Witters v. Washington Department of Services for the Blind and Zobrest v. Catalina Foothills School District, where individual benefits could flow to secular options without systemic subsidy, noting here the program's exclusion of traditional public schools as redemption sites reinforced religious channeling.^[2] Souter rejected the majority's dilution of the Lemon test via Agostini v. Felton, maintaining that the program's primary effect was to advance religion through unchecked fiscal support for doctrinal instruction.^[3] Ultimately, Souter concluded that the program breached the Establishment Clause by politicizing religion and coercing support for it, potentially enabling funding for divisive or extremist sects and eroding voluntary religious exercise.^[3] He described the decision as a "tragic" devaluation of the Clause's protections, foreseeing increased governmental entanglement and public division absent rigorous separation.^[2] Joined fully by his colleagues, Souter's dissent underscored the constitutional imperative to avoid any public scheme where religious education becomes the default beneficiary of state aid.^[3]

Core Legal Principles Established

True Private Choice Test

In Zelman v. Simmons-Harris (2002), the Supreme Court established that government voucher programs do not violate the Establishment Clause of the First Amendment when they constitute a program of "true private choice," meaning aid reaches religious schools only through the genuine, independent decisions of parents rather than direct state action or endorsement.^[14] Chief Justice William Rehnquist's majority opinion emphasized that such programs sever any improper link between government and religion by channeling funds via individual parental selections among diverse educational options.^[14] This standard prioritizes neutrality and voluntariness, ensuring no coercion compels participation in religious education.^[2] The core requirements for true private choice include strict neutrality toward religion, whereby eligibility for aid is determined by secular criteria such as financial need and geographic residence, without reference to religious affiliation or practice.^[14] In the Cleveland program, scholarships of up to $2,250 per pupil were available to children from families at or below 200% of the federal poverty line residing in the Cleveland school district, encompassing a broad class of over 75,000 students without favoring religious participants.^[14] Aid must flow indirectly to schools through parents, who select from a genuine array of alternatives, including traditional public schools, magnet schools (enrolling about 13,000 students), community schools (about 1,900 students), nonreligious private schools, and religious private schools (46 of 56 participating private schools).^[14] This multiplicity of options—secular and religious—ensures parents exercise uncoerced discretion, as evidenced by the program's provision of tutoring for those remaining in public schools.^[2] Further, true private choice demands no structural or financial incentives that deliberately skew selections toward religious institutions.^[14] The Cleveland program's funding levels reflected this: voucher amounts were lower than per-pupil expenditures at community or magnet schools ($7,746 and $7,392, respectively, versus $2,250 for vouchers), providing no economic preference for private religious options.^[14] Although 96% of voucher recipients in the 1999-2000 school year attended religious schools—reflecting limited secular private alternatives and parental preferences for institutions with higher proficiency rates (e.g., 95% of Catholic eighth graders passing state reading tests versus 22% in district public schools)—the Court held this outcome alone does not indicate state endorsement, as it stems from private decisions rather than government design.^[14]^[2] This framework aligns with precedents like Mueller v. Allen (1982), Witters v. Washington Department of Services for the Blind (1986), and Zobrest v. Catalina Foothills School District (1993), where tax credits, vocational grants, and sign-language interpreters followed individual choices into religious settings without constitutional violation, even if most beneficiaries were religious.^[14] It distinguishes programs involving direct aid to pervasively sectarian institutions, as in Committee for Public Education & Religious Liberty v. Nyquist (1973), by requiring that government neutrality and parental autonomy prevent any inference of state promotion of religion.^[14] Thus, the test safeguards against Establishment Clause breaches by confirming that religious advancement, if any, is incidental and attributable to private actors.^[2]

Neutrality and Non-Coercion Standards

In Zelman v. Simmons-Harris (2002), the Supreme Court articulated a neutrality standard under the Establishment Clause requiring that government aid programs, such as school vouchers, must be evenhanded toward religion, providing benefits to a broad class of recipients defined solely by neutral, secular criteria like financial need and geographic eligibility, without regard to religious affiliation or the religious character of chosen schools.^[14] The Ohio Pilot Project Scholarship Program satisfied this standard, as it extended tuition aid up to $2,250 per pupil to families in the Cleveland public school district—where only 1,900 of 5,000 children attended regularly—allowing selection from 56 private schools (46 religious) alongside public magnet and community schools, with no preferential treatment for religious institutions.^[2] This neutrality ensured the program's purpose and primary effect neither advanced nor inhibited religion, distinguishing it from prior cases involving direct subsidies to pervasively sectarian entities.^[14] Complementing neutrality, the non-coercion standard demands that aid flow to religious schools only through genuine, independent private choices by beneficiaries, severing any governmental coercion toward religious observance or endorsement.^[14] In Zelman, the Court emphasized that parents faced no compulsion, as they could opt for secular alternatives including over 13,000 slots in magnet schools funded at $7,746 per pupil or 1,900 in community schools at $4,518 per pupil, alongside 10 nonreligious private schools; the fact that 96% of 1,400 scholarship recipients in 1999–2000 chose religious schools reflected market-driven parental preferences rather than state pressure, with religious schools receiving less per-pupil funding than secular public options.^[2] This principle, rooted in precedents like Mueller v. Allen (1983) and Witters v. Washington Department of Services for the Blind (1986), holds that constitutional validity turns not on the statistical outcomes of choices but on the program's structural safeguards against coercion, ensuring no taxpayer funds are diverted to religion absent voluntary redirection by individuals.^[14] These intertwined standards—neutrality ensuring evenhandedness and non-coercion preserving individual autonomy—established that voucher programs channeling aid via true private choice do not violate the Establishment Clause, even if substantial funds ultimately support religious education, provided secular options exist and no incentives skew selections.^[14] Dissenters, including Justice Souter, contended that predominant religious uptake (effectively 99% of vouchers after accounting for tutorial components) evidenced de facto coercion for low-income families lacking viable secular alternatives, potentially fostering religious division, but the majority rejected such empirical tallies as dispositive, prioritizing formal neutrality and choice mechanics over aggregate effects.^[2] This framework has informed subsequent evaluations of school choice initiatives, prioritizing structural design over incidental religious predominance.^[14]

Empirical and Policy Dimensions

Evidence on Voucher Program Effectiveness

Studies evaluating the Cleveland Scholarship and Tutoring Program (CSTP), central to the Zelman case, have produced varied findings on student academic performance. Initial randomized evaluations by researchers such as Jay P. Greene and others, analyzing data from the program's early years (1996–2000), reported modest gains in math proficiency for African American voucher recipients after two years, with effect sizes around 0.17 standard deviations, though reading scores showed no significant improvement.^[19] Longer-term tracking of CSTP participants through high school graduation indicated higher on-time graduation rates (approximately 10 percentage points above public school peers) and increased college enrollment, particularly among low-income minority students, suggesting sustained benefits beyond short-term tests.^[19] Broader empirical research on voucher programs, including meta-analyses of randomized and quasi-experimental designs, reveals moderate positive effects on participant achievement, with greater impacts observed in reading (average effect size of 0.07–0.15 standard deviations) than math, and heterogeneity favoring disadvantaged subgroups.^[20] ^[21] A global meta-analysis of 19 voucher experiments found overall achievement gains equivalent to 0.12 standard deviations after several years, attributing variability to program design factors like voucher value and private school quality.^[20] However, some studies, including reanalyses of Ohio's EdChoice program (similar to Cleveland's), reported initial declines in test scores for voucher users (up to 0.1–0.2 standard deviations in early grades), potentially due to adjustment challenges, though these effects often attenuated over time with no long-term harm evident.^[22] Evidence on competitive effects—where voucher-induced choice pressures public schools—shows small but positive spillovers. A meta-analysis of 28 studies estimated that school choice policies, including vouchers, yield average public school achievement gains of 0.05 standard deviations, driven by enrollment declines prompting efficiency improvements.^[23] In Cleveland, public school districts exposed to CSTP competition exhibited modest test score increases (around 2–4 percentile points) in affected grades post-1997, per event-study analyses.^[19] Non-academic outcomes further support voucher efficacy. Parental satisfaction surveys in Cleveland reported over 80% of CSTP families rating private schools as superior in safety and discipline, correlating with reduced absenteeism.^[19] Longitudinal data from programs like Milwaukee's (analogous to Cleveland) linked vouchers to 6–10% higher postsecondary persistence rates.^[21] Critics, drawing from selective reviews, argue vouchers fail to consistently outperform public options on standardized metrics, but such claims often overlook subgroup benefits and non-cognitive gains, as rigorous multi-study syntheses affirm net positives for choice-enabled mobility.^[24] ^[20]

Criticisms of Public Funding for Private Education

Critics argue that public funding for private education through vouchers diverts essential resources from public schools, which serve the majority of students and often face chronic underfunding. In states with voucher programs, per-pupil public school funding decreases as taxpayer dollars subsidize private tuition, leading to larger class sizes, teacher layoffs, and program cuts; for instance, Arizona's universal voucher expansion in 2022 resulted in an estimated $723 million drain from public schools in its first year, exacerbating budget shortfalls without corresponding enrollment drops from public to private sectors.^[25] This effect is amplified in rural areas, where few private schools exist, leaving public institutions to absorb losses from vouchers claimed by non-public attendees or homeschoolers, thus undermining local tax bases reliant on consolidated public systems.^[26]^[27] Empirical studies have documented negative academic outcomes for voucher recipients, challenging claims of universal benefits. A randomized evaluation of Washington, D.C.'s Opportunity Scholarship Program found that students offered vouchers scored 7 percentile points lower in math after three years compared to non-voucher peers, with no significant gains in reading or other metrics.^[28] Similarly, Louisiana's program, analyzed via lottery data, showed voucher users lagging 0.08 to 0.15 standard deviations behind in math and English after one year, attributing declines to lower private school standards and reduced accountability absent public oversight.^[29] These findings contrast with proponent assertions of competitive improvements, highlighting selection biases in non-experimental studies and the causal risks of shifting students to unregulated environments.^[30] Private schools receiving public funds often evade the transparency and civil rights mandates imposed on public institutions, fostering discrimination and inefficiency. Unlike public schools, voucher-accepting privates can selectively admit based on religion, disability, or behavior, excluding vulnerable groups; historical data from programs like Milwaukee's reveal persistent racial and socioeconomic segregation, with vouchers subsidizing elite enclaves rather than broad access.^[31] Fraud risks compound this, as seen in Ohio's post-Zelman scandals where loosely regulated providers misappropriated funds for non-educational uses, underscoring the absence of standardized testing or fiscal audits that public systems require.^[32] Opponents, including education policy analysts, contend this creates a two-tier system where public funding props up unaccountable alternatives, eroding democratic oversight without proportional societal gains.^[33]

Impact and Legacy

Influence on School Choice Expansion

The Supreme Court's decision in Zelman v. Simmons-Harris on June 27, 2002, marked a pivotal shift by upholding the Cleveland Scholarship Program's use of vouchers redeemable at religious schools, provided the program adhered to a "true private choice" paradigm where aid flowed through parental decisions rather than direct government endorsement of religion.^[34] This ruling removed a major constitutional barrier under the Establishment Clause, catalyzing what scholars describe as the third wave of voucher expansion in the early 2000s by clarifying that neutral, non-coercive programs do not impermissibly advance religion.^[35] Prior to 2002, voucher programs were confined to a handful of states, such as Wisconsin's Milwaukee Parental Choice Program (established 1990) and Ohio's Cleveland initiative (1995), with participation limited to low-income students and often excluding or restricting religious schools due to legal challenges.^[36] Post-Zelman, states accelerated adoption and broadening of voucher-like mechanisms, including traditional vouchers, education savings accounts (ESAs), and tax-credit scholarships, with programs increasingly incorporating religious options. By 2016, voucher programs operated in 13 states plus the District of Columbia, a marked increase from the pre-2002 era, and by 2024, more than half of states—over 30 including the District of Columbia and Puerto Rico—featured at least one form of private school choice policy.^[37]^[38] Notable examples include Florida's expansion of its Opportunity Scholarship Program (originally 1999) following Zelman-aligned rulings, Indiana's Choice Scholarship Program (2011, serving over 40,000 students by 2023), and Louisiana's expansions despite empirical debates over outcomes.^[35] The decision directly influenced federal efforts, such as the D.C. Opportunity Scholarship Program authorized by Congress in 2004, which provided up to $7,500 annually for low-income students and was upheld in subsequent litigation.^[35] This legal greenlight spurred legislative momentum, with Zelman cited in state courts and policy debates as precedent for neutrality toward religious providers, though expansions often faced state-level hurdles like Blaine Amendments or voter referenda (e.g., Utah's 2007 voucher initiative rejected by 62% of voters).^[39] By enabling broader participation—reaching hundreds of thousands of students nationwide—the ruling facilitated a shift from targeted aid to more universal eligibility in some states, laying groundwork for recent surges, such as eight states enacting near-universal programs between 2022 and 2023.^[36]^[40] Despite criticisms from opponents regarding potential public school funding diversion, the decision's emphasis on parental autonomy empirically correlated with program proliferation, as evidenced by the growth from fewer than 10,000 participants in early programs to over 1 million in private choice options by the mid-2020s.^[35]^[41]

Challenges from Blaine Amendments

Following the U.S. Supreme Court's decision in Zelman v. Simmons-Harris (2002), which upheld Cleveland's voucher program under the federal Establishment Clause by emphasizing neutrality and genuine private choice, opponents shifted focus to state constitutional barriers, particularly Blaine Amendments. These provisions, present in 37 state constitutions as of 2023, prohibit the appropriation of public funds to sectarian or religious institutions and originated in the late 19th century amid anti-Catholic nativism aimed at blocking aid to parochial schools.^[42] Unlike the federal standard clarified in Zelman, state courts have often interpreted Blaine Amendments to bar indirect funding of religious schools via vouchers or scholarships, even when programs are neutral on their face.^[43] Prominent post-Zelman challenges succeeded in several states. In Florida, the state Supreme Court in Bush v. Holmes (2006) struck down the Opportunity Scholarship Program, ruling that its vouchers for low-income students to attend private (including religious) schools violated the state constitution's no-aid clause—a Blaine Amendment analog—by diverting funds from public education and effectively compelling support for religion.^[44] Similarly, in Colorado, the Supreme Court in Taxpayers for Public Education v. Douglas County School District (2015) invalidated a pilot voucher program allowing students to use public funds for private tuition, holding it breached the state constitution's prohibition on aiding sectarian schools, despite federal precedents like Zelman.^[45] These rulings highlighted how Blaine Amendments enable stricter scrutiny, often prioritizing public school uniformity over parental choice, with critics noting their historical roots in religious discrimination rather than neutral separation principles.^[46] Subsequent U.S. Supreme Court decisions, such as Espinoza v. Montana Department of Revenue (2020) and Carson v. Makin (2022), curtailed overt discrimination against religious schools in state aid programs by invoking the Free Exercise Clause, effectively limiting some Blaine Amendment applications that excluded faith-based options from neutral funding streams.^[47] However, state courts in places like Oklahoma and Michigan continued to invoke Blaine provisions to challenge expansions; for instance, in 2024, the U.S. Supreme Court declined certiorari in a Michigan case testing the Blaine Amendment's bar on funding religious schools via tax credits, leaving the restriction intact.^[48] As of 2025, these amendments persist as litigation flashpoints for universal school choice initiatives, prompting reform efforts like constitutional amendments in states such as Arizona and Iowa to permit broader private school participation.^[49]

Subsequent Supreme Court Developments

In Trinity Lutheran Church of Columbia, Inc. v. Comer (2017), the Supreme Court held 7-2 that a Missouri program granting funds for playground resurfacing from recycled tires violated the Free Exercise Clause by excluding a church-operated preschool solely due to its religious status, as the program was neutral and generally available to private nonprofits. Chief Justice Roberts's majority opinion emphasized that disqualifying entities for religious character imposes a penalty on free exercise, though a footnote reserved whether the ruling extended to funding "otherwise express religious instruction." This decision marked an initial shift toward prohibiting status-based discrimination in public benefits, laying groundwork for applying Zelman's neutrality principles beyond vouchers to direct aid scenarios, without altering the core Establishment Clause tolerance for parent-directed choice. Building on Trinity Lutheran, Espinoza v. Montana Department of Revenue (2020) ruled 5-4 that Montana's state constitutional provision barring aid to religious schools could not exclude them from a tax-credit scholarship program for private education, as this discriminated against religious status in violation of the Free Exercise Clause.^[50] The majority, again led by Chief Justice Roberts, rejected the state supreme court's Blaine Amendment application, noting it treated religious schools "worse than nonreligious schools" despite the program's private-choice structure akin to Zelman's.^[50] The decision clarified that Zelman's endorsement of neutral, non-coercive funding mechanisms precluded states from using no-aid clauses to bar religious options in generally available programs, though it distinguished exclusions tied to actual religious use rather than mere status. The trajectory culminated in Carson v. Makin (2022), where the Court held 6-3 that Maine's tuition assistance program for rural students attending private high schools violated the Free Exercise Clause by requiring funded schools to be nonsectarian, effectively excluding religious institutions based on their religious character and use.^[51] Chief Justice Roberts's opinion extended Espinoza by prohibiting not only status-based but also use-based discrimination in neutral aid programs, affirming that Zelman's "true private choice" framework under the Establishment Clause coexists with Free Exercise protections against exclusionary conditions that penalize religious exercise.^[51] Dissenters, led by Justice Sotomayor, argued this eroded traditional no-establishment safeguards, but the ruling solidified that states cannot condition educational funding on secularism where private choice predominates, prompting reevaluation of similar programs nationwide.^[51] These rulings progressively narrowed barriers to religious school participation in state-funded choice initiatives, reinforcing Zelman's validation of neutrality and non-coercion while invoking Free Exercise to override state-level restrictions, thereby expanding empirical opportunities for vouchers and scholarships without evidence of government endorsement of religion.^[50]^[51]

References

[1]
Zelman v. Simmons-Harris | Oyez
The case questioned if Ohio's school voucher program violated the Establishment Clause. The Court ruled it did not, as it was a program of true private choice.
[2]
Zelman v. Simmons-Harris | 536 U.S. 639 (2002)
The court held that a state program providing tuition vouchers for private or religious schools does not violate the Establishment Clause, as it does not ...
[3]
ZELMAN V. SIMMONS-HARRIS - Law.Cornell.Edu
This Court's jurisprudence makes clear that a government aid program is not readily subject to challenge under the Establishment Clause if it is neutral with ...
[4]
ZELMAN V. SIMMONS-HARRIS - Law.Cornell.Edu
Jun 27, 2002 · The Court's majority holds that the Establishment Clause is no bar to Ohio's payment of tuition at private religious elementary and middle schools.Missing: summary | Show results with:summary
[5]
[PDF] ADMINISTRATIVE COSTS OF EDUCATION VOUCHER PROGRAMS
In March 1995 the Ohio General Assembly enacted a Pilot Project Scholarship Program intended to provide a limited number of vouchers to allow Cleveland ...<|control11|><|separator|>
[6]
Section 3313.975 | Pilot project scholarship program. - Ohio Laws
The program shall provide for a number of students residing in any such district to receive scholarships to attend alternative schools, and for an equal number ...
[7]
[PDF] Simmons-Harris v. Goff - Supreme Court of Ohio
{¶ 5} The court of appeals declared the School Voucher Program to be unconstitutional, holding it violative of the Establishment Clause of the First. Amendment ...
[8]
Simmons-Harris v. Zelman, 72 F. Supp. 2d 834 (N.D. Ohio 1999)
On cross-motions for summary judgment, the court concluded that the 1995 program violated neither the United States Constitution nor the Ohio Constitution. The ...
[9]
SIMMONS HARRIS v. 00 3063 (2000) - FindLaw Caselaw
On May 27, 1999, the Ohio Supreme Court issued a judgment in favor of Plaintiffs, holding that the 1995 voucher program had been enacted in violation of the one ...<|separator|>
[10]
https://supreme.justia.com/cases/federal/us/330/1/
[11]
Lemon v. Kurtzman | 403 U.S. 602 (1971)
Lemon v. Kurtzman: The test for determining whether a law meets the requirements of the Establishment Clause is whether it has a legitimate secular purpose, ...
[12]
https://supreme.justia.com/cases/federal/us/463/388/
[13]
Agostini v. Felton | 521 U.S. 203 (1997)
The Court found that the program violated the Establishment Clause's prohibition against "government-financed or government-sponsored indoctrination into the ...
[14]
ZELMAN V. SIMMONS-HARRIS - Law.Cornell.Edu
Jun 27, 2002 · The Ohio Supreme Court rejected respondents' federal claims, but held that the enactment of the program violated certain procedural ...
[15]
Mitchell v. Helms | 530 U.S. 793 (2000)
As this passage indicates, the private choices helped to ensure neutrality, and neutrality and private choices together eliminated any possible attribution to ...
[16]
Establishment Clause Tests | U.S. Constitution Annotated | US Law
The Supreme Court has used the Lemon test (secular purpose, no effect on religion, no excessive entanglement), strict scrutiny, and historical traditions to ...
[17]
ZELMAN v. SIMMONS-HARRIS | Supreme Court - Law.Cornell.Edu
Chief Justice Rehnquist delivered the opinion of the Court. The State of Ohio has established a pilot program designed to provide educational choices to ...
[18]
ZELMAN V. SIMMONS-HARRIS - Law.Cornell.Edu
Jun 27, 2002 · This Court has held that parents have the fundamental liberty to choose how and in what manner to educate their children.
[19]
The Empirical Evidence on Cleveland's School Vouchers - EdChoice
This policy brief examines empirical studies on Cleveland's CSTP, a publicly funded voucher program that includes religious schools.
[20]
The participant effects of private school vouchers around the globe
We find moderate evidence of positive achievement impacts of private school vouchers, with substantial effect heterogeneity across programs and outcome years.
[21]
The impact of voucher programs: A deep dive into the research
most rigorous studies found positive short- and long-term impacts for participating students—especially for Black and disadvantaged students ...
[22]
[PDF] The Effects of Ohio's EdChoice Voucher Program on College ...
Prior research on. EdChoice found that the program modestly improved the test scores of public school students but harmed the achievement of students who used a ...
[23]
The Competitive Effects of School Choice on Student Achievement
This systematic review and meta-analysis tests this theory by synthesizing the empirical literature on the competitive effects of school choice on student ...
[24]
Vouchers do not improve student achievement, Stanford researcher ...
Feb 28, 2017 · Education professor Martin Carnoy analyzed 25 years of research and found that voucher programs do not significantly improve test scores.
[25]
How vouchers harm public schools: Calculating the cost of voucher ...
Dec 19, 2024 · EPI has developed a tool that estimates fiscal externalities—the dollar costs to school districts from students leaving public schools with a ...
[26]
Vouchers Deliver Blow to Rural Schools: 'They're Taking Money ...
Jun 13, 2025 · School voucher programs have a proven record of failure and are especially harmful to rural students, families, and communities.
[27]
How the School Choice Agenda Harms Rural Students
May 15, 2025 · Private school voucher programs do not provide “choice” for the vast majority of rural families; instead, they divert funding from their public schools.<|separator|>
[28]
[PDF] School Vouchers and Student Outcomes: Experimental Evidence ...
School vouchers are the most contentious form of parental school choice. Vouchers provide government funds that parents can use to send their children to.
[29]
Research on school vouchers suggests concerns ahead for ...
Aug 15, 2023 · Finally, there is modest evidence that traditional vouchers may compel small improvements in the achievement levels of at-risk public schools.
[30]
Private school vouchers: Research to help you assess school choice ...
This explainer examines academic research on how private school vouchers and other school choice programs affect student achievement.
[31]
[PDF] Research Shows Private School Vouchers Don't Work for Students ...
Studies have found that students who attended private schools using vouchers actually performed worse than their similar peers in public schools. Page 2 ...
[32]
Taxpayer-funded private school vouchers and market failure
Sep 2, 2025 · We found little evidence that private school vouchers created markets that produced valued educational outcomes. In many cases, policymakers ...
[33]
State Policymakers Should Reject K-12 School Voucher Plans
Mar 21, 2023 · School vouchers raise the risk of harm to students, do little to expand opportunity, and cut funding to public schools. State policymakers should reject these ...
[34]
https://www.supreme.justia.com/cases/federal/us/536/639/
[35]
https://www.brookings.edu/articles/three-waves-of-school-vouchers-a-history-of-expansion-and-exclusion/
[36]
[PDF] The ABCs of School Choice - EdChoice
As the months ticked by, a total of seven states enacted new choice programs and. 10 expanded ones already in operation. As of this writing, eight states have ...<|separator|>
[37]
School vouchers - Ballotpedia
Voucher programs exist in 13 states and the District of Columbia as of April 2016. History. State support for private school education is not entirely new.
[38]
Navigating Constitutional Waters: The Legality of School Choice ...
Jun 4, 2025 · First, in Zelman v. Simmons-Harris (2002), the Court upheld Cleveland, Ohio's voucher program. The Cleveland program, created in response to a ...
[39]
[PDF] A Practical Guide to School Choice after Zelman V. Simmons-Harris
Feb 4, 2003 · All federal Establishment Clause questions are analyzed using a modified version of the. “Lemon test,” an inquiry first developed by the.
[40]
A historical overview of private school vouchers - EdNC
Jan 23, 2024 · Sentiments around school choice proposals, including vouchers, began to shift in the 1980s and 1990s following the release of “A Nation at Risk, ...
[41]
Fast Facts - EdChoice
There are 23 voucher programs in 15 states—Georgia, Indiana, Louisiana (2), Maine, Maryland, Mississippi (2), New Hampshire, North Carolina, Ohio (5), Oklahoma, ...
[42]
Blaine Amendments - The Institute for Justice
Blaine Amendments are controversial state constitutional provisions rooted in 19th century anti-Catholic bigotry.Missing: 2023-2025 | Show results with:2023-2025
[43]
Blaine Amendments - Free Speech Center - MTSU
Jul 31, 2023 · Simmons-Harris (2002) appeared to remove First Amendment obstacles to voucher programs, pro-voucher forces turned their attention to rescinding ...
[44]
[PDF] How the Courts Struck Down the Florida School Voucher Program
This Part will address why the District Court of Appeal erred in its refusal to either strike down the State's no-aid provision as unconstitutional under the.
[45]
Colorado Supreme Court Strikes Down School Vouchers
Jun 29, 2015 · Earlier today, the Colorado Supreme Court ruled that Douglas County's school voucher program violates the state constitution.<|separator|>
[46]
School Choice and State Constitutions' Religion Clauses
The two most significant are compelled support clauses and Blaine Amendments. Both are frequently misinterpreted by state courts as applied to school choice ...
[47]
What the Espinoza Ruling Means for Blaine Amendments and ...
Jul 22, 2020 · After the U.S. Supreme Court ruled that vouchers are constitutional in Zelman v. Simmons-Harris, 536 U.S. 639 (2002), Dick Komer, who ...Missing: striking post-
[48]
U.S. Supreme Court Denies Blaine Amendment Challenge
Oct 7, 2024 · Michigan's discriminatory Blaine Amendment will continue to restrict school choice after the United States Supreme Court declined to hear a case challenging ...Missing: 2023-2025 | Show results with:2023-2025
[49]
Court Cases and the Legal Environment Around School Choice in ...
Jul 28, 2025 · Back in 2023, we posted about court cases impacting school choice, parental rights, and education in general. Here's an update for 2025.
[50]
[PDF] 18-1195 Espinoza v. Montana Dept. of Revenue (06/30/2020)
Jun 30, 2020 · The case involves Montana's tax credit program for private school scholarships, which was challenged for excluding religious schools, and ...
[51]
[PDF] 20-1088 Carson v. Makin (06/21/2022) - Supreme Court
Jun 21, 2022 · See, e.g., Zelman v. Simmons-Harris, 536 U. S.. 639, 662 (2002) (allowing government funds to flow to reli- gious schools if private individuals ...